RESEARCH IN GLOBAL STRATEGIC MANAGEMENT Series Editor: Alan M. Rugman Recent Volumes: Volume 1: International Business Research in the 21st Century Volume 2: Global Competition and European Community Volume 3: Corporate Response to Global Change Volume 4: Global Competition: Beyond Generics Volume 5: Beyond the Diamond Volume 6: Multinational Location Strategy Volume 7: International Entrepreneurship: Globalization of Emerging Business Volume 8: Leadership in International Business Education and Research Volume 9: Multinationals, Environment and Global Competition
RESEARCH IN GLOBAL STRATEGIC MANAGEMENT VOLUME 10
NORTH AMERICAN ECONOMIC AND FINANCIAL INTEGRATION EDITED BY
ALAN M. RUGMAN Indiana University, USA
2004
Amsterdam – Boston – Heidelberg – London – New York – Oxford Paris – San Diego – San Francisco – Singapore – Sydney – Tokyo
ELSEVIER B.V. Sara Burgerhartstraat 25 P.O. Box 211 1000 AE Amsterdam The Netherlands
ELSEVIER Inc. 525 B Street, Suite 1900 San Diego CA 92101-4495 USA
ELSEVIER Ltd The Boulevard, Langford Lane, Kidlington Oxford OX5 1GB UK
ELSEVIER Ltd 84 Theobalds Road London WC1X 8RR UK
© 2004 Elsevier Ltd. All rights reserved. This work is protected under copyright by Elsevier Ltd, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier’s Rights Department in Oxford, UK; phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail:
[email protected]. Requests may also be completed on-line via the Elsevier homepage (http://www.elsevier.com/locate/permissions). In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555; fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier’s Rights Department, at the fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. First edition 2004 British Library Cataloguing in Publication Data A catalogue record is available from the British Library. ISBN: 0-7623-1094-4 ISSN: 1064-4857 (Series) ∞ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Printed in The Netherlands.
CONTENTS LIST OF CONTRIBUTORS
ix
LIST OF FIGURES
xi
LIST OF TABLES
xv
PREFACE
xix
PART I: REGIONAL ECONOMIC INTEGRATION IN NORTH AMERICA NORTH AMERICAN INTRA-REGIONAL TRADE AND FOREIGN DIRECT INVESTMENT Alan M. Rugman
3
CANADA-U.S. ECONOMIC INTEGRATION FOLLOWING NAFTA Steven Globerman and Paul Storer
17
REGIONAL VS. GLOBAL FINANCING STRATEGIES FOR U.S. MNEs Joseph P. Daniels, Walid Hejazi and Marc von der Ruhr
47
REGIONAL INTEGRATION OF U.S. BORDER STATES WITH CANADA: EVIDENCE FROM U.S. STATE EXPORTS Larry Davidson
69
v
vi
PROTECTING CANADA AND THE U.S. AGAINST TERRORISM: A COMMON SECURITY PERIMETER? Georges A. Tanguay and Marie-Christine Therrien
85
WAL-MART IN MEXICO Syed Tariq Anwar
97
PART II: NORTH AMERICA FINANCIAL INTEGRATION AND A COMMON CURRENCY ASSESSING THE BORDER: TESTS OF THE LAW OF ONE PRICE IN CANADA AND THE U.S. Janet Ceglowski
117
THE IMPACT OF NAFTA ON THE INTEGRATION OF THE CANADIAN, MEXICAN, AND U.S. EQUITY MARKETS Claire G. Gilmore and Ginette M. McManus
137
A SINGLE CURRENCY FOR NAFTA? Patrick M. Crowley
153
NORTH AMERICAN INTEGRATION AND THE SINGLE CURRENCY? Martin Coiteux
175
CANADIAN INVESTMENT IN THE UNITED STATES: A REGIONAL PERSPECTIVE Prem Gandhi and Wayne Glass
193
PART III: NAFTA AT AGE TEN: SECTORAL STUDIES AND RELATED ISSUES CANADA-U.S. AGRICULTURAL TRADE Andrew Schmitz and Hartley Furtan
213
vii
THE CANADA-U.S. SOFTWOOD LUMBER DISPUTES Harry Nelson and Ilan Vertinsky
237
TRUCKING SERVICES UNDER NAFTA Christine Drennen
263
COSTS OF THE U.S.-CANADA BORDER John C. Taylor, Douglas R. Robideaux and George C. Jackson
283
A CONTINENTAL ELECTRICITY INDUSTRY Paul S. Ciccantell
299
PARTICIPATION BY CIVIL SOCIETY AND NON-GOVERNMENTAL ORGANIZATIONS IN WESTERN HEMISPHERE ECONOMIC INTEGRATION Jonathan P. Doh and Barbara Kotschwar
317
HAS NAFTA FULFILLED ITS PROMISE? Bernard L. Weinstein
339
AUTHOR INDEX
349
SUBJECT INDEX
355
LIST OF CONTRIBUTORS Syed Tariq Anwar
West Texas A&M University, Canyon, TX, USA
Janet Ceglowski
Bryn Mawr College, Bryn Mawr, PA, USA
Paul S. Ciccantell
Western Michigan University, Kalamazoo, MI, USA
Martin Coiteux
HEC Montreal, Montreal, Canada
Patrick M. Crowley
Texas A&M University, Corpus Christi, TX, USA
Joseph P. Daniels
Marquette University, Milwaukee, WI, USA
Larry Davidson
Indiana University, Bloomington, IN, USA
Jonathan P. Doh
Villanova University, Villanova, PA, USA
Christine Drennen
Bowling Green State University, Bowling Green, OH, USA
Hartley Furtan
University of Saskatchewan, Saskatchewan, Canada
Prem Gandhi
Plattsburgh State, Plattsburgh, NY, USA
Claire G. Gilmore
King’s College, Wilkes-Barre, PA, USA
Wayne Glass
Center for the Advancement of Sustainable Tourism, Saranac Lake, NY, USA
Steven Globerman
Western Washington University, Bellingham, WA, USA
Walid Hejazi
University of Toronto, Toronto, Canada
George C. Jackson
Wayne State University, Detroit, MI, USA
Barbara Kotschwar
Organization of American States and Georgetown University, Washington, DC, USA
Ginette M. McManus
St. Joseph’s University, Philadelphia, PA, USA ix
x
Harry Nelson
University of British Colombia, Vancouver, BC, Canada
Douglas Robideaux
Grand Valley State University, Grand Rapids, MI, USA
Alan M. Rugman
Indiana University, Bloomington, IN, USA
Andrew Schmitz
University of Florida, Gainesville, FL, USA
Paul Storer
Western Washington University, Bellingham, WA, USA
Georges A. Tanguay
University of Maine, Orono, ME, USA
John C. Taylor
Grand Valley State University, Grand Rapids, MI, USA
Marie-Christine Therrien
University of Maine, Orono, ME, USA
Ilan Vertinsky
University of British Colombia, Vancouver, BC, Canada
Marc von der Ruhr
St. Norbert College, De Pere, WI, USA
Bernard L. Weinstein
University of North Texas, Denton, TX, USA
LIST OF FIGURES Rugman Figure 1: Exports in the Broad Triad, 2000 (U.S. $ Billion) Figure 2: Total Outward FDI in the Broad Triad, 1999 (U.S. $ Billion) Figure 3: Intra-Regional Trade and FDI in the Triad, 1980-2000 Globerman/Storer Figure 1: Average of Logged Canada-U.S. Deviations from the Law-of-One-Price Figure 2: Variance of Canada-U.S. Deviations from PPP Figure 3: BLS Indexes of Hourly Manufacturing Compensation Costs (U.S. Dollar Basis) Figure 4: Ratio of U.S. to Canadian Average Weekly Earnings, 1991–2001 (in U.S. Dollars) Figure 5: Returns on Equity and Investment Figure 6: Manufacturing Net Profit Rates in Canada and the U.S. Figure 7: Behavior of the Nominal Exchange Rate Daniels/et al. Figure 1: Distribution of FDI Stock Figure 2: Distribution of FDI Across Industries Figure 3: Total, Regional, and Global Distribution of FDI
6 7 8
28 29 31 31 33 34 38 52 53 57
Anwar Figure 1: Major Global Retailers: Sales and Net Income (1997–2002)
106
Gilmore/McManus Figure 1: Standardized Stock Price Indexes of United States, Canada, and Mexico
141
Crowley Figure 1: BIC by Numbers of Clusters Figure 2: 4 Cluster Classification from EM Using EEV Model (a) GDP vs. Unemployment (b) GDP vs. Lagged GDP
xi
164 165
xii
Coiteux Figure 1: Relative Capital Inflows and the Bilateral Real Exchange Rate Figure 2: Relative Terms of Trade and the Bilateral Real Exchange Rate Figure 3: Relative Productivity and the Bilateral Real Exchange Rate Figure 4: Relative Capital Inflows and the Bank of Canada’s Relative Price of Commodities Series Schmitz/Furtan Figure 1: Canada-U.S. Agricultural Exports, 1989–2001 Figure 2: Canada-U.S. Dairy Exports, 1989–2001 Figure 3: Canada-U.S. Corn and Wheat Trade, 1990–2002 Figure 4: Canada-U.S. Fresh and Frozen Pork Exports, 1989–2001 Figure 5: Canada-U.S. Live Pork Exports, 1989–2001 Figure 6: Canada-U.S. Fresh and Frozen Beef Exports, 1989–2001 Figure 7: Canada-U.S. Live Cattle Exports, 1989–2001 Figure 8: Canada-U.S. Aggregate Producer Subsidy Equivalents, 1986–2001 Figure 9: (a) U.S. Commodity Payments vs. Direct Support, 1986–2001 (b) Canadian Commodity Payments vs. Direct Support, 1986–2001 Figure 10: Canada-U.S. Individual Commodity Producer Subsidy Equivalents, 1986–2001 Figure 11: Canada-U.S. Tomato Exports, 1989–2001
184 184 185 186
214 215 215 217 218 218 219 224
225 226 231
Nelson/Vertinsky Figure 1: U.S. Lumber Demand and Domestic Supply, 1984–2001 Figure 2: Canadian Share of U.S. Lumber Consumption, 1984–2001 Figure 3: Domestic Lumber Supply in the U.S. from Selected Regions, 1978–2001
243
Drennen Figure 1: Total U.S.-Canada-Mexico Trade by Truck (U.S. $ Billions)
268
Taylor et al. Figure 1: Total U.S.-Canada-U.S. Economic Activity, Canadian Land Exports to U.S. and Trucks in October 2000 to June 2001 to June 2002 (Percentage Monthly and Cumulative Period Change)
286
241 242
xiii
Doh/Kotschwar Figure 1: Framework for Understanding NGO Influence in Agreements Weinstein Figure 1: U.S. Exports to Canada and Mexico, 1996–2002 (in Billions of U.S. Dollars) Figure 2: U.S. Imports from Canada and Mexico, 1996–2002 (in Billions of U.S. Dollars) Figure 3: U.S. Trade Balances (in Millions of U.S. Dollars) Figure 4: Foreign Direct Investment in the U.S.: Capital Inflows (in Millions of U.S. Dollars) Figure 5: Foreign Direct Investment in the U.S.: Capital Outflows (in Millions of U.S. Dollars) Figure 6: Agricultural Trade Between Mexico and the U.S. Figure 7: Breakdown of Subsidies to Mexican Farmers Figure 8: Farmers in Mexico are Better Subsidized than in the U.S. (Guaranteed Income in Dollars per Ton)
327
340 341 341 342 342 345 345 346
LIST OF TABLES Rugman Table 1: Canada Direct Investment Position Abroad, 1986–1999 Table 2: Bilateral Stocks of U.S. Foreign FDI: Canada and the United States, 1982–2000 (Millions of U.S. $) Table 3: United States: The Relative Importance of Trade in Services 1971–2000 Globerman/Storer Table 1: Distribution of U.S. Exports by Country (Percent of Total) Table 2: Distribution of U.S. Imports by Country (Percent of Total) Table 3: Distribution of Canada’s Exports by Country (Percent of Total) Table 4: Distribution of Canada’s Imports by Country (Percent of Total) Table 5: Geographic Distribution of Inward FDI Stock in Canada (Percent) Table 6: Geographic Distribution of Canadian FDI Stock Abroad (Percent) Table 7: Geographical Distribution of Inward FDI in the U.S. (Percent) Table 8: Geographical Distribution of U.S. Outward FDI (Percent) Table 9: Non-Permanent Immigration from Canada to the U.S. Table 10: Share of Canada’s Total Goods Trade (%)
9 10 12 21 21 22 22 24 24 25 25 26 35
Daniels/et al. Table 1: Pattern of Accumulated FDI Table 2: U.S. Bilateral Tax Agreements Table 3: Estimation Results for FDI Stocks Table 4: Estimation Results for FDI Flows with Tax Variable Intercept Dummy Variables Table 5: Estimation Results for FDI Flows with Vintage Effects
62 63
Davidson Table 1: Export Sales of Border States to the World, Canada, and Mexico, 1996 (in Billions of Dollars)
74
xv
56 58 61
xvi
Table 2: Export Sales of Border and Non-Border States to the World, Canada, and Mexico, Percentage Change from 1996 to 2001 (in Percent) Table 3: Export Sales Share of Border and Non-Border State Exports to World Destinations, 1996 and Change from 1996 to 2001 Table 4: Similarity Index Values by Export Destination in 1996 and Percentage Change from 1996 to 2001 (Index Values and Percent)
75 76
77
Anwar Table 1: Profile of Wal-Mart (August 2002) Table 2: Global Retailers Ranked by Sales (2002) Table 3: Global vs. Regional Strategies: Issues and Concepts
99 100 103
Ceglowski Table 1: Cross-Border Price Differences (%) Table 2: Intranational Price Differences (%) Table 3: Univariate ADF Tests for Cross-Border Prices Table 4: Univariate ADF Tests for Intranational Prices Table 5: National Average Cross-Border Prices (% Change)
121 122 124 127 130
Gilmore/McManus Table 1: Summary Statistics of Weekly Index Returns Series Table 2: Correlation Coefficients of Weekly Stock Return Series Table 3: Unit Root Tests for Weekly Stock Indices Table 4: Bilateral Johansen Cointegration Tests Results Table 5: Multilateral Johansen Cointegration Test Results Table 6: Vector Error-Correction Model Results Table 7: Variance Decomposition of Forecast Errors Results
142 142 143 144 144 145 146
Crowley Table 1: Parameterizations of the Covariance Matrix Table 2: BIC Values by Number of Clusters Using the EM Algorithm Table 3: Cluster Membership Using EEV Model and EM Algorithm
160 163 164
Coiteux Table 1: Dependence on Regional Trade as Measured by the Combined Intensity Index Table 2: Regional Dependence Index
179 180
xvii
Gandhi Table 1: Properties of Canadian Companies Investing in the United States (Percent Responding) Table 2: Motives for Canadian Companies to Invest in the U.S.: A Regional Perspective (Percent Responding) Table 3: Most Important and Least Important Factors for Locating in the U.S.: A Regional Perspective (Percent Responding) Table 4: Consistency of Results with Previous Studies Table 5: Company Attitude Toward Further Integration (Percent Responding) Table 6: Effects on Canadian Companies’ Operations Under Expanded Integration (Percent Responding) Schmitz/Furtan Table 1: U.S. Agricultural Commodity Loan Rates (1991 and 2002) Table 2: U.S. Agricultural Commodity Target Prices Table 3: U.S. Oilseed and Pulse Corps: Loan Rates Table 4: Average of Canadian Non-Tarrif-Rate Quotas (TRQs) and Over-Quota TRQs, 2001 Table 5: U.S. Agricultural Trade Issues and Concerns Under CUSTA, 1989–2001 Table 6: U.S. Agricultural Trade Issues and Concerns Under CUSTA, 1989–2001 Table 7: Canadian Agricultural Trade Issues and Concerns Under UCSTA Nelson/Vertinsky Table 1: Annual Harvest by Ownership in the U.S. in 2001 (Million/ft3 ) Table 2: Softwood Lumber Production in Canada for Selected Provinces in 1992 and 2002, in Billion Board Feet Table 3: Ownership of Timberlands in the U.S. and Canada Drennen Table 1: Domestic, International, & NAFTA Activity (Average Annual Growth Rate 1995–2000) Table 2: Trade Dependence of NAFTA Partners (NAFTA Trade as a Percentage of World Trade) Table 3: NAFTA Truck Traffic, 2000
200 202 203 205 206 207 219 220 220 226 227 229 230
243 244 247
269 269 270
xviii
Taylor et al. Table 1: Two Way Traffic into Canada and USA by Vehicle Type 000’s of Vehicles Table 2: Cross-Border Freight Rate/Cost Penalty (Billions of U.S.$) Table 3: Transit Time and Uncertainty Related Impact Summary (Millions of U.S.$) Table 4: Truck Primary and Secondary Delay Cost Impacts (Millions of U.S.$) Doh/Kotschwar Table 1: Trade Agreements Under Negotiation or in Effect in the Western Hemisphere Table 2: Role of NGOs in Key Trade Agreements in Which U.S. and Canada are Members Table 3: Submissions by Civil Society Groups Under Free Trade Agreement of Americas Process Topical Distribution Table 4: Submissions by Civil Society Groups Under Free Trade Agreement of Americas Process Geographical Distribution
285 290 291 292
324 326 331 332
PREFACE This book contains selected papers first presented by invitation at the “CanadaUnited States Business Conference” held at the Kelley School of Business, Indiana University, on April 11–12, 2003. The conference was co-sponsored by the Canadian Embassy of the Government of Canada and the IU Center for International Business Education and Research (IU CIBER). The primary objective of this conference was to bring together U.S. and Canadian-based business scholars to help develop a network of specialists on Canadian-U.S. trade and investment issues, in particular, and bilateral business issues more generally. With a primary emphasis on Canadian-U.S. issues, the conference was broad enough to incorporate NAFTA-related business issues and researchers with expertise on the business relations of North America. Within a broad framework, issues of current interest and concern – NAFTA at age ten, the impact of the FTAA on NAFTA, were examined. Of particular importance in the post-September 11, 2001 environment, other issues included cross border trade and security. Key policy areas discussed included regulatory issues, issues of labor mobility, productivity, and specific sectoral issues (energy, transportation, etc.). In addition to an examination of key research issues, conference participants also considered the public policy implications of their findings – the likely impact of resulting policy options on business practice and its environments. Key members of the Canadian, U.S., and Indiana-specific business and policy making communities were involved in the conference. Indiana is a major exporting state, ranking 14th of the fifty U.S. states. Canada is the State’s largest export market, receiving some 50% of its annual exports. Highlighting the importance of North American trade for the state is the fact that Indiana’s second largest trading partner, Mexico, receives a total of 13% of its annual exports. Indiana University is well known as a major research university, and its Kelley School of Business is home to a significant number of highly-regarded research scholars. Like the University, the Kelley School of Business has been involved in international research and programmatic activities over the last five decades – in Europe, Asia, Latin America, and elsewhere. The IU Center for International Business Education and Research (IU CIBER), a designated U.S. National Resource Center, has been located in the School for more than a decade. IU xix
xx
CIBER encourages and supports the international research endeavours of Kelley faculty and particularly encourages collaboration among academics and, as an extension of that work, their outreach to the business community. The IU CIBER is directed by the Kelley’s School’s L. Leslie Waters Chair of International Business, a chair designated for an internationally recognized scholar in the IB field. Of the 30 papers prepared for the conference, a set of 18 were accepted for publication in this book. These are grouped into three major sections. In Part I, six papers deal with economic integration in North America, including papers on the strategies of multinational enterprises and exporting firms. In Part II, there are five papers dealing with financial integration in North America and the issues of a common currency. Finally in Part III, seven papers examining specific sectors and issues of public policy across the Canada-U.S. border. I wish to thank Louise Siffin, Director of Global Programs at the Kelley School, for her help with the design and implementation of the conference. Teena Albright provided superb organisational skills to facilitate the attendance and presentation of a large number of visiting scholars. Finally, Mildred Harris prepared the manuscript for this book with great diligence and attention to detail, interacting with the authors to prepare a book of lasting merit. Alan M. Rugman Bloomington, Indiana University October 2003
PART I: REGIONAL ECONOMIC INTEGRATION IN NORTH AMERICA
NORTH AMERICAN INTRA-REGIONAL TRADE AND FOREIGN DIRECT INVESTMENT Alan M. Rugman ABSTRACT Since the North American Free Trade Agreement (NAFTA) of 1994, the economic interdependence of Canada and the United States has continued to grow. With Mexico now beginning to integrate itself much more strongly with the U.S. economy, NAFTA has emerged as a strong economic block with nearly as high a degree of integration as the E.U. Furthermore, NAFTA’s economic integration is growing at a faster rate. The data to support these findings are examined in this paper.
INTRODUCTION This paper examines recent empirical data on the nature and extent of Canada-U.S. economic integration, especially since the North American Free Trade Agreement (NAFTA) of 1994. The statistical evidence is that the increase in regional integration in the NAFTA region has been achieved mainly through trade. Intra-regional trade in NAFTA is now at 55.7%, up from 33.6% in 1980, and 49.2% in 1996. At the same time, intra-regional FDI in NAFTA has declined, particularly U.S. FDI in Canada. In contrast, very recently, the share of Canadian FDI in the United States has increased slightly, after decreasing every year since the Canada-U.S. North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 3–16 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10001-6
3
4
ALAN M. RUGMAN
Free Trade Agreement (FTA) started in1989. Trade in the automotive sector and related industries, such as in specialty chemicals and steel products, heavily influences this pattern of manufacturing regionalization by intra-regional trade. In contrast, services are now being integrated in NAFTA, more by FDI than by trade, as many barriers to trade in services remain. To examine these trends in the bilateral relationship, and to relate them to broader trade and FDI issues, data on NAFTA-based service trade and FDI are related to changes in the manufacturing sector. The data for NAFTA indicate a pattern of increased intra-regional trade in manufacturing over the last 20 years but decreased intra-regional FDI. Similar trends exist in the E.U.; however, there is increasing inter-regional FDI between the E.U. and NAFTA. The latter is part of the “triad” effect of large MNEs operating mainly within their home-market triads of NAFTA, the E.U. and Asia, see Rugman (2000), but also seeking to access each other’s triad market. The data indicates that MNEs are strongly home-triad market-based, but there is some evidence of increased E.U.-NAFTA inter-regional FDI in the last few years.
THEORETICAL BACKGROUND Previous work on Canada-U.S. trade and FDI appeared in Rugman (1990). One of the main findings of that study was that Canadian FDI in the United States was growing, over the 1975–1987 period, at twice the pace of U.S. FDI in Canada. In 1975, the percentage of Canadian FDI in the United States to U.S. FDI in Canada (in stocks) was 18.7%. Already by 1980 it was 33.7%, by 1984 it was 48.0%, and by 1987, at the start of the FTA, it was 57.6%, Rugman (1990, p. 12). The theoretical explanation behind this dramatic trend was that Canadian MNEs needed to access the U.S. market through FDI. As the United States was at least 10 times greater than Canada’s, investing in a U.S. business became a huge commitment for Canadian firms who needed to compete with larger U.S. rival firms. However, access to the U.S. market through FDI allowed Canadian MNEs to develop continental capabilities, following the example of Northern Telecom, Alcan, Bombardier and Seagram. An earlier explanation of this phenomenon of the growth of Canadian multinationals and Canadian outward FDI appears in Rugman (1987). At the time of the FTA this information about the growth and maturity of Canadian multinationals was a counterpoint to the popular Canadian concern about U.S. ownership of the Canadian economy. A related finding, in Rugman (1990), was that U.S. foreign control of Canadian industry actually decreased from 28.8% in 1970 to 21.9% in 1985. In manufacturing, U.S. control fell from
North American Intra-Regional Trade and Foreign Direct Investment
5
46% in 1970 to 39.8% in 1985 and, in petroleum from 77.0% in 1970 to 39.6% in 1985. These two events, the decline of U.S. foreign control of Canadian industry, coupled with the dramatic increase in Canadian FDI in the United States, reduced and alleviated Canadian concerns about U.S. economic domination. These two trends indicated that Canadian business was capable of capturing an increased relative share of the North American business system. Canada’s dependence on the U.S. market is well known. In 2000, 87% of Canada’s exports went to the United States (up from 76% in 1991). In 1999, 52.2% of Canada’s outward FDI stock went to the United States. In a complementary manner, U.S. economic involvement with Canada is also strong and is its largest bilateral relationship. In 2000, 22.6% of all U.S. exports went to Canada, see Rugman and Hodgetts (2003, p. 513). Back in 1991, U.S. exports to Canada were 20.2%, and in 1981 they were 16.9% of all U.S. exports. In contrast to the increase in intra-regional trade between Canada and the United States, the intra-regional FDI has been falling relative to FDI with non-NAFTA countries. In 1982, the share of the outward U.S. stock of FDI in Canada was 20.9%. By 1989 it was 16.7%, and by 2000 it was 10.2% of all U.S. outward FDI. This decrease in the proportion of U.S. FDI in Canada is a puzzle which merits more attention than has been paid to it, although a similar result appears in Safarian and Hejazi (2001) and in Hejazi and Safarian (2002). In this paper the reasons for the relative decline in U.S. FDI in Canada are explored and implications for business and public policy are considered. The Canada-U.S. bilateral relationship, which experienced ever deeper degrees of economic interdependence over the 1950–1988 period, was finally recognized in the Canada-U.S. Free Trade Agreement in 1989. This provided a set of trade and investment rules to govern and further facilitate this economic relationship. Subsequently, the trade and investment provisions of the FTA were used as the basis for NAFTA which began in 1994. The FDI aspects of NAFTA have been examined in Rugman (1994). The main finding of that study was that the Canada-U.S. economic relationship would not be affected much by NAFTA but that Mexico would experience the greatest burden of adjustment and also make the largest gains from NAFTA. This prediction has been borne out in practice by subsequent events which have led to a large increase in intra-regional trade, but not in FDI in NAFTA. Safarian and Hejazi (2001) have also found that inward FDI into Canada has been falling in the 1990s, while outward FDI from Canada has been increasing. In a more recent study Hejazi and Safarian (2002), using a gravity model for trade and an augmented gravity model for FDI, found that NAFTA has led to less FDI between Canada and the United States, but to more intra-regional trade. They argue that U.S. MNEs are now better able to serve the Canadian market by trade
6
ALAN M. RUGMAN
rather than FDI. These studies are entirely consistent with the data and analysis in this paper.
RECENT CANADA-U.S. TRADE AND FOREIGN DIRECT INVESTMENT The increase in regional integration under the FTA and NAFTA has been achieved mainly through trade. With the elimination of tariffs and most other barriers to trade in goods and traded services, the U.S. and Canadian economies have a situation of virtual free trade. However, there remain barriers to trade in most of the service sectors; indeed as services are largely produced and consumed locally there is little in the FTA and NAFTA to facilitate increased trade in services. Furthermore, over half of the service sectors were excluded from the national treatment provisions of the FTA and NAFTA, which mainly applied to business services. Sectors exempted include: cultural services; health; education; social services; transportation; and financial services, Rugman (1994). Several implications arise from the institutional fabric of the FTA and NAFTA: (1) Free trade for goods and traded services has resulted in an increase in intraregional trade, as shown in Fig. 1; intra-regional trade is now at 55.7% (up from 33.6% in 1980 and 49.19% in 1996). One key reason for the increasing
Fig. 1 Note: Asia and intra-Asia is the addition of Japan and Developing Asia in the IMF, Direction of Trade Statistics Yearbook. It does not include New Zealand and Australia. Source: Adapted from International Monetary Fund, Direction of Trade Statistics Yearbook, 2001 (Washington, DC: IMF, 2001).
North American Intra-Regional Trade and Foreign Direct Investment
7
Fig. 2 Source: OECD, International Direct Investment Statistics Yearbook, 2001. (1) Data for the European Union only include Austria, Finland, France, Germany, Italy, Netherlands, Sweden, United Kingdom, Portugal, Greece and Denmark, but include FDI to all other E.U. countries. Data for other E.U. countries are not available. For Italy, data for FDI to Asia (excluding Japan and Korea) are for 1994, the latest year available. Data for Denmark, France, Netherlands, Austria, Germany and Finland are for 1998. (2) Data for Asia do not include Australia and New Zealand. Data for Japan are 1998.
degree of intra-regional trade in NAFTA is the increasing dependence of both Canada and Mexico on the U.S. market. In 2000, 87% of Canada’s exports went to the United States and 88.7% of Mexico’s to its main NAFTA partner. (2) As can be seen in Fig. 2 the situations with regard to FDI is more complex. There has been a decrease in intra-regional FDI as a percentage of total FDI for 1999 for NAFTA at 18.2% (down from 21.1% in 1997 and 30.3% in 1986). The main reason is there has been a decrease in the percentage of U.S. FDI to Canada, as described below. The key trade and FDI stock data available for years 1980, 1986, 1997, 1999 and 2000 are summarized in Fig. 3. Figure 3 confirms the two main points made above. First, intra-regional trade has been increasing at a steady pace within each of the markets of the triad, but it is increasing at the fastest rate in NAFTA in recent years. Going back to 1980, both NAFTA and Asia have shown similar rates of increase in intra-regional trade, but Asia’s integration largely occurred before 1997, whereas NAFTA’s has accelerated since then.
8
ALAN M. RUGMAN
Fig. 3 Note: International Direct Investment Statistics Yearbook, 2001. Source: IMF, Direction of Trade Statistics Yearbook, 1983–2001 and OECD.
Second, in terms of FDI, there is a slowing down of intra-regional FDI stocks in all three parts of the triad between 1997 and 1999. This reflects greater interregional FDI, especially both ways between the E.U. and NAFTA in recent years. However, over the 1980–1999 period, there has been an increase in intra-regional FDI in two parts of the triad – the exception is NAFTA. The NAFTA decrease in intra-regional FDI is almost entirely due to changes in the U.S.-Canadian economic relationship. Back in 1986 there were still major barriers to intercontinental business, but the Canada-U.S. Free Trade Agreement of 1989 and NAFTA in 1994 served to open up both markets to trade, thereby reducing the need for FDI to overcome trade barriers. This bilateral relationship is discussed in more detail below. These two statements indicate clearly that as NAFTA facilitates free trade, U.S. business can access the Canadian market better than before from production bases in the United States, so less FDI is required to overcome regulatory barriers to traded goods. Unlike the situation in the E.U., the Canadian market was heavily resistant to U.S. foreign ownership in the 1970s and early 1980s, Safarian (1966). Not until the Mulroney government abolished the Foreign Investment Review agency in 1985 was a strong signal sent welcoming U.S. FDI in Canada. In 1980 to 1981, following Canada’s discriminatory National Energy Program, there was a huge withdrawal of U.S. FDI in energy-related goods and services. In fact, this was the largest single outflow of FDI in any sector in world history, see
North American Intra-Regional Trade and Foreign Direct Investment
9
Rugman (1990). In contrast, within the E.U., there were fewer restrictions on intra-regional FDI. Data from the U.S. Survey of Current Business (2002) report there was only 10.2% of the U.S. outward stock of FDI in Canada in 2000, compared to 20.9% in 1982 (and 16.7% in 1989 at the start of the FTA and 12.1% in 1994 at the start of NAFTA). The U.S. stock of FDI in Mexico was 2.8% in 2000, about the same as in 1994, at the start of NAFTA. Canada has very little FDI in Mexico, and Mexico has little in the rest of NAFTA. In addition, the United States has increased its FDI position in the E.U. and Asia, at the expense of Canada, as barriers to trade remain relatively stronger in these other parts of the triad. By 2000, the U.S. stock of FDI in Europe was 52.1% of its total (up from 44.5% in 1982); in Asia it was 16.0% (up from 13.6% in 1982). In terms of world-wide FDI stocks, OECD (2000) data support this analysis. By 1999, the U.S. stock of FDI in the E.U. was up to 45.2% (from 38.0% in 1986); in Asia it was at 12.9% (up from 10.3% in 1986). In contrast, the U.S. stock of FDI in the rest of NAFTA (Canada and Mexico) was 12.9% in 1999 compared to 21.3% in 1986. For Canada, its stock of FDI in NAFTA (almost entirely in the United States) in 1999 was 53.3% compared to 68.7% in 1986; 58.3% in 1991; and 53.1% in 1996. In 1999, Canada’s stock of FDI in the E.U. was 18.8%; in Asia it was 5.8%. Table 1 reviews Canada’s FDI stock in the United States alone, not across NAFTA. In 1999, the Canadian stock in the United States was 52.2% (as compared Table 1. Canada Direct Investment Position Abroad, 1986–1999. Year
Total
U.S.
% of total
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
61,498 70,639 79,763 89,851 98,402 109,068 111,691 122,427 146,315 161,237 180,616 209,678 246,313 257,408
42,027 46,091 51,025 56,578 60,049 63,379 64,502 67,677 77,987 84,562 93,886 105,683 124,405 134,281
68.3 65.2 64.0 63.0 61.0 58.1 57.8 55.3 53.3 52.4 52.0 50.4 50.5 52.2
Source: Adapted from OECD, International Direct Investment Statistics, 2001 and 1995.
10
ALAN M. RUGMAN
Table 2. Bilateral Stocks of U.S. Foreign FDI: Canada and the United States, 1982–2000 (Millions of U.S.$). Year
Canada’s FDI Position in the U.S. (1)
U.S. FDI Position in Canada (2)
Net Position [(1) − (2)]
(1)/(2)
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
11,708 11,434 15,286 17,131 20,318 24,684 26,566 30,370 29,544 36,834 37,843 40,487 42,133 45,618 54,799 64,022 72,696 76,526 100,822
43,511 44,779 47,498 47,934 52,006 59,145 63,900 63,948 69,508 70,711 68,690 69,922 74,987 83,498 91,301 99,859 98,200 111,051 126,421
−31,803 −33,345 −32,212 −30,803 −31,688 −34,461 −37,334 −33,578 −39,964 −33,877 −30,847 −29,435 −32,854 −37,880 −36,502 −35,837 −25,504 −34,525 −25,599
0.27 0.26 0.32 0.36 0.39 0.42 0.42 0.47 0.43 0.52 0.55 0.58 0.56 0.55 0.60 0.64 0.74 0.69 0.80
13.5
6.5
(1.3)
6.6
Average % (82–00)
Note: Data are on a historical-cost basis. Sources: Adapted from BEA, Survey of Current Business, June, 2002; June 1999 and July 1996.
to 53.3% in NAFTA, including Mexico with the United States). Other than a small increase in 1998, the 2000 figure reverses a steady decrease in the share of Canadian FDI in the United States, which fell from 68.3% in 1986 to 63.0% in 1989 (with the FTA) and to 53.3% in 1994 (with NAFTA). Rather than this relative decrease being offset by a marked increase in FDI to the other two triad regions of the E.U. and Asia, most of the Canadian FDI no longer going to the United States instead appears to be going into tax havens (listed as “other”), which accounted for 22.1% of Canada’s FDI in 1999. Table 2 relates Canada’s FDI stock in the United States to the U.S. stock in Canada. The former has continued to grow at 13.5% per year, double the U.S. rate of 6.5% per year. As a result the ratio of Canadian FDI in the United States to U.S. FDI in Canada has continued to increase. In 1982 it was 27%; in 1989 it was 47% with the FTA. In 1994 it was 56%, and by 2000 it was 80%. These data are from a U.S. source which actually understates the ratio, compared to Statistics Canada data used in Rugman (1990), where these ratios are much higher. The reason, as discussed in Rugman (1987, 1990), is that the Canadian-based data evaluate the
North American Intra-Regional Trade and Foreign Direct Investment
11
Canadian stock of FDI in the United States as about 50% greater than the U.S. evaluation, due to different methodological approaches. Overall, these data reveal a decrease in the intra-regional FDI of NAFTA, although the decrease mainly occurred before the FTA and NAFTA started. This may have been in anticipation of the FTA, Rugman (1990). Between 1986 and 1996, the U.S. stock in NAFTA (mainly in Canada) declined from 21.3% to 13.7%, but then only to 12.9% by 1999. Similarly, the Canadian stock in the United States fell from 68.7% in 1986 to 53.1% in 1996, but has actually increased slightly to 53.3% in 1999. The largest change involves an increase in Canada’s FDI in “other” from 12.8% in 1986 to 22.1% in 1999. In short, both the United States and Canada are now less dependent upon each other in terms of FDI compared to the situation in 1986. To keep this decrease in NAFTA intra-regional FDI in perspective, however, let us consider the situation in Europe, as was shown in Figs. 2 and 3. In 1986, the intra-regional stock of EC FDI was 35.8%; in 1997 it was 49.3%; but in 1999 it fell to 45.7%. This recent decrease in E.U. intra-regional FDI stock, over the last five years of data availability, is actually greater in absolute values than the decrease in NAFTA intra-regional stock of FDI over the same period. The principal reason for this switch is an increase in the E.U. stock in NAFTA from 22.8% in 1996 to 30.8% in 1999. In other words, European MNEs are increasingly seeking access to the NAFTA market through FDI at a time when intra-regional trade in the E.U. is still increasing. The European situation is therefore exactly similar to that of NAFTA, increasing intra-regional trade but FDI going more inter-regional. Returning to the importance of increased intra-regional trade in NAFTA, the United States has become more open, especially in terms of imports, which in 2000 represented almost 15% of GDP. Exports accounted for nearly 11% of GDP in 2000. These ratios have increased since NAFTA and have more than doubled over the last 30 years. The U.S. economy is now more open than the Japanese economy. The U.S. openness coefficient is 16% of GDP as compared to 10% of GDP for Japan. The European Union continues to be the most open trade block, according to these data, with a 35.6% openness coefficient. However, the vast of majority of E.U. trade is intra-regional, and this significantly biases upward this ratio.
SECTORAL CHANGES IN MANUFACTURING AND SERVICES Another way of interpreting the data on increased intra-regional trade yet decreased intra-regional FDI within NAFTA is to take into account the differences in sectoral performance. For example, it is well known that about one-third of all
12
ALAN M. RUGMAN
U.S.-Canadian trade is in automobiles and auto-parts, transportation equipment and affiliated industries. This integration of the largest manufacturing sector has occurred over the last 40 years. When specialty chemicals, plastics, steel, and related manufacturing sectors linked to the automobile sector are brought into consideration, it is apparent that probably half of all U.S.-Canadian trade is automobile related. The dominance of this manufacturing sector in the bilateral trade statistics is extraordinary and shows the special nature of the bilateral relationship. In contrast, in services, market access on both sides of the border is much more diversified. No one service sector dominates FDI or trade data. Therefore, the service sector will be subject to different explanations than the manufacturing sector, since the latter is dominated by automobiles; services are much more diversified. Theory shows that access by Canadian business to the U.S. service sector has to be by FDI as services are produced and consumed locally. In contrast, market access for manufacturers in NAFTA is now largely achieved through trade. Consistent with this point, Hejazi (2003) shows that the share of Canadian FDI in the United States has grown by 19.2% over the 1983–1995 period, whereas the share of U.S. FDI in Canada in services has only grown by 4.5% over the same period. Table 3 shows that the United States has halved its ratio of services to goods imports over the 1971–2000 period. However, since just after NAFTA started there has only been a slight decrease from 18.8% in 1996 to 17.8% in 2000. In contrast, Table 3. United States: The Relative Importance of Trade in Services 1971–2000. Year
1971 1976 1981 1986 1991 1996 2000 % Change (71–00)
Services as % of Total
Ratio of Services to Goods
Exports
Imports
Exports
Imports
27.41 19.60 19.48 27.96 28.40 28.17 27.46
25.25 16.51 14.65 17.87 19.44 15.81 15.14
37.76 24.38 24.20 38.81 39.67 39.21 37.86
33.8 19.8 17.2 21.8 24.1 18.8 17.8
0.18
(40.04)
0.25
(47.18)
Note: Goods are calculated on a balance of payment basis. The data on ratio of services to goods are expressed as services as a percentage of goods. Source: BEA, Gross Domestic Product by Industry and the Components of Gross Domestic Income, 1947–2000 BEA, Balance of Payments and Historical Estimates.
North American Intra-Regional Trade and Foreign Direct Investment
13
the ratio of U.S. services to goods exported has remained high at 37.86% in 2000, slightly down from 39.21% in 1996. Coupled with earlier information about the increase in intra-regional trade in NAFTA and the decrease in intra-regional FDI in NAFTA, this reduced ratio of imported services suggests a switch in the United States from trade in services towards FDI in services. This is apparent only when we look at service imports to the United States. Using the trade data in Table 3 we note that the ratio of service imports to total imports decreased from 25.15% in 1971 to 15.81% in 1996 and then to 15.14% in 2000. So, the United States must have increased its share of manufactures in total imports. As about one-quarter of all U.S. imports come from Canada we can infer that there must be a large increase in intra-regional manufacturing trade in NAFTA over the last three years. In contrast, access to the U.S. market for services is now more likely to be by FDI. In terms of services, NAFTA is a different cultural animal than the E.U. Within the E.U. there is a policy to promote harmonization and homogenization across its internal market. There is a social charter, a single currency, and every six months new waves of directives stemming from the commissioners in Brussels. These institutional developments are necessary due to the historical, political, and cultural divisions between and within the fifteen member states of the E.U. In contrast, in NAFTA, there is huge degree of existing similarity between wealthy U.S. and Canadian consumers, and the existence of variation in the service sector stems more from regulatory sovereignty than from cultural differences. In other words, due to size asymmetries, Canadians have already taken on board the dominant cultural and social characteristics of the United States. Today, much attention is paid in the Canadian media to what are really trivial differences between middle-class Canadians and Americans. In Europe, the more balanced sizes of the six large and nine smaller economies has led to a smaller signaling effect from a dominant partner. Furthermore the close geographic proximity of Canada to the United States has resulted in Canadian business achieving a high degree of market access to its neighboring economy. The large size of the United States acts as a magnet to business. This has resulted in both greater exports and growth of FDI stocks by Canada in the United States over the last 20–30 years. In contrast, as the Canadian economy is only one-tenth the size of the U.S. economy, there has been less incentive for U.S. business to locate in Canada (in a relative sense). In particular, U.S. FDI in Canada in the service sector is hindered because many sectors are exempted from national treatment. So this reduces potential U.S. interest in Canadian sectors, such as banking, health care, culture (including newspapers, book publishing, and other media), transportation, etc. Consequently, Canada does not attract U.S. services FDI today in the same proportion as it did in U.S. FDI in manufacturing 20–30 years ago.
14
ALAN M. RUGMAN
One key aspect of the size asymmetries has been that Canadian business has only to learn how to deal with one set of U.S. institutions (and Americans with one set of Canadian institutions) whereas within the E.U. the complex array of regulations and institutional peculiarities of each of the 15 member states have taken a longer time to resolve into a common set of institutions.
FIRM-LEVEL DATA ON INTRA-REGIONAL BUSINESS A final checkpoint on this aggregate data comes from looking into firm-level data. The largest 500 MNEs in the world account for well over 90% of the world’s stock of FDI and over half of the world’s trade, Rugman (2000). In a recent analysis of the intra-regional sales of the largest 100 of these MNEs it was found that 67 of the 78 cases, for which data were available, had a majority of the sales in their home triad, Rugman and Brain (2003). For these 67 MNEs the average ratio of home-triad foreign-to-total (F/T) sales was 79.3%. For all 78 MNEs the average of home-triad intra-regional sales was 71.6%. Of the largest 100 MNEs, only three were “global” in the sense of having at least 20% of their sales in each of the three parts of the triad: IBM, Sony, and Nestl´e. Another four are bi-regional, with 20% of the sales in two parts of the triad and under 50% in the home triad: BP, Toyota, Nissan, and Unilever. A final set of four MNEs are host-triad oriented: Daimler-Chrysler, ING, Royal Ahold, and Honda. Overall, this is confirmation of the importance of doing business in the home triad, and it supports a focus on NAFTA for Canadian-based firms. In a related study all 49 retail MNEs in the world’s largest 500 were examined, including Wal-Mart which is currently the largest MNE. Only one of these 49 MNEs is “global” in the above sense. This is LVMH, the luxury goods retailer, see Rugman and Girod (2003), Rugman (2005).
CONCLUSIONS The major trend over the last 20 years has been an increase in intra-regional trade within each triad block. This is especially noticeable in NAFTA. The U.S. economy is now more open (at 11% exports to GDP and 15% imports to GDP). This is similar to Japan, where exports account for 10.8% of GDP, but it is behind the E.U. average of 35% of GDP. Due to increased intra-regional trade in NAFTA, U.S. FDI stocks in NAFTA, as a percentage of worldwide FDI, have decreased over the last 20 years. In contrast,
North American Intra-Regional Trade and Foreign Direct Investment
15
intra-regional FDI stocks have continued to increase within the E.U. and Asian triad blocks. The unusual situation in NAFTA is dominated by the U.S.-Canadian manufacturing relationship. In turn, this reflects the overwhelming importance of the automobile sector in North America. This sector accounts for one-third of U.S.-Canadian trade. In this sector free trade has existed for over 40 years, and it has replaced the need for FDI. In contrast, across the service sectors affected by NAFTA, FDI is still important. Indeed, it remains the only viable mode of entry in sectors which are still regulated, such as financial services, transportation, cultural industries, health, education, and social services. Furthermore, national treatment for FDI is still denied in these service sectors. This reduces the incentive for U.S. FDI in the service sector in Canada but (due to size asymmetries) does not discourage Canadian FDI in the United States. For example, privatization and de-regulation in the U.S. financial service sectors now present opportunities for Canadian banks to expand into the United States. They will need to do this through FDI rather than trade due to the need to deliver services locally and the residual risks in this quasi-regulated sector. At firm level, data on foreign sales by MNEs reveals that very few of the world’s largest MNEs are “global” in the sense of operating in all three triad markets. The vast majority of large MNEs mainly operate in their home-triad market. These firm-level data support the large degree of intra-regional trade in NAFTA.
ACKNOWLEDGMENTS Helpful research assistance has been provided by Cecilia Brain of Toronto. Helpful comments have been received from Walid Hejazi of the University of Toronto.
REFERENCES Hejazi, W. (2003). Foreign direct investment and domestic capital formation. Journal of International Business Studies (forthcoming). Hejazi, W., & Safarian, A. E. (2002, June). Explaining Canada’s changing FDI patterns, Mimeo. Rotman School of Management, University of Toronto. Rugman, A. (1987). Outward bound: Canada’s direct investment in the United States. Toronto: C. D. Howe Institute. Rugman, A. (1990). Multinationals and Canada-United States free trade. Columbia, SC: University of South Carolina Press. Rugman, A. (Ed.) (1994). Foreign investment and NAFTA. Columbia, SC: University of South Carolina Press. Rugman, A. (2000). The end of globalization. London: Random House and New York: AmacomMcGraw-Hill.
16
ALAN M. RUGMAN
Rugman, A. (2005). The regional multinationals. Cambridge: Cambridge University Press. Rugman, A., & Brain, C. (2003). Multinational enterprises are global, not regional. Multinational Business Review, 11(1), 3–12. Rugman, A., & Girod, S. (2003, February). Retail multinationals and globalization: The evidence is regional. European Management Journal, 21(1). Rugman, A., & Hodgetts, R. (2003). International business (2nd ed.). London: Pearson Financial Times/Prentice-Hall. OECD (1995, 2000). International Direct Investment Statistics Yearbook. Safarian, A. (1966). Foreign ownership of Canada’s industry. Toronto: McGraw-Hill of Canada. Safarian, A., & Hejazi, W. (2001). Canada and foreign direct investment: A study of determinants. Toronto: Rotman School Centre for Public Management, University of Toronto.
CANADA-U.S. ECONOMIC INTEGRATION FOLLOWING NAFTA Steven Globerman and Paul Storer ABSTRACT This paper evaluates the extent and implications of Canada-U.S. economic integration in the wake of two formal trade liberalization agreements. The paper considers how quantity and price measures can be used to assess integration, then surveys the evidence on the extent of integration. Overall, we find little evidence that these trade agreements had significant incremental impacts on economic integration between Canada and the United States. We find some evidence that exchange rate variability may discourage integration. Microeconomic efficiency has not been enhanced through alignment of prices and costs and the volatility of the Canada-U.S. exchange rate may also account for this. The finding provides some tentative evidence in favor of a common currency arrangement.
1. INTRODUCTION Substantial controversy surrounded the negotiation and implementation of both the Canada-U.S. Free Trade Agreement (CUSTA) and the North American Free Trade Agreement (NAFTA). Supporters of the agreements tended to emphasize the substantial economic benefits that would be associated with a closer integration of input and output markets across the countries involved. Opponents of the agreements emphasized the potential for closer economic integration to lead to North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 17–45 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10002-8
17
18
STEVEN GLOBERMAN AND PAUL STORER
a flight of capital from Canada to the United States (in the case of CUSTA) and from the United States to Mexico (in the case of NAFTA). Whether and how to proceed with new initiatives to promote further regional economic integration remain issues today. In particular, recent reports by the Canadian government call for broader and deeper economic integration with the United States. While public opinion towards regional economic integration between Canada and the United States is relatively positive, significant new initiatives to encourage even closer economic integration are bound to spark controversy. In this context, it seems useful to assess how far the two countries have come in terms of being economically integrated both to help assess the potential importance of future integration initiatives, as well as to evaluate the consequences of the measures that have been implemented to date. For pragmatic reasons, it is easiest to focus on the integration that has taken place between Canada and the United States. In particular, data for the two countries are relatively available and reliable. Moreover, integration within North America is most clearly pronounced in the Canada-U.S. case. For example, increased labor mobility within the North American market subsequent to a new visa arrangement incorporated into the NAFTA primarily affects the bilateral flow of skilled workers between Canada and the United States.1 For these reasons, our primary interest is the closer economic integration between Canada and the United States subsequent to the 1989 implementation of the CUSTA. Nevertheless, we will make reference to the Mexican experience subsequent to the 1994 implementation of NAFTA, when possible, to compare and contrast it to the bilateral experience. The paper proceeds as follows. Section 2 identifies and discusses measures of economic integration. It also summarizes the existing literature focusing on North American economic integration particularly distinguishing between quantity and price-based measures of integration. Section 3 presents statistical evidence that we develop bearing upon the degree of regional economic integration. Explanations for this evidence other than CUSTA are discussed in Section 4. In Section 5, we offer a policy analysis and recommendations for future policy initiatives. A brief summary and several conclusions are offered in Section 6.
2. MEASURING ECONOMIC INTEGRATION Economists and international business scholars employ both de jure and de facto definitions of economic integration. The former identifies the comprehensiveness of legal arrangements between countries. The latter encompasses a range of actual macroeconomic variables. The most typical variable cited is trade flows between a set of countries. Flows of foreign direct investment (FDI) are also frequently
Canada-U.S. Economic Integration Following NAFTA
19
cited. Thus, one well-known trade policy expert cites growing trade and FDI flows between Canada and the United States subsequent to the implementation of CUSTA as illustrative of convergence between the two economies since the implementation of CUSTA (Hufbauer, 2001). Rugman (2000) also highlights flows of exports from Canada and Mexico to the United States as an indicator of regional economic integration, along with U.S. FDI stocks in Canada and Canadian FDI stocks in the United States. More recently, immigration flows between Canada and the United States under temporary visa arrangements have been identified as an indicator of increasing economic integration.2 The latter indicator is much less frequently referenced than trade or FDI flows, presumably reflecting the relatively small (to date) value of human capital imputed to recent flows (Globerman, 2000). In addition to quantity-based measures of the volume of economic activity potentially related to regional integration initiatives, one might also look at the behavior of prices within a region. Specifically, as barriers to the movement of goods, services and factors of production between members of a regional integration arrangement are reduced, or dismantled, theory suggests that there should be an intensification of trade between the economies. In the neoclassical economic model, an intensification of trade should lead to an equalization of prices net of transport costs and taxes (Hine, 1994). Furthermore, since trade is a substitute for factor movements in the neoclassical model, increased trade should also lead to a convergence of wages and returns to capital. To the extent that direct factor movements are stimulated by differences in wage rates and rates-of-return, increased cross-border flows of capital and labor should further contribute to a convergence of returns to factors of production within the integrating region. Is there any reason to favor quantity or price-based measures of economic integration? The theory of contestable markets suggests that price convergence is perhaps a more generally relevant indicator of regional economic integration than are quantity flows of outputs and inputs. Specifically, the theory of contestable markets makes the fundamental point that the threat of substantial new entry into domestic industries can cause monopoly prices to decline to competitive levels without actual entry taking place. Moreover, the threat of new entry can lead to reductions in X-inefficiency. The latter refers to higher than necessary costs that are sustained by firms being protected from more efficient competitors. In the extreme, the threat of new competition from imports can promote significantly lower prices in domestic markets without any significant increase in import volumes. As a consequence of this insight, as well as the fact that much less attention has been paid to price-based measures of regional economic integration, we focus prominent attention upon that measure of bilateral economic integration; however, we begin by addressing and assessing conventional measures of output and input flows between the two countries.
20
STEVEN GLOBERMAN AND PAUL STORER
3. STATISTICAL EVIDENCE OF INTEGRATION USING QUANTITY AND PRICE MEASURES In this section, we first identify evidence of recent regional economic integration using “quantity” measures. As noted earlier, observers have typically looked at trade and FDI with reference to quantity-based measures of economic integration.
3.1. Trade Flows Obviously, absolute quantity flows are not meaningful, since they fail to standardize for the potential volume of flows given economic growth in the regional economies. One approach that has been taken to identify the volume of trade flows relative to the size of the trading bloc is to estimate gravity models.3 Such models effectively hold constant the influence of economic growth on trade flows so that the impact of relevant trade agreements on trade flows can be identified. In this regard, Lee (2002) reviews a number of statistical studies suggesting that the buoyant U.S. economy and the depreciation of the Canadian dollar were mainly responsible for a dramatic increase in Canadian exports to the United States in the 1990s. The CUSTA and NAFTA Agreements, in contrast, are estimated to have increased Canadian exports to the United States by only around 9%. Estimates of so-called border effects build upon the structure of gravity models by comparing trade flows across countries or regions to flows within countries or regions where the “potential” volume of trade is identified through a gravity model. McCallum (1995) and Helliwell (1998) document the existence of substantial border effects in the Canada-U.S. context. Simply put, there is much less North-South trade relative to intra-Canada trade than one would expect given the size of the U.S. economy relative to provincial economies in Canada. Helliwell also examines the impact of the CUSTA on border effects for the period 1988–1996. He finds that the average border effect was constant from 1988 to 1990 and then fell substantially from 1990 to 1993. The border effect in 1996 was the same as in 1993 and about 60% of the estimated 1990 value. Interestingly, Helliwell also finds that export border effects fell more than import border effects over the sample period. Given the lower average tariff levels in the U.S. compared to Canada in the period immediately preceding the CUSTA, one would have expected the direction of the relative border effect change to be the opposite of what Helliwell finds. That is, one might have expected a greater relative increase in trade going from the U.S. to Canada. The inference one might draw is consistent with Lee’s (2002) previously cited conclusion. Namely, that the CUSTA, per se, has been only a small factor influencing bilateral trade flows with
Canada-U.S. Economic Integration Following NAFTA
21
Table 1. Distribution of U.S. Exports by Country (Percent of Total).
Canada Mexico Europe Other
1980
1981–1985
1986–1990
1991–1995
1996–2001
17 7 23 53
19 6 23 52
20 7 25 48
18 9 24 49
19 9 24 48
Source: U.S. Department of Commerce, Bureau of Economic Analysis.
other, more important factors strongly encouraging Canadian shipments to the United States. The findings of the aforementioned statistical studies are largely supported by data presented in Tables 1–4. These data look at relative trade flows, in effect normalizing for factors that would increase trade generally on both a bilateral and multilateral basis. Specifically, Tables 1 and 2 report the distribution of U.S. merchandise exports and imports by country or region for select periods since 1980. Tables 3 and 4 report similar data for Canada. With respect to U.S. exports, Canada became a relatively more important trading partner over the period 1980–1990; however, Canada became marginally less important in relative terms in the post-1990 period. With respect to imports, Canada was relatively more important at the end of the sample period than at the beginning, although there is essentially no increase in relative imports from Canada from the period immediately preceding CUSTA to later periods. Tables 3 and 4 identify changes in Canada’s trade patterns. Relative exports from Canada to the United States rose sharply over the period 1980–2001. The increase was fairly consistent over the entire period suggesting that CUSTA, per se, may not have been a dominant influence on the behavior of Canadian exports. This inference is further supported by the data in Table 4 showing that the U.S. share of Canadian imports increased modestly in the latter half of the 1990s after declining over the period 1986–1995. The latter is the period during which the impacts of CUSTA would have been most prominent. Conversely, while the NAFTA was implemented in 1994, it was Table 2. Distribution of U.S. Imports by Country (Percent of Total).
Canada Mexico Europe Other
1980
1981–1985
1986–1990
1991–1995
1996–2001
16 5 17 62
18 6 20 56
17 5 22 56
17 7 21 55
18 8 21 53
Source: U.S. Department of Commerce, Bureau of Economic Analysis.
22
STEVEN GLOBERMAN AND PAUL STORER
Table 3. Distribution of Canada’s Exports by Country (Percent of Total).
U.S. Mexico Asia Other
1980
1981–1985
1986–1990
1991–1995
1996–2001
61 1 3 35
70 1 3 26
72 1 4 23
79 0.4 4 17
85 0.5 3 12
Source: International Monetary Fund, Direction of Trade Statistics, various issues.
not expected to have much significance directly for Canada-U.S. trade patterns, as CUSTA essentially eliminated tariff barriers to Canada-U.S. trade. In summary, the main observed change in bilateral trade patterns is a major increase in Canadian exports to the U.S. Other dimensions of bilateral trade changed only modestly or not at all in the post-CUSTA period. This pattern is not dissimilar to U.S. trade patterns with Mexico. Specifically, Mexico’s share of total U.S. imports increased by only one percentage point in the post-NAFTA period, while the share of total U.S. exports going to Mexico also increased by only one percentage point in the post-NAFTA period. As in the case of Canada, the share of Mexican exports going to the U.S. increased consistently over the period 1980–2001. Conversely, imports from the United States as a share of Mexico’s total imports actually declined in the latter half of the 1990s, i.e. the period following the implementation of NAFTA.4 While the preceding discussion focused on merchandise trade, exports and imports of services are also relevant.5 In the case of services, however, the trend towards greater Canadian export intensity with the U.S. is much less marked than in the case of goods. For example, in 1989, service exports to the U.S. accounted for almost 57% of Canada’s total service exports. In 2000, service exports to the United States accounted for around 59% of total service exports. For the two same years, service imports from the United States accounted for around 62 and 63%, respectively, of total Canadian service imports. Moreover, the Table 4. Distribution of Canada’s Imports by Country (Percent of Total).
U.S. Mexico Asia Other
1980
1981–1985
1986–1990
1991–1995
1996–2001
68 0.5 2 29.5
69 1 3 27
65 1 6 28
64 2 8 26
68 3 8 21
Source: International Monetary Fund, Direction of Trade Statistics, various issues.
Canada-U.S. Economic Integration Following NAFTA
23
category of service trade (both exports and imports) enjoying the fastest growth over the period 1989–2000 was transportation, arguably reflecting the growth in merchandise trade. The information summarized in this section supports Helliwell’s conclusion that the main border effect change in the 1980s and 1990s was manifested in increased Canadian merchandise exports to the United States, although the importance of formal trade agreements as a contributor to reduced border effects is, at best, questionable. We explore this issue further in a later section.
3.2. Foreign Direct Investment Gravity-type models have also been used to estimate the impact of formal trade agreements on FDI flows. In the FDI models, national or regional GDP variables are used to standardize for the influence of private sector economic determinants of FDI flows. Several gravity models identify a potential influence of CUSTA and NAFTA on bilateral FDI flows. For example, Buckley, Clegg, Forsans and Reilly (2000) find that U.S. FDI into Canada was encouraged by the two formal trade agreements, although changes in the exchange rate also fostered FDI. Globerman and Shapiro (1999) identify an increase in Canadian inward and outward FDI flows in the period subsequent to CUSTA (including the period covered by NAFTA) holding other determinants of FDI flows constant. Outward flows are larger than inward flows; however, there are reasons to doubt whether the CUSTA and NAFTA, per se, explain the increase in outward FDI, since growth in outward FDI from Canada was primarily directed at Western Europe in the late 1980s and early 1990s. Eden and Monteils (2000) are also skeptical about the impact of the formal free trade agreements on the magnitude of bilateral FDI flows. Specifically, they conclude that MNCs making intra-regional foreign investments in North America engaged in “locational reshufflings” as they rationalized their investments on a continental basis. The statistical evidence questioning the impact of the CUSTA and NAFTA on bilateral FDI flows is basically supported by the data reported in Tables 5–8. Again, we focus on relative magnitudes, thereby indirectly performing a simple gravity-model analysis. Specifically, Tables 5 and 6 report the geographic distribution of inward and outward FDI for Canada. The data show that the U.S. became a relatively less important source for inward FDI to Canada, as well as a less important destination for Canadian outward FDI over the period 1986–2001. Western Europe became an increasingly important source and destination of Canadian FDI flows. Tables 7 and 8 show similar information for the United States.6 The main point is that Canada became relatively less important as an
24
STEVEN GLOBERMAN AND PAUL STORER
Table 5. Geographic Distribution of Inward FDI Stock in Canada (Percent).
U.S. U.K. Other Europe Other
1986
1990
1995
2001
72 12 10 6
64 13 11 12
67 8 13 12
67 8 19 6
Source: Evans (2002) and Industry Canada (2001).
FDI partner for the United States over the period 1980–2001. While we do not show data for Mexico, the U.S. share of Mexico’s total FDI stock decreased from around 74% in 1980 to around 62% in 1998. In summary, the bilateral intensity of FDI flows actually decreased over the period following the CUSTA. In the context of the earlier data on bilateral trade flows, the only clear indication of closer bilateral economic relations subsequent to the CUSTA is the increased share of Canadian exports going to the United States.
3.3. Labor Mobility The CUSTA facilitated temporary entry of workers on a reciprocal basis between Canada and the U.S. by waiving requirements for a non-immigrant’s visa, prior petition, and labor certification or other prior approval, as long as appropriate documentation was presented establishing citizenship or permanent residence, and showing professional engagement in one of the occupations listed in the qualifying occupation schedule. The NAFTA visa provision extends the CUSTA provisions for the two countries. The TN-1 (NAFTA) visa has compelling advantages compared to the H-1B visa governing the temporary entry of professionals.7 The inference one might draw is that the implementation of the TN visa facilitated Table 6. Geographic Distribution of Canadian FDI Stock Abroad (Percent).
U.S. U.K. Rest of Europe Other
1986
1990
1995
2001
69 8 6 17
61 14 7 18
52 10 11 26
51 10 13 26
Source: Evans (2002) and Industry Canada (2001).
Canada-U.S. Economic Integration Following NAFTA
25
Table 7. Geographical Distribution of Inward FDI in the U.S. (Percent).
Canada U.K. Netherlands Japan Other
1980 (Stock)
1986 (Stock)
1992 (Stock)
1996–2001 (Flow)
14 17 25 6 38
9 25 19 12 35
9 23 15 23 30
7 79a 5 9
Source: Graham and Krugman (1995) and the U.S. Department of Commerce, Bureau of Economic Analysis. a For 1996–2001, the number shown is for all of Western Europe.
the intra-North American mobility of labor given economic pressures towards greater integration of labor markets in North America. Table 9 provides a snapshot of the migration of temporary workers from Canada to the United States compared to migration of temporary workers to the United States from all countries. The table highlights migration under three main categories: the H1-B visa; intra-company transferees and the NAFTA visa.8 These are the three main sources of non-permanent immigrants entering under a work-related visa status. Unfortunately, consistent data on H1-B visa issuance are unavailable for a time series sufficiently long to examine the pre and post-CUSTA experience. Hence, a broader category entitled “Temporary Workers and Trainees” is referenced for the 1986 observations. The last column of Table 9 reports the percentage of the total number of non-permanent immigrants under the three categories that originated from Canada. As seen, there was essentially no change in the percentage from 1986 to 1992; however, there was a significant increase (from around 11.6% to around 18.7%) from 1992 to 2000. The data on business-related non-permanent immigration therefore suggest that intra-North American labor markets became more integrated in both absolute and relative terms. The main reason is the substantial growth in Canadians entering Table 8. Geographical Distribution of U.S. Outward FDI (Percent).
Canada Western Europe Japan Latin America Other
1980 (Stock)
1986 (Stock)
1996–2001 (Flow)
21 44 3 20 12
17 46 5 18 14
10 49 3 16 22
Source: Rugman and Gestrin (1994) and the U.S. Department of Commerce, Bureau of Economic Analysis.
26
STEVEN GLOBERMAN AND PAUL STORER
Table 9. Non-Permanent Immigration from Canada to the U.S. Year
H1-Ba
Intra-Company Transferees
TN
Total
Percentage from Canada
1986
Canada Total
12,382 85,359
5,045 66,925
N/A N/A
17,427 152,284
11.4
1992
Canada Total
4,633 110,193
5,664 75,347
12,675 12,675
22,972 198,215
11.6
2000
Canada Total
12,929 355,605
19,221 294,658
110,740 113,460
142,890 763,723
18.7
Source: U.S. Immigration and Naturalization Service, Statistical Yearbook of Immigration and Naturalization Service. Washington, DC: U.S. Government Printing Office, various years. a In 1986, this category is “Temporary Workers and Trainees.”
the United States under the NAFTA TN visa-status. While Congress sets a limit to the number of H1-B visas that can be issued in any year, this limit is arguably well below the number of visas that would be issued if “market forces” only determined their issuance. That is, U.S. employers would arguably hire a far greater number of foreign workers than they actually do, while the supply of foreigners willing (and qualified) to enter the U.S. under that visa status is also greater than the number that actually enter. As Table 9 documents, the number of Canadians entering under the H1-B status was essentially unchanged between 1986 and 2000, whereas the total of all entries under that status more than quadrupled. Put simply, limited entry under the H1-B status affects entry into the U.S. primarily from other countries besides Canada. As such, Table 9 arguably overstates the implied economic integration between Canada and the U.S. In short, the information on Table 9 should not be seen as convincing evidence of market forces promoting closer labor market integration between Canada and the U.S. relative to other countries. Furthermore, the number of skilled workers migrating to the U.S. from Canada is relatively small compared to the domestic work force in each country. For example, total civilian employment in the United States was around 135 million in 2000. Temporary immigration from Canada in that year therefore represents less than one-tenth of 1% of the total employed labor force in the United States.9
3.4. Convergence of Prices of Goods and Services An important measure of increased economic integration is the convergence of costs paid by businesses and prices paid by consumers. With integrated markets,
Canada-U.S. Economic Integration Following NAFTA
27
there should be a single price for any commodity if transportation costs, pricecost margins and taxes don’t impede price equalization. Hence, the degree of convergence of prices and costs is a measure of the success of trade liberalization and integration.10 If the law of one price holds for all pairs of goods prices in two countries and each good has the same weight in aggregate price level indices, then purchasing power parity (PPP) will hold between the two countries. PPP is a relationship between aggregate price levels such as the GDP deflator and states that $U.S. P U.S. = Q × P CAN × $CAN where Q is the real exchange rate. Most existing tests of price convergence such as Johnson (1990) have used overall price indices such as the GDP deflator for PU.S. and PCAN . A recent Bank of Canada survey of PPP by Lafrance and Schembri (2002) concluded there are frequent departures from PPP and that there are fairly persistent cycles in the exchange-rate adjusted relative prices of goods and services between Canada and the United States. Recently, studies of commodity-level prices have become more common. For example, Engel and Rogers (1996) studied ratios of CPI indexes (including 14 categories of prices) for 23 different cities in Canada and the United States. They found, when comparing cities in different countries, that the deviations from the law of one price are larger than predicted based on distance alone. Engel and Rogers concluded that the combination of sticky nominal prices and exchange rate volatility explained some of this border effect. A separate study by Engel and Rogers (1998) found greater price divergence between than within regions. This is analogous to border effects for trade and FDI flows. Their findings cannot be attributed to trade liberalization per se because distance is shorter within regions, regions are more likely to use a common currency, and distribution networks are more likely to be integrated within a region. Hence, the tendency to have lower trade barriers within a region is only one possible explanation for lower price dispersion within regions than between regions. The effect of trade liberalization on price convergence was evaluated by Beiling Yan (2002). Yan used paired Canada-U.S. final user prices of 168 private business commodities for 1985, 1990, 1993, and 1996 to calculate deviations from the law of one price for each commodity. Yan looked at averages of these deviations for three types of general groups: non-tradable commodities such as services and traderestricted goods such as milk, differentiated tradable goods such as appliances and clothing, and homogeneous tradable goods such as rice, fresh fruit, and fish. The results of Yan’s study that are relevant to the CUSTA are illustrated by Fig. 1. This chart shows the average values of the logarithm of U.S.-Canada price differences
28
STEVEN GLOBERMAN AND PAUL STORER
Fig. 1. Average of Logged Canada-U.S. Deviations from the Law-of-One Price. Source: Table 6 of Yan (2002).
by commodity for Yan’s three groups. The figures are available for 1985, 1990, 1993, and 1996. The declining absolute value of the average difference for homogeneous goods from 1990 through 1996 is consistent with CUSTA contributing to price convergence. Of course, the absolute values of average price differentials also fall for differentiated tradable goods over this same period, and the rejection of the null hypothesis of a zero average difference reflects the distance from zero after 1985. What is striking about this graph is the common V-shaped pattern that mirrors the cycle of appreciation of the Canadian dollar from 1985 through 1990 followed by depreciation. This is consistent with the sticky nominal price/volatile exchange rate explanation for deviations from the law of one price outlined by Engel and Rogers (1996). Given that both non-tradable and homogeneous tradable good prices are following the same patterns, it seems unlikely that trade liberalization is the common underlying cause of all three patterns. What average price deviations in Fig. 1 don’t reveal is the degree of convergence for individual goods. The patterns observed by Yan for the variance of individual deviations from the law of one price are shown in Fig. 2. While trade liberalization should lower this variance most for homogeneous traded goods, the variance of individual deviations from the law of one price generally rose for homogeneous traded goods, and by 1996 was roughly double its 1985 value. In contrast, the variance of individual deviations actually fell for both non-tradable commodities and differentiated tradable goods, with the non-tradable commodities showing the largest decline in variance. It is difficult to believe that
Canada-U.S. Economic Integration Following NAFTA
29
Fig. 2. Variance of Canada-U.S. Deviations from PPP. Source: Table 6 of Yan (2002).
the convergence observed for these two categories is due to trade liberalization after CUSTA. The patterns in Fig. 2 are consistent with Yan’s results for homogeneous traded goods in Fig. 1. The key to reconciling her results is the recognition that the variance of price differences can increase even as the average value of the differences gets closer to zero. Quite simply, what seems to have happened is that deviations from the law of one price have become larger in absolute value for homogeneous goods, but there has also been an increased tendency for negative and positive deviations to average to zero for homogeneous traded goods. This result was also found by Crucini, Telmer and Zachariadis (2001), who examined the law of one price for more than 5,000 goods and services in European countries in 1975, 1980, 1985, and 1990. Specifically they found little decline in the dispersion of prices over time but did find that weighted averages of price differences tended to be close to zero. This suggests that similar factors may have limited the impact of integration on goods price convergence in Europe in the 1970s and 1980s and in North America in the 1990s. Overall, the results of this section reveal no evidence that CUSTA led to increased convergence of prices, at least at the individual product level. 3.5. Convergence of Costs of Labor To the extent that labor and capital are increasingly able to move across the Canada-U.S. border, there should be some convergence in labor costs. This may happen because workers migrate to higher wage employers or because firms
30
STEVEN GLOBERMAN AND PAUL STORER
invest in locations with lower labor costs. The CUSTA and NAFTA both increased labor mobility for workers with academic degrees and increased the ability of firms to seek low-cost production locations by facilitating the flow of finished and intermediate goods across the border. To date, there has been relatively little analysis of the labor market effects of trade liberalization on relative wages in Canada and the U.S. Gaston and Trefler (1997) found that the high interest rates associated with the anti-inflation policy of the early 1990s had a greater impact on the Canadian labor market than the trade liberalization due to CUSTA. Beaulieu (2000) presented evidence that “. . . the CUSTA tariff reductions had no effect on average annual earnings in the manufacturing industries for either skilled or less skilled workers.” He attributed this finding to the fact that Canadian real wages didn’t vary much after trade liberalization, so that there is not much variation to attribute to trade liberalization. These existing studies of Canadian earnings do not permit analysis of relative labor costs in Canada and the U.S. This is partly due to the lack of comparable occupational data, at least for long periods. Canada does not provide earnings data by occupation, and the industry data used in studies such as Gaston and Trefler and Beaulieu are not always comparable to industry definitions in the U.S. This latter problem will be reduced as data using the NAICS industry classifications become available for both countries in 2003. One current source of Canada-U.S. labor cost comparisons is the U.S. Bureau of Labor Statistics (BLS) indexes of hourly manufacturing compensation that are available for the U.S. and several foreign countries, including Canada. These series are currently available from 1991 through 2001, and the results are shown in Fig. 3. This graph shows a divergence in labor costs over the post-CUSTA period, with declining relative labor costs in Canada.11 Similar trends appear when average weekly earnings are compared for manufacturing, transportation equipment, and lumber in Fig. 4. This chart plots the ratios of U.S. to Canadian earnings and for all three industrial sectors the weekly wage is increasing in the U.S. relative to Canada so by the end of the sample period, labor costs are uniformly lower in Canada. While the levels of relative labor costs differ by industry, the trends are almost identical and seem to reflect the large depreciation of the Canadian dollar combined with the relative stickiness of nominal wages.
3.6. Convergence of Costs of Capital Integration of the Canadian and U.S. economies should lead to a convergence of costs of capital and rates of return on investment. At the margin, the cost of capital should equal the return on capital. Cross-border investment flows should
Canada-U.S. Economic Integration Following NAFTA
31
Fig. 3. BLS Indexes of Hourly Manufacturing Compensation Costs (U.S. Dollar Basis). Sources: BLS web site, series INU0007US0, and INU0007CA0.
tend to equate these returns and costs. Divergence between returns on capital in the two countries could reflect barriers to non-resident investment in certain sectors (such as banking, broadcasting, or health care in Canada) or risk premia related to exchange rate or political risk.
Fig. 4. Ratio of U.S. to Canadian Average Weekly Earnings, 1991–2001 (in U.S. Dollars).
32
STEVEN GLOBERMAN AND PAUL STORER
One method of examining the convergence of rates of return in Canada and the U.S. is to examine firm-level data on profitability such as return on equity (ROE) or return on investment (ROI). The Compustat database has measures of these two returns using the following definitions: ROE = ROI =
Income Before Extraordinary Items Common Equity as Reported Income Before Extraordinary Items Long-Term Debt + Common Equity + Preferred Stock + Minority Interest
Values for these two measures of the return on capital invested are presented in Fig. 5. The U.S. series is the average of returns for the companies in the S&P 500 index while the Canadian series is for the TSE 300 index.12 Unfortunately, the Compustat data for Canada begin in 1988 (for ROI) and 1989 (for ROE) and this does not permit a long-term comparison. The bottom panel of Fig. 5 examines the spread between returns in the U.S. and Canada and shows little evidence of convergence of rates of return on capital, with the possible exception of 2001 where the deeper economic downturn in the U.S. is apparent. Another source of profitability data is the national accounts. Professor John Rodgers of Western Washington University has compiled comparable measures of the profit rate for Canada and the United States. Rodgers defines the net profit rate as: NPR =
Output − Total Compensation − Depreciation Net Capital Stock
One advantage of this approach to measuring the return on capital is that it does not require the use of firm-level accounting data but rather relies on national accounts data. Recent concern over standards at public accounting firms has led to increased reliance on profitability measures based on national accounts. Rodgers’ data (shown in Fig. 6 for the manufacturing sector) does show a definite trend toward convergence of net profit rates in Canada and the U.S., but it appears that this trend mainly occurred before 1980. Moreover, the convergence is mainly due to a marked decline in the net profit rate in U.S. manufacturing from 1965 through 1980. While increasing integration during this period (particularly that related to the Auto Pact) could lower differences in rates of return, it doesn’t seem plausible that this integration-driven equalization would happen almost exclusively through adjustment of the net profit rate in the U.S.
Canada-U.S. Economic Integration Following NAFTA
Fig. 5. Returns on Equity and Investment.
33
34
STEVEN GLOBERMAN AND PAUL STORER
Fig. 6. Manufacturing Net Profit Rates in Canada and the U.S.
4. INTERPRETING THE EVIDENCE ON INTEGRATION Our analysis to this point leads to the inference that economic integration between Canada and the United States in the post-CUSTA period has been relatively
Canada-U.S. Economic Integration Following NAFTA
35
modest. Certainly, one possible explanation of these findings is that the Canadian and U.S. economies were already so tightly integrated prior to the CUSTA that additional efforts by governments and business to link the two economies further were likely to have quite modest results.13 In addition, existing manufacturing tariffs in the two countries were relatively low, especially those facing Canadian exporters to the United States. Further insight into possible explanations might be gained from identifying changes in the mix of Canada’s merchandise exports and imports and the Canada-U.S. exchange rate.
4.1. U.S. Demand Patterns If the CUSTA was an important influence promoting the increased bilateral concentration of trade with the U.S., one would expect to see relatively rapid growth in Canadian export sectors that enjoyed especially liberalized access to the U.S. market as a result of the CUSTA. In this regard, Table 10 reports the sectoral distribution of Canada’s goods trade at a broad level of aggregation for 1989 and 2000. Specifically, it shows the share of exports and imports accounted for by seven broad industrial sectors in Canada.14 One point to note is that the share of each sector in total imports is relatively unchanged across the two years. However, the sectoral distribution of exports does exhibit significant differences. In particular, industrial goods and materials and machinery and equipment appear to be the sectors experiencing the greatest change in export intensities over the 1990s. Industrial goods include intermediate inputs such as metals, chemicals, plastics and fertilizers. Machinery and equipment includes telecommunications equipment, computers and other office machines. In principle, there is no reason to expect the CUSTA or the NAFTA to favor increased Table 10. Share of Canada’s Total Goods Trade (%). Exports
Agriculture and fishing Energy Forestry Industrial Machinery and equipment Automotive Consumer goods
Imports
1989
2000
1989
2000
7.9 9.3 14.6 22.0 16.3 23.1 1.8
6.6 12.6 9.9 15.5 25.3 23.1 3.5
6.0 4.5 1.0 19.3 31.1 22.9 10.8
5.1 4.9 0.9 19.4 33.8 21.3 11.0
Source: Government of Canada, Department of Foreign Affairs and International Trade (2001).
36
STEVEN GLOBERMAN AND PAUL STORER
exports of machinery and equipment relative to industrial goods, especially as computers and office machinery had very low U.S. tariffs prior to CUSTA. Moreover, the major barrier to exporting telecommunications equipment was arguably local purchasing preferences by the Bell Operating Companies in the U.S., and this preference would not have been addressed by the formal trade agreements. One possible explanation for the relatively rapid growth of exports in machinery and equipment was the capital-spending boom in the United States during the 1990s, especially in technology-intensive sectors such as telecommunications and Internet-related businesses.15 It is interesting in this regard that telecommunications equipment was one of the few “high-tech” sectors in which Canadian suppliers increased their share of overall U.S. imports over the period 1980–1999.16 One is tempted to conclude that the absolute and relative surge of Canadian exports of machinery and equipment largely reflects the remarkable success of Nortel over the period of the 1980s and 1990s combined with the explosive growth in infrastructure spending by U.S. telecommunications carriers and by other businesses expanding their own in-house telecommunications and data transmission capabilities.17 Indeed, the 39% decline in exports of telecommunications between 2000 and 2001 contributes much of the 7.6% decline in exports of machinery and equipment over the same period.
4.2. The Canada-U.S. Exchange Rate An obvious candidate to explain the observed international trade pattern for Canada is the Canada-U.S. exchange rate. Specifically, the long and sustained decline in the value of the Canadian dollar from the mid-1970s through 2001 could well be a contributor to the increase in Canada’s export intensity with the United States.18 In this regard, it might be noted that the substantial decline in the value of the Canadian dollar relative to its U.S. counterpart is in contrast to the general appreciation of major European currencies and the Japanese Yen relative to the U.S. dollar over the same period. The importance of a declining exchange rate as an important stimulus to Canadian exports is further suggested by the fact that productivity growth did not enhance Canada’s underlying “cost-competitiveness” relative to the U.S. over the period in question. For example, The Centre for the Study of Living Standards (2002) reports that real GDP per hour worked in the United States increased at a 1.5% average annual rate over the period 1973–2001, whereas real GDP per hour worked in Canada increased at a 1.33% average annual rate over the same period. The marked decline in the value of the Canadian dollar is also consistent with the increase in temporary migration from Canada to the U.S. noted earlier. As
Canada-U.S. Economic Integration Following NAFTA
37
Harris (2000) argues, the decline in the Canadian dollar represents a substantial decline in relative living standards for Canadians and provides a strong impetus for talented Canadians to move south of the border.19 Harris also suggests that the decline in the value of the Canadian dollar should have encouraged an increase in inward FDI in Canada, as the international value of Canadian assets would also have declined. Since the value of the Canadian dollar declined even more against other major currencies, the impetus to inward FDI would not have been unique to U.S. investors. Moreover, there are clearly other factors motivating FDI flows that would need to be standardized before one could identify the influence of the exchange rate with any confidence.20 In short, it is likely that the marked decline of the Canadian dollar has influenced trade-related measures of integration. Nevertheless, it is still of interest that Canada’s substantially increased export intensity with the United States has not been matched by an increase in FDI intensity given the traditional assumption that trade and FDI are net complements.21 Clearly, Canadian firms were able to increase their exports to the U.S. market without proportionately increasing their direct investments in the U.S. As a consequence, while the dollar value of the FDI stock increased among all of the NAFTA partners, relatively more investments were directed outside of North America. Part of the rationalization process also involved more “contracting-out” and less intra-firm trade by North American MNCs (Acharya, Sharma & Rao, 2001). The behavior of the exchange rate may also explain why there seems to be little evidence of increased convergence of prices of goods or costs of productive inputs since the CUSTA was implemented. As Engel and Rogers (1998) note: nominal prices may be sticky within each country so that volatile exchange rates lead to departures from the law of one price. Even if the exchange rate only accounts for half of the departures from goods price convergence, this is still a substantial effect. Moreover, given that nominal wages tend to be even stickier than prices, this explanation likely accounts for much of the lack of labor cost convergence, as well. Also, exchange rate variability might affect capital cost convergence if exchange rate uncertainty creates a risk premium that discourages foreign direct investment. Figure 7 illustrates the behavior of the Canada-U.S. bilateral exchange rate in the post-war period. The three panels of this graph examine thirteen post-CUSTA years, the thirteen years prior to CUSTA, and the preceding portion of the postwar period. This first panel for 1945–1975 loosely corresponds to the Bretton Woods period in which fixed exchange rates were more common. Simple comparison of standard deviations to measure exchange rate volatility shows that the post-CUSTA period has the highest level of exchange rate volatility.22 We would expect this to translate into larger deviations from the law of one price given sticky nominal prices.
38
STEVEN GLOBERMAN AND PAUL STORER
Fig. 7. Behavior of the Nominal Exchange Rate.
Canada-U.S. Economic Integration Following NAFTA
39
The simple measure of exchange rate volatility in Fig. 7 may be relevant for price convergence in the markets for goods or labor, but the standard deviation may not be the most appropriate measure of exchange rate uncertainty for capital market decisions. If there is a predictable trend to the exchange rate, the exchange rate might fluctuate a great deal without causing uncertainty for firms contemplating direct investment. We address this possibility by estimating the forecasting equation: E t = ␣ + E t−1 + u t The measure of residual uncertainty in this graph is simply the value of ut obtained after an ordinary least squares (OLS) regression is fitted to this forecasting equation.23 Our results show an increase in exchange rate uncertainty in the post-CUSTA period.
4.3. Comparison with the European Experience When trying to explain the apparently small relative impact of recent de jure North American trade liberalization, the European experience is instructive. Specifically, integration among the original, major European economies also “stalled out,” notwithstanding continuing formal agreements to promote increased trade and investment within Europe. For example, Germany’s trade intensity with its four major EU trading partners (U.K., France, Italy and Netherlands) was actually lower at the end of 2000 than it was at the beginning of 1980 (Globerman, 2002). For all of Western Europe, intra-regional trade as a share of total trade increased from around 54% to around 67% from 1958 to 1973 but then increased to only 70% by 1993 (Brunelle & Deblock, 2001). Dunning (2000, p. 131) concludes from evidence for the first phase of macroeconomic regional integration in Europe (in the late 1960s and 1970s) that while the removal of intra-tariffs and the establishment of a common external tariff, increased both intra-and external FDI – and sometimes significantly soother determinants of FDI (e.g. market size, market growth, relative factor costs, agglomeration economies, etc.) were at least as, if not more, important, particularly once the initial (and sometimes once-for-all) effects of the European Common Market had worn off. Furthermore, over the period of the establishment of the Common Market (1958–1985), intra-EC trade grew much faster than intra-EC FDI. He interprets the available evidence after 1985 as indicating that the Internal Market Program (IMP) stimulated both extra- and intra-regional FDI (the former more than the latter); however, other variables had more significant impacts.
40
STEVEN GLOBERMAN AND PAUL STORER
In short, the Western European experience with regional economic integration is not dissimilar to the Canada-U.S. experience. Specifically, border effects persist even in the face of increasingly ambitious formal efforts to integrate the regional economy. One is tempted to conclude that there are strong diminishing returns to reducing border effects through traditional tools of trade liberalization, especially after significant initial measures to reduce barriers to trade and investment have been taken. 4.4. Comparison with Earlier Integration in North America Prior to CUSTA and NAFTA, the deepest economic integration experienced in North America was introduced by the Canada-U.S. Auto Pact of 1965. According to Hart (2002), prior to the Auto Pact “Canadians paid considerably more for cars than did Americans” and “Canadian workers earned about 30% less than their U.S. counterparts.” After the Auto Pact came into effect, there was marked convergence of wages, prices, and rates of return as Canadian producers rationalized product lines and increased production runs. It is also noteworthy that the Centre for the Study of Living Standards (2002) reports an increase in total factor productivity for the Canadian motor vehicle industry from 73% of the U.S. level in 1961 to 107% in 1995, with most of the increase observed between 1961 and 1973. These results for the managed sectoral free trade of the Auto Pact stand in contrast to the relative absence of price, wage, and rate of return convergence after CUSTA and NAFTA. Also, while exploiting economies of scale and enhancing productivity in Canada was a major objective of both the Auto Pact and the CUSTA, there has been some debate over whether CUSTA has actually contributed to improved productivity in Canada. While Trefler (2001) found faster productivity growth for those industries that were most affected by CUSTA tariff cuts,24 he also found that “. . . it appears that scale and capital deepening are not part of the FTA story.” Rather, the increases were due to “a mix of plant turnover and rising technical efficiency within firms.” Taken together, the evidence on prices, wages, rates of return and productivity suggest that the impact of CUSTA was rather different from that of the Auto Pact. Further study is required to identify the causes of the observed differences.
5. IMPLICATIONS FOR FUTURE POLICY INITIATIVES A number of observers have recently argued that major new initiatives must be undertaken to promote any significant further movement towards economic
Canada-U.S. Economic Integration Following NAFTA
41
integration between Canada and the United States. They include: adopting a customs union, liberalizing goods trade in excluded sectors such as agriculture, liberalizing trade in services, removing ownership restrictions in protected sectors (such as broadcasting and finance), allowing for freer movement of labor, and adopting a common currency. It is obviously impossible for us to critically assess these and other strategies for future integration between Canada and the United States in this paper. It is certainly reasonable to suggest that efforts to streamline customs clearance procedures, border crossings for tourists and surface transporters, and the like, as well as continuing attempts to mitigate the impact of national and sub-national regulations on cross-border trade are worthy policy objectives, especially in light of the challenges to cross-border commerce posed by heightened security considerations. We would also support initiatives to open previously protected sectors (mostly in Canada) to foreign direct investment including telecommunications and transportation. However, even in these areas, one can only guess at the likely response of North American businesses to the policy initiatives. One suggestion is for Canada to vigorously promote the adoption of a common currency with the United States or, failing that, to adopt the U.S. dollar as its domestic currency.25 The suggestion rests upon the notion that exchange rate instability is a significant discouragement to trade and direct investment flows. While the magnitude and precise nature of the linkage between a common currency and regional integration are uncertain, the literature does support a conclusion that a significant positive linkage exists.26 As noted earlier, the Canada-U.S. exchange rate has become more volatile in the post-CUSTA period. This result is additional indirect evidence that the two economies have not become substantially more integrated since the CUSTA, and this finding suggests that regional economic integration might have been both broader and deeper had Canada adopted the U.S. dollar as its domestic currency or negotiated a common currency with the United States.
6. CONCLUSIONS This paper outlines a comprehensive methodology for assessing the degree of economic integration between two countries. Application of this methodology reveals that Canada and the U.S. experienced very modest additional integration in the period following the implementation of the CUSTA and the NAFTA. Nevertheless, there may still be significant gains from bilateral trade, since the evidence suggests that economies of scale have yet to be exploited outside of sectors such as automobiles. Moreover, microeconomic efficiency has arguably
42
STEVEN GLOBERMAN AND PAUL STORER
not been enhanced through a closer alignment of prices and costs within the North American economy. An important potential explanation for continued segmentation of markets within North America is the volatility of the Canada-U.S. exchange rate. Opponents of a common currency point to the need for exchange rate flexibility in Canada to enhance adjustments to external influences such as terms of trade shocks. However, if the increase in the level of Canadian GDP due to a common currency is sufficiently high, it might compensate for the social cost of greater business cycle fluctuations in Canada under a common currency vs. a floating currency regime.
NOTES 1. For an overview of the NAFTA visa provision and subsequent bilateral migration under that and related provisions, see Globerman (2000). 2. See, for example, DeVoretz and Coulombe (2002) and Globerman (2000). 3. A technical discussion of gravity models and their interpretation can be found in Helliwell (1998). 4. See Globerman (2002) for a fuller discussion of recent U.S.-Mexico trade and investment flows. 5. By way of illustration, in 2000, exports of services accounted for almost 12% of Canada’s total exports of goods and services, while imports of services accounted for around 8.6% of total imports. See Government of Canada, Department of foreign affairs and international trade (2001). 6. For the period 1996–2001, the FDI data are estimates of flows rather than stocks. 7. For a detailed discussion of the CUSTA/NAFTA immigration provisions and an assessment of their effects, see Globerman (1999). 8. It should be noted that the reported TN visa number does not control for multiple border crossings by the same visa holders. Thus, a person living in Windsor and commuting five days a week to work in Detroit would generate five visa counts a week. The assumption made here is that the resulting upward bias has not increased over time. Hence, the increase over time in TN visas reflects more than an increase in multiple crossings. 9. There has been a much faster growth of TN visas issued by the U.S. to Canadians than vice versa. For example, the number of permits issued by the U.S. increased from 2,677 in 1989 to 26,794 in 1996, while the number of visas issued by Canada rose from 2,748 to 13,337 (Egelton & Wojciech, 1999). 10. Even if prices were not equalized, we would expect them to move closer over time with the elimination of trade barriers, other things constant. 11. These charts use the market exchange rate to convert Canadian dollars, and the decline in relative costs in Canada reflects departures from PPP for overall prices. 12. Both returns are in local currencies, so we do not correct for possible time-varying country risk premia. 13. This assertion is made by Helliwell (2001), among others. 14. The percentages shown do not aggregate to 100 since miscellaneous industrial categories are excluded.
Canada-U.S. Economic Integration Following NAFTA
43
15. This explanation is suggested in Government of Canada, Department of Foreign Affairs and International Trade (2001). 16. Ibid. 17. Nortel is Canada’s largest supplier of telecommunications equipment and a world leader in the development of digital switching systems. 18. In March 1973, the U.S. dollar was worth CDN $0.9967. The average value of the U.S. dollar in 2001 was CDN $1.5487. See Council of Economic Advisers (2002). 19. The strength of the declining Canadian dollar as an impetus to temporary migration flows is empirically uncertain. For a review of evidence on the determinants of temporary worker flows within North America, see Globerman (1999). 20. There are also strong theoretical objections that can be raised against the thesis that a declining domestic currency promotes inward FDI, all other things constant. For a discussion of these objections, see Safarian and Hejazi (2001). 21. For a review of the literature on this issue, see Globerman (2002). 22. The standard deviation for the third period (13.35) exceeds that for the second (11.21) or the first (4.80). 23. The fitted equation for the January 1945 to May 2002 period was E t = −0.16 + 1.002 × E t−1 . Where E is the Canada-U.S. exchange rate. If the coefficient on the lagged exchange rate is constrained to equal one a random walk with drift model is obtained, and the measure of exchange rate uncertainty is essentially unchanged. 24. Trefler found that since 1996 there has been a 2.1% compounded annual increase in labor productivity for industries that were the most affected by tariff cuts versus 0.6% for the entire manufacturing sector. 25. This suggestion is promoted by Grubel (1999) among others. 26. See, for example, Engel and Rogers (1996), Flam and Jansson (2000) and Frankel and Rose (2002).
REFERENCES Acharya, R. C., Prakash, S., & Someshwar, R. (2001). Canada’s trade and foreign direct investment patterns with the United States. Ottawa: Industry Canada, mimeo. Beaulieu, E. (2000). The Canada-U.S. free trade agreement and labour market adjustment in Canada. Canadian Journal of Economics, 33(2), 540–563. Brunelle, D., & Deblock, C. (2001). Economic regionalism under the FTAA and MERCOSUR: James Monroe or Simon Bolivar? Montreal: University of Quebec at Montreal, mimeo. Buckley, P. J., Clegg, J., Forsans, N., & Reilly, K. (2000). Assessing NAFTA’s impact on the strategies of multinational firms in Canada. Leeds: University of Leeds, mimeo. Centre for the Study of Living Standards (2002). Raising Canadian living standards: A framework for analysis. Paper prepared for the TD forum on Canada’s living standards, Ottawa, Ontario, October 7–8, mimeo. Council of Economic Advisers (2002). Economic report of the president. Washington, DC: U.S. Government Printing Office. Crucini, M., Telmer, C., & Zachariadis, M. (2001). Understanding European real exchange rates. Working Paper, June 2001. DeVoretz, D., & Coulombe, D. (2002). Labor mobility between Canada and the United States: Quo Vadis 2002. Burnaby, BC: Simon Fraser University, mimeo.
44
STEVEN GLOBERMAN AND PAUL STORER
Dunning, J. H. (2000). The Impact of the completion of the European internal market on FDI. In: J. H. Dunning (Ed.), Regions, Globalization and the Knowledge-Based Economy (pp. 131–169). Oxford: Oxford University Press. Eden, L., & Monteils, A. (2000). Regional integration: NAFTA and the reconfiguration of North American industry. In: J. H. Dunning (Ed.), Regions, Globalization and the Knowledge-Based Economy (pp. 170–220). Oxford: Oxford University Press. Egelton, R., & Wojciech, S. (1999, March 24). Trends in Canada-U.S. Migration: Where’s the flood? Economic Analysis. Engel, C., & Rogers, J. (1996). How wide is the border? American Economic Review, 86(5), 112–125. Engel, C., & Rogers, J. (1998). Regional patterns in the law of one price: The roles of geographies vs. currencies. In: J. Frankel (Ed.), The Regionalization of the World Economy (pp. 153–183). Chicago: University of Chicago Press. Evans, T. (2002). Foreign direct investment monitor. Ottawa: Export Development Canada, mimeo. Flam, H., & Jansson, P. (2000). EMU Effects on international trade and investment. Stockholm University: Institute for International Economic Studies. Seminar Paper No. 683. Frankel, J., & Rose, A. (2001). An estimate of the effect of common currencies on trade and income. The Quarterly Journal of Economics (May), 437–466. Gaston, N., & Trefler, D. (1997). The labour market consequences of the Canada-U.S. free trade agreement. Canadian Journal of Economics, 30(1), 18–41. Globerman, S. (1999). Trade liberalisation and the migration of skilled workers. Research publications program, Paper No. 3. Ottawa: Industry Canada. Globerman, S. (2000). Trade liberalization and the migration of skilled professionals and managers: The North American experience. World Economy, 23(7), 901–922. Globerman, S. (2002). Trade, FDI and regional economic integration: Cases of North America and Europe. Paper presented at conference entitled Enhancing Invest Cooperation in Northeast Asia, Honolulu, Hawaii, August 7–9, mimeo. Globerman, S., & Shapiro, D. (1999). The impact of government policies on foreign direct investment: The Canadian experience. Journal of International Business Studies, 30(3), 513–532. Government of Canada, Department of foreign affairs and international trade (2001). Second Annual Report on Canada’s State of Trade, http://www.dfait-maeci.gc.ca/eet/trade/sot 2001/section04en.asp. Grubel, H. (1999). The case for the amero: The economics and politics of a North American monetary union. Vancouver: Fraser Institute. Harris, R. G. (2000). The case for North American monetary union. ISUMA, 1(1), 1–6. Hart, M. (2002). A trading nation: Canadian trade policy from colonialism to globalization. Vancouver: UBC Press. Helliwell, J. F. (1998). How much do national borders matter? Washington, DC: Brookings Institution Press. Helliwell, J. F. (2001). Canada – Life beyond the looking glass. Journal of Economic Perspectives, 15(1), 107–124. Hine, R. C. (1994). International economic integration. In: D. Greenaway & L. A. Winters (Eds), Surveys in International Trade (pp. 234–272). Oxford: Blackwell. Hufbauer, G. C. (2001, March 30). North American convergence: An American perspective. remarks at the annual policy conference of the Canadian association of business economists. Washington, DC. Johnson, D. (1990). Co-integration, error correction, and purchasing power parity between Canada and the United States. Canadian Journal of Economics, 23(4), 839–855.
Canada-U.S. Economic Integration Following NAFTA
45
Lafrance, R., & Schembri, L. (2002). Purchasing power parity: Definition, measurement, and interpretation. Bank of Canada Review (Autumn), 27–33. Lee, J. (2002). Canada-Mexico relations. Report prepared for the house of commons standing committee on foreign affairs and international trade. Ottawa: Parliamentary Research Branch, mimeo. McCallum, J. (1995). National borders matter: Regional trade patterns in North America. American Economic Review, 85(3), 615–623. Rugman, A. (2000). The end of globalization. London: Random House Business Books. Trefler, D. (2001). The long and short of the Canada-U.S. free trade agreement. NBER Working Paper 8293. Yan, B. (2002). Purchasing power parity: A Canada/U.S. exploration. Statistics Canada micro economic analysis division research paper 11F0027MIE No. 2.
REGIONAL VS. GLOBAL FINANCING STRATEGIES FOR U.S. MNEs Joseph P. Daniels, Walid Hejazi and Marc von der Ruhr ABSTRACT Despite declining in 2001, foreign direct investment (FDI) surged during the 1990s. As a result, current levels of FDI flows are triple their 1990 levels. It is well documented in the literature that FDI occurs in large part among countries that are geographically close. It is also well established that the NAFTA had a significant impact on both U.S. FDI flows and hence FDI stocks. In addition, tax policies and tax treaties have been shown to be important drivers of U.S. FDI. The analysis presented in this chapter confirms these earlier results. We extend the analysis, however, to show that tax treaties have a significant impact on financing patterns of U.S. MNE activities abroad. Based on these results, we argue that bilateral tax treaties should be an important part of trade agreements between the United States and Latin American partners in anticipation of a Free Trade Agreement of the Americas (FTAA).
1. INTRODUCTION Since the 1994–1995 peso crisis, a great deal of attention has been given to financial globalization and the destabilizing consequences of unbridled short-term capital North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 47–68 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10003-X
47
48
JOSEPH P. DANIELS ET AL.
flows. In spite of this attention, the process of financial globalization does not depend on the actions of short-term investors. Rather, as argued by Rugman (2000), the correct focus of financial globalization is on the activities of multinational enterprises (MNEs) and the long-term capital flows generated through their foreign direct investment (FDI) activities. Rugman further argues that because the majority of MNE activity is regionally-based, true financial globalization has yet to occur. A recent study by the OECD (2003, p. 5) echoed this view, arguing that a significant share of FDI takes place among countries bound by regional trade agreements and among geographically close countries. The defeat of the Multilateral Agreement on Investment (MAI), the failed Seattle trade round, the sluggish progress of the Doha trade round, and chilled diplomatic relations among the world’s economic superpowers signals that if further economic integration is to occur at all, it will result from the actions of MNEs, and therefore proceed on a regional basis rather than a multilateral basis. In other words, further economic integration depends on management decisions on where to engage in FDI and how to finance these FDI flows. Hence, prudent policy making requires a recognition of the regional nature of long-term capital flows and an understanding of their financing patterns. As a result, countries can avoid distortionary policy actions so as to ensure an efficient allocation of productive resources. The purpose of this chapter is to review the patterns of U.S. outward FDI and to highlight differences regionally vs. globally. Though these patterns have been discussed elsewhere, little attention has been given to the financing patterns of U.S. MNEs operating abroad, and again making comparisons regionally vs. globally. Because further economic integration is likely to be to the south, we consider potential differences between triad FDI activity and North-South FDI activity. We find a stark difference in corporate tax treatment among the triad nations as opposed to north-south relationships. Specifically, there is a preponderance of long-standing bilateral tax treaties among triad nations and a lack of bilateral tax treaties between the United States and its potential Free Trade Agreement of the Americas (FTAA) partners to the south. Our contribution is twofold: First, though a number of studies conclude that tax polices affect FDI and that FDI is becoming more sensitive to its tax treatment, we follow the advice of Hines (1997) and test the robustness of these conclusions by controlling for important macroeconomic factors. Second, we extend the analysis of Blonigen and Davies (2000) on the impact of tax treaties on FDI in two important ways. First, we extend the period of analysis to cover the entire 1990s, a period in which FDI flows rose dramatically before settling to a level three times that prior to the 1990s (OECD, 2003), and by considering the impact of bilateral treaties on MNE financing patterns. We find that total FDI flows are significantly affected by both the 1986 U.S. tax reform and by the existence of a bilateral tax treaty
Regional vs. Global Financing Strategies for U.S. MNEs
49
between the United States and the host nation. However, when we decompose total FDI flows into their three forms – those financed by new equity, retained earnings, and intercompany debt – we see that both of these tax variables have different effects on equity financing vs. retained earnings financing. Unfortunately, our model performs poorly in explaining intercompany debt flows. We conclude that tax policies have important consequences for FDI. Bilateral tax treaties, in particular, present an interesting regional dimension because U.S. bilateral treaties exist primarily with NAFTA and European nations.1 Attempts to create an FTAA need to include tax treaties in the similar manner that the NAFTA did. The remainder of the chapter is organized as follows. Section 2 briefly describes the existing microeconomic theory on FDI. Section 3 describes general trends in U.S. outward FDI stock while Section 4 describes U.S. outward FDI flows and how these flows are financed. Section 5 provides empirical evidence on tax treaties and other key determinants of these patterns. Section 6 offers conclusions.
2. THEORETICAL BACKGROUND In this section, we outline the literature that identifies reasons for the rise of MNEs. In the interest of brevity, we identify only the major themes of the literature, and refer the reader to Hejazi and von der Ruhr (2002) for a summary of specific articles. The development of microeconomic theories to explain FDI are linked by their focus on firms’ efforts to reduce both explicit and implicit transactions costs. Early models focus on the explicit costs that may be reduced by internalizing a transaction via FDI as opposed to engaging in an external, market-based transaction. More recent models build upon earlier work by considering the effect of implicit costs, which may arise as a consequence of incomplete contracting. The additional costs arising from incomplete contracting are important factors that influence how a firm organizes its transactions. For example, the difficulty of generating incentive-compatible contracts with foreign sales agents as a means to serve a foreign market will have a significant influence on the FDI decision. These models often hinge on the role of information and uncertainty about quality. Uncertainty about quality implies that the firm must reveal its competitive advantage to receive its full value in the foreign market. This may lead an agent who deals with the firm at arm’s length to behave opportunistically, or for the firm to sacrifice too much of its rents to a local agent. Dunning (1988, 1993a) summarizes the different motives for FDI in a single cohesive theory via his ownership, location, internalization framework (OLI).
50
JOSEPH P. DANIELS ET AL.
According to this framework, for a firm to enter a foreign market via FDI rather than other potential options (such as direct exporting or licensing) the firm must have three unique advantages: (1) an ownership advantage; (2) an internalization advantage; and (3) a location advantage. Ownership advantage refers to the firm’s unique assets which confer upon it market power. The internalization advantage refers to the firm’s inability to realize the full value of its ownership advantage through market transactions, thus causing the firm to deliver its product internally through FDI. Finally, the location advantage refers to the firm locating a production process abroad to benefit from that country’s comparative advantage, or to locate close to the consumer in order to realize the value of its ownership advantage. Though the majority of research on FDI has focused on manufacturing FDI, the OLI framework is particularly well suited to explaining the increasing role of service industries in FDI. Services have unique characteristics that deserve attention in examining their contributions to FDI flows. Depending on the immediacy of production and consumption of the service, exporting may not be feasible. Second, services are used increasingly to coordinate fragmented production processes, and require a local presence where production is located. Third, services are often knowledge-intensive, and therefore replicated at a low cost. Finally, a key determinant of a service provider’s long-run success involves its ability to establish a reputation for consistently providing a high-quality service, and is best done through FDI. Evidence suggests that FDI is the leading method by which service providers access a foreign market. For instance, we may consider the ratio of affiliate sales to affiliate sales plus export sales to illustrate how important FDI sales are to services (Dunning, 1993b). For accounting services this ratio is 92%, and for advertising services it is 85%. The rationale for engaging in FDI as established in the literature helps explain the industrial and geographic distribution of U.S. outward FDI seen in the data. Services are the leading industry of both U.S. outward and inward FDI, and typically originate from or are located in other developed countries. This may be attributed to the demand for services in developed economies as well as the fact that as production processes become increasingly fragmented, service firms are needed to coordinate various production processes. The last issue to consider relates to what method a firm uses to finance its FDI flows: new equity, retained earnings, or intercompany debt. Little work has been done in this area. A notable exception is Lipsey (1993) who compares financing flows across U.S. inward and outward FDI from 1950 through 1991. He finds that U.S. firms tend to finance FDI abroad more consistently through retained earnings whereas foreign firms have a more varied pattern of financing.
Regional vs. Global Financing Strategies for U.S. MNEs
51
3. GENERAL TRENDS IN U.S. OUTWARD FDI STOCK Before proceeding, something should be said about the construction of the FDI stock data. U.S. FDI stock data are reported on a historical cost basis. Retained earnings and new flows are in current dollars and are simply added to the previous year’s FDI stock (which are not in current dollars). This forms the balance of payments definition of FDI. The market value of FDI stock at any point in time, however, is defined as the level of FDI in the previous period, plus retained earnings, plus net new flows of FDI, plus price appreciation (or less depreciation). It is this last component which is needed to convert FDI from historical costs to market values. Since these adjustments are unavailable, most FDI stock data are at historical costs as is the data presented here.2 Examination of the data clearly indicates that U.S. firms’ FDI decisions on where to locate have changed over time. Figure 1 illustrates the distribution of the stock of U.S. FDI in five host regions: Europe, Canada, Latin America, Japan, and Australia and New Zealand. The figure illustrates that Europe remains the dominant and increasingly important destination for U.S. outward FDI, accounting for about 52% of U.S. outward FDI in 2000. On the other hand, Canada has received less U.S. FDI over time. Canada reached a high of 22% of U.S. outward FDI in 1984, and a low of 10% by 2000.3 An interesting fact to draw from this figure refutes a popular belief about developed-country FDI, namely that MNEs are driven to enter foreign markets in order to exploit cheap labor. For world stocks of FDI in 2000, 67% of all inward FDI stocks were located in developed countries. Furthermore, 88% of these stocks originate in developed countries (UNCTAD, 2001). For the United States specifically, approximately three quarters of its outward FDI stock was located in other developed countries and over 90% of its inward stock originates in other developed countries. Therefore, the popular perception that MNEs, American and otherwise, engage in FDI primarily to exploit low-wage labor is clearly mistaken: market access and access to technology are far more important determinants of FDI. It may nevertheless be true that to the extent MNEs undertake FDI in developing countries it is for reasons of factor-price differences. However, this is by no means the prime motivation for MNE activity abroad. Figure 2 illustrates the changing composition of U.S. outward FDI among the following industries: petroleum, manufacturing, and producer services (the sum of wholesale trade, banking, finance, insurance, real estate, and other business services) sectors.4 This figure reveals an interesting trend regarding the increasing importance of services FDI. The importance of petroleum FDI has fallen dramatically whereas manufacturing has fallen by less. The chart illustrates
52 JOSEPH P. DANIELS ET AL.
Fig. 1. Distribution of FDI Stock.
Regional vs. Global Financing Strategies for U.S. MNEs
53
Fig. 2. Distribution of FDI Stock Across Industries.
that services have significantly overtaken manufacturing in its share of U.S. outward FDI. As noted above, these trends can be attributed to several broad factors: restrictions on FDI in services have been liberalized relatively recently, there has been dramatic growth in services in the economies of the world, and the changing nature of international production. These trends are likely to continue given the increased importance of services in the GDP of the U.S. and other developed countries. The importance of the third factor, namely the changing nature of international production – that is, the increasing fragmentation of production processes – should be highlighted. These sub-production processes require business services to coordinate them. Hence, we see an increasing role for services in FDI. It is also logical that the increase in services FDI followed manufacturing FDI. Manufacturers led the FDI process and as their production is refined, they call on services to be provided in the foreign market (Raff & von der Ruhr, 2001). The magnitudes of these stocks of services FDI also reflect on the importance of services in international trade. The United States is the world leader in the export of services. It dominated the share of total service exports in 1998, accounting for 18.1% of total service exports; with the next highest share, 7.7%, coming from Britain (Economist, 1999). Many of these services are not easily traded for a number of reasons. For instance, they may need to be produced and consumed at
54
JOSEPH P. DANIELS ET AL.
the same time, be subject to high trade barriers, or simply require frequent contact between the service provider and the consumer. Thus, the majority of trade in these services is actually carried out through FDI rather than exports. In fact, in many industries, foreign markets cannot be serviced without a local presence, and hence FDI. Additional understanding of FDI and trade in these service industries is important for policy makers, as the liberalization of trade in services has received considerable attention in recent trade negotiations. Our interest in regional FDI prompts consideration of the industrial breakdown of the stock of U.S. FDI in both Canada and Mexico. Though not presented here, the FDI data suggests that some significant changes took place around 1994 in both countries. All industries experienced significant increases in the stock of FDI in each country between 1994 and 2000. Further, in each country, the top three industries experiencing growth were banking, FIRE, and other business services FDI. Perhaps because of market potential for growth, Mexico’s growth was larger in each category than Canada’s. Interestingly, the growth in regional manufacturing FDI has lagged behind the growth of manufacturing FDI globally. This may be explained by the NAFTA; as more goods are traded freely, the need for manufacturing FDI to jump tariff and non-tariff trade barriers is decreased. Further, the growth in services FDI has grown more regionally than globally, reflecting the increased amount of trade that must be done “in person” thereby requiring a local presence.
4. U.S. FDI FLOWS AND FINANCING To this point, we have focused on U.S. FDI stocks abroad. We now turn our focus to U.S. FDI outflows and how these flows are financed. We break U.S. outward FDI flows into their components: equity capital, retained earnings, and intercompany debt. The Bureau of Economic Analysis defines each type of flow as follows. Equity capital outflows occur when a U.S. parent company increases its equity investment in one of its existing foreign affiliates or makes a new equity investment in a foreign business enterprise, either by acquiring an existing business or establishing a new one. Equity capital inflows occur when a U.S. parent company reduces its equity investment in one of its existing foreign affiliates. Intercompany debt flows are of two types: U.S. parent receivables and U.S. parent payables. U.S. parent receivable represent loans that a U.S. parent extends to its foreign affiliate. An outflow on U.S. parent receivables occurs when the parent extends a new loan to its affiliate; an inflow occurs when an affiliate repays part or all of a loan from its U.S. parent. U.S. parent payables represent loans that a foreign affiliate extends to its U.S. parent. An outflow on U.S. parent payables
Regional vs. Global Financing Strategies for U.S. MNEs
55
occurs when the parent repays part or all of its loan from its foreign affiliate. An inflow occurs when an affiliate extends a new loan to the U.S. parent. Retained earnings are the parent’s claim on undistributed after-tax earnings of its foreign affiliate. They are computed as the difference between the parent’s claim on its affiliate’s current earnings and the dividends that the affiliate pays to a parent in a given period. They are positive when a parent has a claim on positive current earnings of its affiliate in excess of the dividends that it receives from its affiliate. To better understand the general patterns of these flows, we accumulate the data and present it for the entire period, 1982 through 2000, in Table 1. The top panel of each table provides the dollar value of the three forms of financing, and the bottom panel provides the share of financing by country or region considered. As expected, there is a great amount of heterogeneity across countries and regions. There are, however, some generalizations. First, retained earnings are almost always more important than either new equity financing or intercompany debt. (The exceptions to this are petroleum FDI in Mexico and Central America, and services investment in Mexico, Australia, and South America.) The bottom three rows of Table 1 give the total, regional, and global distributions of U.S. outward FDI, where regional FDI is U.S. FDI located in Canada and Mexico, whereas global FDI is U.S. FDI to the rest of the world. These aggregates are plotted in Fig. 3. The following patterns clearly emerge. First, retained earnings are the most important source of financing for U.S. MNEs operating abroad, and this is true at both the aggregate and industry levels. Furthermore, retained earnings are more important regionally than globally, and this is true at the aggregate and industry levels. Although equity finance is less important as a source of finance for U.S. MNEs operating abroad than is retained earnings, equity finance is more important than intercompany debt. Furthermore, equity finance is more important globally than regionally for total outward FDI as well as in manufacturing, whereas it is more important regionally than globally in services. Of course, the interesting question that arises is, what one is to make of these differences. This is the question to which we now turn. Much attention has been paid to the corporate decision as to whether to finance capital internally or externally. It is generally the case that internal financing sources are less costly than are external ones, and hence it is advised that corporations first use internal financing sources. In addition, firms that do not wish to disclose information about their operations to outsiders are also less likely to go to outside sources for funds. Finally, issues relating to asymmetric information often place limits on a company’s ability to raise capital in the external capital market. For example, when a company issues new shares, it can be read as a signal that management believes that shares are overvalued, and hence may result in a reduction in share prices.
56
JOSEPH P. DANIELS ET AL.
Table 1. Pattern of Accumulated FDI.
Millions, U.S. dollars Canada Mexico Europe Japan Asia and Pacific Australia New Zealand Other Asia Latin America South America Central America Total outward Regional Global Percentage distribution Canada Mexico Europe Japan Asia and Pacific Australia New Zealand Other Asia Latin America South America Central America Total outward Regional Global
Total Flows
Equity Flows
Retained Earnings
Inter Company Debt
98,159 34,068 570,190 30,565 158,580 30,326 3,055 150,553 198,018 75,738 50,062
25,894 13,330 221,008 3,605 45,244 15,220 857 43,448 70,133 29,096 15,826
64,669 21,560 275,878 20,957 102,255 15,044 1,928 93,766 113,435 38,762 34,202
7,599 −820 73,298 6,006 15,001 −51 1,220 13,339 14,450 7,880 30
1,051,817 132,227 919,590
367,795 39,224 328,571
570,236 86,229 484,007
113,788 6,779 107,009
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
26.4 39.1 38.8 11.8 28.5 50.2 28.1 28.9 35.4 38.4 31.6
65.9 63.3 48.4 68.6 64.5 49.6 63.1 62.3 57.3 51.2 68.3
7.7 −2.4 12.9 19.6 9.5 –0.2 39.9 8.9 7.3 10.4 0.1
100.0 100.0 100.0
35.0 29.7 35.7
54.2 65.2 52.6
10.8 5.1 11.6
We hypothesize that the patterns of financing we have seen bilaterally, by region and globally, should depend on host country characteristics such as the volume of real economic activity, openness to trade, openness to direct investment, distance, and adjacency. In addition, it has been shown in the literature that FDI activity is sensitive to changes in tax policies and that this sensitivity has risen over time.5 For example, the Tax Reform Act of 1986 attempted to address perceived incentives for MNEs to invest abroad as opposed to domestically as created in the former
Regional vs. Global Financing Strategies for U.S. MNEs
57
Fig. 3. Total, Regional, and Global Distribution of FDI.
tax code. Specifically, changes in cross crediting, or foreign tax averaging, and changes in tax definitions were designed to balance U.S. tax rules in this regard. The reform act also resulted in different foreign tax credits for different types of activity, such as financial services, creating potential differences in FDI activity across sectors (U.S. Congress, 1987). Though the Tax Reform Act of 1986 affected income earned in all other nations the same, bilateral tax treaties created the possibility for differential tax treatment. Blonigen and Davies (2000) argue that bilateral tax treaties result in four outcomes; standardization of tax definitions, promotion of the exchange of tax information thereby improving the ability of policy makers to enforce tax laws, prevention of tax arbitrage or “shopping” for the lowest tax treatment, and affecting overall taxation of MNEs by reducing the coincidence of double taxation and taxes on remittances. They show that over time bilateral tax treaties affect U.S. inward and outward FDI activity. Table 2 provides the countries in
58
JOSEPH P. DANIELS ET AL.
Table 2. U.S. Bilateral Tax Agreements. Country Argentina Australia Austria Belgium Brazil Canada Chile Columbia Costa Rica Denmark Ecuador Finland France Germany Greece Guatemala Honduras Hong Kong Ireland
Date of Tax Treaty NA 1953 (1982) 1956 (1996) 1948 (1970) NA 1941 (1980) NA NA NA 1948 NA 1940 (1994) 1940 (1994) 1954 (1989) 1950 NA NA NA 1952 (1997)
Country
Date of Tax Treaty
Israel Italy Japan Luxembourg Mexico Netherlands New Zealand Norway Panama Peru Portugal South Africa Spain Sweden Switzerland Turkey United Kingdom Venezuela
1975 (1993) 1955 (1984) 1954 (1971) 1963 1994 1949 (1992) 1948 (1982) 1951 (1971) NA NA 1994 1953–1987 (1997) 1991 1940 (1994) 1951 (1996) 1997 1945 (1975) 2000
Note: Date of Last Revision in Parentheses. Source: Blonigen and Davies (2000), U.S. Internal Revenue Service, www.irs.gov/pub/irs-trty.
our sample and dates of their bilateral tax treaties with the United States, if one exists. We consider both the effect of bilateral tax treaties and the changes in foreign taxation brought about by the Tax Reform Act of 1986. We contribute to the findings of Blonigen and Davies-who consider tax treaties in the context of a gravity model-by including more recent data and considering the impact of bilateral tax treaties on methods of financing.6 We also contribute to the literature on the Tax Reform Act of 1986 by considering the impact of the act in the context of the main empirical framework of FDI analysis, a gravity model, and by allowing for vintage effects of the tax act. To this point we have considered the theory of FDI location and the actual patterns of U.S. outward FDI and patterns of financing. We have also acknowledged the potential impact tax polices can have on FDI flows as well as methods of financing. Together with a reconsideration of the data presented in Table 2 leads us to the following conclusions: U.S. outward FDI is concentrated among nations with whom the United States has longstanding bilateral tax treaties, primarily Western European nations. The exceptions are as follows: Canada, who has the second oldest tax treaty with the United States; Mexico, whose tax treaty coincided with
Regional vs. Global Financing Strategies for U.S. MNEs
59
the NAFTA; and the 2000 tax treaty with Venezuela. As argued by Hines (1997), increased magnitudes of FDI mean that costs associated with inefficient tax laws become much greater. If indeed tax policies affect FDI patterns and financing strategies, then greater attention must be paid to tax policies to ensure that the potential benefits of an FTAA are achieved. Given the on-going negotiations with respect to a FTAA and the evidence of the importance of tax policies to the location decisions of MNEs, it is surprising that there has been little or no attempt to harmonize corporate tax treatment between the U.S. and Latin America generally. We argue here that such treaties will serve to stimulate further long-term capital flows and hence contribute to further economic integration between the United States and Latin America. Furthermore, such treaties also impact the methods of financing used by MNEs as they expand their operations abroad.
5. EMPIRICAL RESULTS To examine this issue we extend the standard gravity model to include tax policies and treaties. We examine the impact the 1986 U.S. tax reform as well as the introduction (or existence) of bilateral tax treaties have had on both the patterns of FDI and how these FDI flows are financed. The model estimated includes host county real GDP, the exchange rate (measured as foreign currency per U.S. dollar), host country openness to trade, host country openness to FDI, an adjacency dummy, a NAFTA dummy, distance measures, and dummy variables to capture the Tax Reform Act of 1986 and the existence of a bilateral tax treaty.7 The Penn world tables provide data on host real GDP, the bilateral exchange rate, and host country openness to trade. We calculate host openness to FDI as the ratio of host stock of inward FDI to host GDP for each year. All of these factors are expected to have a positive influence on U.S. outward FDI. Distance between the host country and the United States is gathered from the Bali Online Corporation, and may be argued to have either a positive or negative effect on FDI, though the gravity model suggest a negative influence. The adjacency dummy is expected to have a positive influence on attracting U.S. FDI. The NAFTA dummy is also expected to have a positive influence to reflect increased regional economic integration. To capture the influence of the 1986 U.S. tax reform and the effect of bilateral tax treaties, we allow for each policy to have an immediate effect on U.S. FDI patterns. Furthermore, following Blonigen and Davies (2000), we acknowledge that the impact of these treaties may change over time, and possibly in a non-linear fashion. We therefore include a variable reflecting the age of each tax treaty, in years, as well as the square of their ages.
60
JOSEPH P. DANIELS ET AL.
We further investigate the effects of the 1986 U.S. tax reform and bilateral tax treaties on U.S. FDI by examining the effect each has on various measures of FDI. That is, we estimate the model first with U.S. outward FDI stocks as the dependent variable, and then we employ U.S. outward FDI flows be the dependent variable. These flows are then decomposed in their components: new equity flows, retained earnings flows, and intercompany debt flows. That is, we consider the effects of our control variables on both the stocks and flows of U.S. FDI abroad. This strategy allows us to interpret the estimated coefficients from each specification to generate additional insights. The coefficient estimates on the independent variables, when the dependent variable is U.S. FDI stock abroad, represent incremental impacts of the independent variables on changes in stocks. Thus, the coefficient estimates in these stock regressions can be interpreted as the impact each control variable has on U.S. outward flows. When we regress FDI flows on the independent variables, the coefficient estimates reflect changes in flows, thereby showing rates of change of FDI. The estimation results for various specifications are given in Tables 3 through 5. Table 3 provides the results with FDI stock serving as the dependent variable. Table 4 considers the results with FDI flows as the dependent variable. These estimates allow us to test whether, when we include dummy variables to account for the presence of a bilateral tax treaty and the 1986 U.S. tax reform, there is any statistically significant effect on total FDI stocks, flows, or the means by which these flows are financed (new equity, retained earnings, or intercompany debt). Finally, Table 5 reconsiders FDI flows, allowing the tax treaties and the U.S. tax reform to have a non-linear effect over time. Following Blonigen and Davies (2000), we omit the intercept dummy variables associated with the bilateral tax treaties and the U.S. tax reform and replace each with its age and the square of its age. The coefficient of the age variable may be interpreted to show whether the impact of the policy change on the FDI flow increased or decreased over time. Further, the coefficient of the square of the age variable may be interpreted to show whether the increasing or decreasing impact of the policy change occurred at a rate that increased or decreased over time. Our empirical results for the standard gravity model are generally in conformity with the existing literature. As a result, we focus here only on the analysis as it relates to the U.S. tax code change and tax treaty variables. Table 3 shows the results for the FDI stock regressions. The first column shows that the tax reform of 1986 as well as the presence of bilateral tax treaties have a positive and statistically significant impact on U.S. FDI stocks abroad. The regression results in the second column exclude the tax-policy dummy variables and instead include the age and squared age of each of the tax-policy variables. The results indicate that the square of the age of the tax treaty is positive and significant
Regional vs. Global Financing Strategies for U.S. MNEs
61
Table 3. Estimation Results for FDI Stocks. Variable
Intercept Exchange rate RGDP Trade openness FDI openness Distance Adjacency NAFTA Treaty U.S. 1986 tax code
Dependent Variable: FDI Stock Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
−20663.23** (9755.31) −0.019745 (0.0135) 0.0000485*** (0.000003) 297.4142*** (25.3679) 20345.45*** (4645.303) −1.952348** (0.8429) 10596.02 (7345.539) −854.5614 (5800.074) 8500.283*** (2336.437) 2260.303** (1198.009)
−24782.17** (10100.12) 0.012156 (0.0129) 0.0000381*** (0.000003) 246.9644*** (26.4958) 16985.63*** (4292.188) −1.422986* (0.7925) 13299.46** (6871.225) 4564.763 (5301.922)
−32401.85*** (11084.85) 0.014741 (0.0129) 0.0000395*** (0.000003) 270.2192*** (25.9367) 14993.37*** (4279.27) −1.243606 (0.8113) 18053.69*** (6755.616) 5448.69 (5318 238)
602.995*** (52.7951)
702 0.8897 0.8880 1.8619 506.1747 0.0000
Treaty age
Rho
0.944136*** (0.0138)
79.9778 (179.6918) 7.200424*** (2.389) −783.2232 (630.5547) 57.44306 (44.3477) 0.9522*** (0.013)
Number of observations R-squared Adjusted R-squared Durbin-Watson stat F-statistic Prob (F-statistic)
702 0.8688 0.8669 1.8897 457.6761 0.0000
702 0.8911 0.8892 1.8609 469.9210 0.0000
(Treaty age)2 U.S. 1986 tax reform age (U.S. 1986 tax reform age)2
∗p
≤ 10%. ≤ 5%. ∗∗∗ p ≤ 1%.
∗∗ p
−606.2756 (630.1629) 53.35066 (44.5087) 0.961043*** (0.0116)
62
JOSEPH P. DANIELS ET AL.
Table 4. Estimation Results for FDI Flows with Tax Variable Intercept Dummy Variables. Variable
Dependent Variable Total FDI Flows
Intercept Exchange rate RGDP Trade openness FDI openness Distance Adjacency NAFTA Treaty U.S. 1986 tax reform Rho Number of observations R-squared Adjusted R-squared Durbin-Watson stat F-statistic Prob (F-statistic)
Equity Flows
Retained Earning Flows
Intercompany Debt Flows
Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
−207.8401 (874.2349) 0.000222 (0.0034) 0.00000274*** (0.000000458) 15.67699*** (4.6228) 3257.319*** (1014.357) −0.260427*** (0.097) 376.0301 (1261.609) 2917.158** (1315.193) 882.3984** (440.2783) 715.7634*** (283.7865) 0.708854*** (0.0278)
29.55263 (557.5244) 0.001021 (0.0016) 0.00000144*** (0.000000267) 7.663654*** (2.5342) 884.1798* (523.6234) −0.152427*** (0.0611) 338.6784 (718.3889) −432.967 (671.9994) 455.4984* (243.7028) 190.845 (140.6796) 0.794994*** (0.0239)
−302.0731 (350.3701) −0.000467 (0.0016) 0.00000119*** (0.000000192) 10.535*** (2.0133) 1735.163*** (445.3041) −0.108939*** (0.039) 505.0062 (525.35) 2044.055*** (583.5947) 352.9329* (185.8513) 348.8743*** (128.3686) 0.655154*** (0.031)
37.66901 (152.2639) −0.000389 (0.0013) 0.000000285*** (0.000000084) −0.123284 (0.9607) 483.842* (253.4623) −0.030756* (0.0162) −194.8092 (239.5595) −22.33261 (360.8061) 128.6305 (89.1313) 153.6402* (93.1505)
702 0.5875 0.5815 2.1906 98.4217 0.0000
702 0.6401 0.6348 1.8790 122.8718 0.0000
702 0.553718 0.54726 2.176348 85.73488 0
703 0.0366 0.0241 2.0448 2.9221 0.0021
∗p
≤ 10%. ≤ 5%. ∗∗∗ p ≤ 1%.
∗∗ p
while the age variable is statistically insignificant. We interpret the combination of a significant squared age variable with an insignificant age variable to suggest that at tax treaty has a (near) linear, rather than quadratic, effect over time. To test this, we estimate the regression using only the treaty age variable. The third column
Variable
Dependent Variable Total FDI Flows
Intercept Exchange rate RGDP Trade openness FDI openness Distance Adjacency NAFTA Treaty age (Treaty age)2 U.S. 1986 tax reform age
Equity Flows
Retained Earning Flows
Intercompany Debt Flows
Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
Coefficient (Std. Err.)
300.4218 (779.1982) 0.001366 (0.0034) 0.00000186*** (0.000000417) 8.013055* (4.3312) 2818.76*** (947.0936) −0.180247** (0.0847) −84.97626 (1140.372) 2921.612** (1237.374) −108.1034*** (32.5678) 2.245301*** (0.4949) −86.65722 (127.2042)
54.63233 (548.2537) 0.002073 (0.0017) 0.00000106*** (0.000000271) 4.940807* (2.6214) 786.3846 (514.9333) −0.133573** (0.0587) 239.7794 (704.1054) −235.0481 (657.338) −12.99999 (19.0432) 0.489803* (0.2765) 20.28378 (73.851)
−81.91138 (578.4873) 0.002199 (0.0017) 0.00000111*** (0.000000283) 5.712946** (2.6932) 746.721 (520.1674) −0.153198*** (0.0616) 438.7118 (717.8738) −278.1693 (662.0345) 20.56483*** (5.195)
3.765414 (322.3828) −0.000121 (0.0015) 0.000000803*** (0.000000176) 7.103034*** (1.8724) 1494.018*** (415.844) −0.077159** (0.0351) 379.7751 (483.5247) 1938.673*** (548.6477) −45.46203*** (14.1167) 0.963627*** (0.217) −102.0224* (55.1129)
76.83281 (148.2962) 9.98E-05 (0.0013) 0.000000279*** (0.000000084) 0.200699 (0.9745) 427.1399* (252.2901) −0.022747 (0.0161) −253.771 (241.2128) −16.66406 (362.5942) −20.04512** (8.3694) 0.370436*** (0.1325) 17.1975 (29.6821)
36.5026 (74.5655)
Regional vs. Global Financing Strategies for U.S. MNEs
Table 5. Estimation Results for FDI Flows with Vintage Effects.
63
64
Table 5. (Continued ) Variable
Dependent Variable Total FDI Flows
Equity Flows
Retained Earning Flows
Intercompany Debt Flows
Coefficient (Std. Err.) Coefficient (Std. Err.) Coefficient (Std. Err.) Coefficient (Std. Err.) Coefficient (Std. Err.) (U.S. 1986 tax reform age)2 Rho
∗p
≤ 10%. ≤ 5%. ∗∗∗ p ≤ 1%.
∗∗ p
−0.622872 (5.235) 0.783606*** (0.0252)
−0.993508 (5.3238) 0.800925*** (0.0243)
9.861172** (3.9462) 0.627993*** (0.0314)
−1.360113 (2.2516)
702 0.6170 0.6104 2.1908 92.5059 0.0000
702 0.6494 0.6432 1.8950 106.3274 0.0000
702 0.6479 0.6423 1.9049 115.4327 0.0000
702 0.5929 0.5858 2.1633 83.6254 0.0000
703 0.0501 0.0349 2.0733 3.3103 0.0002
JOSEPH P. DANIELS ET AL.
Number of observations R-squared Adjusted R-squared Durbin-Watson stat F-statistic Prob (F-statistic)
10.29135 (9.0952) 0.667406*** (0.0291)
Regional vs. Global Financing Strategies for U.S. MNEs
65
provides the results with the treaty age included without including the square of the age variable. The significance of the treaty age variable suggests that the effect of the treaty is linear over time. Table 4 measures the impact of tax reform and bilateral tax treaties on FDI flows, both aggregate and disaggregated into their components. The results indicate that both the presence of tax treaties and the tax reform of 1986 had a positive and statistically significant impact on U.S. outward FDI flows. Breaking FDI flows down into forms of financing, the tax reform act is significant for retained earnings flows, but not for equity flows (we briefly discuss intercompany debt below). The tax treaty variable is positive and significant for both equity flows and retained earnings flows. Thus, we see that the policies have different impacts on FDI flows by mode of financing. Table 5 extends the analysis in Table 4 to allow for an examination of the effects of tax policies over time. It also shows that for retained earnings the impact of the tax reform act decreases over time and, based on the quadratic term, at a diminishing rate. This is the only mode of financing that was statistically significantly affected by the tax reform act. The impact of a bilateral tax treaty on total FDI flows and retained earnings flows decreases at a diminishing rate over time. The age of a tax treaty is not significant in the regression on equity financing, while the square of the age is, indicating that the effect is closer to a linear effect than a quadratic effect. This is confirmed when we estimate the same regression with only the linear term, which is highly significant. Thus, we see that equity flows tend to be positively affected over time by the existence of a bilateral tax treaty. This stands in contrast to the diminishing impact the treaties have on retained earnings flows over time. The unusual results for intercompany debt may not be surprising when one considers the overall fit of the model to explaining intercompany debt. While the goodness for fit is reasonable for the other specifications, it is very low for the intercompany debt regressions. The results suggest that a model other than the gravity model is needed to explain intercompany debt flows of FDI. Consequently, we make no inferences from the results of the regressions for intercompany debt. In general, we find that there are important regional aspects to FDI. This is reflected first by the NAFTA variable. Its significance, however, varies by type of FDI flow suggesting that regionalization will impact the form of financing used to engage in FDI. Second, an implication for the negotiation of regional trade agreements is reflected by the impact of the tax treaties on FDI, as the preponderance of U.S. bilateral tax treaties are with European nations, rather than Latin American nations. Given the evidence that tax policies affect FDI flows and financing, policy makers may want to increase the attention paid to tax policies in order to maximize the potential benefits of regional trade agreements such as the
66
JOSEPH P. DANIELS ET AL.
FTAA. We suggest that such policies will help to foster further long-term capital flows and contribute to further economic integration between the United States and Latin America.
6. CONCLUSION The recent trade pact signed between the United States and Chile signals continuing and perhaps renewed interest in creating a Free Trade Agreement of the Americas. Such an agreement is likely to strengthen regional integration vis-`a-vis foreign direct investment. The results presented here are consistent with previous literature in that host country characteristics are important determinants of FDI. In addition, we show that tax policies – both aggregate tax changes and bilateral treaties – affect FDI. The latter has an important regional aspect because the 2000 tax treaty with Venezuela is the only existing bilateral treaty between the United States and a Latin American nation. We conclude that bilateral tax treaties with Latin American partners should be an important component of trade agreements. We suggest that additional research continue to focus on regional aspects of FDI. In addition, future research should also focus on explaining the differential effects we document of tax policies on alternative forms of FDI financing.
NOTES 1. The second oldest U.S. bilateral treaty is with Canada. The bilateral tax treaty with Mexico, however, coincided with the NAFTA. 2. The U.S. Department of Commerce (1995) has published U.S. stock figures on the basis of historical cost, replacement cost, and market values, but the country and sectoral data are available only on a historical (book value) basis. There are a variety of private and semi-official estimates of the different valuations for the U.S. and U.K. stocks of FDI (Bellak & Cantwell, 1996). A straightforward way to adjust stock values is through changes in security prices, as utilized in Gray and Rugman (1994) but this is subject to a number of criticisms as noted in Bellak and Cantwell. We use the unadjusted data. 3. The reduction in Canada’s share of U.S. FDI has been attributed in part to the Canada-U.S. FTA and the NAFTA: the importance of a local presence by U.S. firms in Canada has been significantly reduced as border restrictions have fallen (Rugman, 1990; Safarian & Hejazi, 2001). 4. The role of other business services (the majority of which are business services such as accounting, management consulting, marketing, etc.) has overtaken banking as an outward source of service FDI. 5. See Hines (1997) for an extensive summary of the literature on taxation and FDI. See Altshuler, Grubert and Newlon (2001) for evidence on the rising sensitivity of FDI to taxation.
Regional vs. Global Financing Strategies for U.S. MNEs
67
6. Blonigen and Davies examine the period 1966 through 1992. Our data period is more recent, 1982 through 2000, and covers a period when FDI activity increased dramatically. 7. The adjacency dummy is always unity for Mexico and Canada and zero for all others. The NAFTA dummy is zero for all countries until 1994 and thereafter when it is unity for Mexico and Canada. Both dummies are included so as to discern between the two effects.
ACKNOWLEDGMENTS Special thanks for assistance and advice from James McGibany and participants at the Canada-United States Business Conference held at Indiana University and the Kelley School of Business Friday, April 11 – Saturday, April 12, 2003.
REFERENCES Altshuler, R., Grubert, H., & Newlon, T. S. (2001). Has U.S. investment abroad become more sensitive to tax rates? In: J. R. Hines (Ed.), International Taxation and Multinational Activity (pp. 9–38). NBER Conference Report Series, Chicago: University of Chicago Press. Bellak, C., & Cantwell, J. (1996, April). Foreign direct investment – How much is it worth? A comment. Transnational Corporations, 5(1), 85–97. Blonigen, B. A., & Davies, R. B. (2000, October). The effects of bilateral tax treaties on U.S. FDI activity. Working Paper 7929, National Bureau of Economic Research. Dunning, J. H. (1993a). Multinational enterprises and the global economy. Wokingham, England: Addison-Wesley. Dunning, J. H. (1993b). Multinational enterprises and the growth of services: Some conceptual and theoretical issues. In: K. P. Sauvant & P. Mullampally (Eds), Transnational Corporations in Service (pp. 33–74). New York: Routledge. Dunning, J. H. (1988). The eclectic paradigm of international production: A restatement and some possible extensions. Journal of International Business Studies, 19(1), 1–30. The Economist (1999). Financial indicators. Economist Magazine (May 8), 109. Gray, S. J., & Rugman, A. (1994). Does the United States have a deficit with Japan in foreign direct investment? Transnational Corporations, 3(2), 127–137. Hejazi, W., & von der Ruhr, M. (2002). U.S. firms in world finance. In: American Macromanagement. Northampton, MA: Edward Elgar. Hines, J. R. Jr. (1997). Tax policy and the activities of multinational corporations. In: A. J. Auerbach (Ed.), Fiscal Policy: Lessons from Economic Research (pp. 401–446). Cambridge, MA: MIT Press. Lipsey, R. E. (1993). Foreign direct investment in the United States: Changes over three decades. In: K. A. Froot (Ed.), Foreign Direct Investment (pp. 113–172). Chicago. Organization for Economic Cooperation and Development (2003). Economic Outlook, 73. Paris: OECD. Raff, H., & von der Ruhr, M. (2001). Foreign direct investment in producer services: Theory and empirical evidence. CESifo Working Paper No. 598, Center for Economic Studies and Ifo Institute for Economic Research, Munich, Germany.
68
JOSEPH P. DANIELS ET AL.
Rugman, A. M. (1990). Multinational and Canada-United States free trade. Columbia, SC: University of South Carolina Press. Rugman, A. M. (2000). From globalisation to regionalism: The foreign direct investment dimension of international finance. In: K. Kaiser, J. J. Kirton & J. P. Daniels (Eds), Shaping and New International Financial System: Challenges of Governance in a Globalising World (pp. 203–219). Aldershot: Ashgate. Safarian, A. E., & Hejazi, W. (2001). Canada and foreign direct investment: A study of determinants. University of Toronto Centre for Public Management, Toronto: University of Toronto Press. United Nations Conference on Trade and Development (UNCTAD) (2001). World investment report: Promoting linkages. New York: United Nations Publications. U.S. Congress, Joint Committee on Taxation (1987, May 7). General explanation of the Tax Reform Act of 1986. Chicago: Commerce Clearing House. US Department of Commerce (1995). Foreign direct investment in the United States: Detail for historical-cost position and related capital and income flows. Survey of Current Business (August), 53–78.
REGIONAL INTEGRATION OF U.S. BORDER STATES WITH CANADA: EVIDENCE FROM U.S. STATE EXPORTS Larry Davidson ABSTRACT This paper examines the nature of regional trade integration between the United States and Canada by using a Similarity Index that summarizes the behavior of exports of states along the U.S./Canadian border relative to U.S. states that are not on the Canadian border. An export Similarity Index is used to show the considerable importance of industry mix relative to distance. Similarity Index changes suggest that increased export sales between the U.S. and Canada between 1996 and 2001 were not primarily driven by proximity factors that underlie a regional phenomenon. Industry factors independent of location and distance were important contributors to changes in U.S. exports to Canada. The upshot is that global, not regional, factors may underlie increasing trade between the U.S. and Canada. That is, an apparent global phenomenon may have been mistaken for a regional one.
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 69–84 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10004-1
69
70
LARRY DAVIDSON
INTRODUCTION Alan Rugman challenged conventional wisdom with data and analysis that shows that when the history books are closed on the last half of the 20th century, the chapter heading will say “The Age of Regionalization” not “The Age of Globalization.” This distinction clearly matters for researchers and policy makers, whether they are practicing in business, government, or academe. My paper dives down deep into the muck of disaggregated sub-national data to add evidence to this issue. I do not find compelling evidence of regionalization with respect to the United States and Canada for the 1996–2001 time period. My results are compatible with Wall (2003).1 The next section is a review of Rugman’s work and his conclusion that favors regionalization over globalization. I then explain why disaggregated data – data for 50 states and 97 industries – can be used to shed more light on this issue and how a Similarity Index can be used to add evidence to the regional/global question. Similarity Index changes are presented and compared for U.S. exports to Canada and to the world. The Similarity Index values measure increasing similarity or dissimilarity of the industry mix of exports from two sets of states – the ones that border Canada and the ones that don’t. The final section contains my summary. While the overall Similarity Index numbers support the idea that distance or proximity matters for trade, a closer look at the details suggests there is some room for doubt. Thus, my analysis finds no compelling evidence for regionalization and suggests that existing and historical links among the industries of various countries was more important than distance in determining changes in trade between Canada and the U.S. between 1996 and 2001.
DISTANCE, REGIONALIZATION, AND GLOBALIZATION Alan Rugman in a series of papers argues that international trade has not become more global – it has become more regional. Rugman and Verbeke (2003a) look at foreign sales as a percentage of total sales for the world’s 500 largest multinational enterprises (MNEs) and is able to classify only 10 of them as truly global. Examining 100 of the world’s largest (MNEs), Rugman and Verbeke (2003b) conclude,2 “Therefore, in the overall set of 20 highly internationalized MNEs, the case of global strategy and structure can be made for only six firms, with the additional observation that even these firms exhibit regional elements.” Rugman and Brain (2002) found additional evidence with respect to the regionalization/globalization issue. They found that while intra-regional trade within NAFTA countries increased significantly, foreign direct investment
Regional Integration of U.S. Border States with Canada
71
(FDI) flows did not. FDI was more apt to develop between the EU and NAFTA. The investment result, they explain, is the result of many decades of free trade in the automotive manufacturing sector in North America that precluded the need for cross border investments. They point out that desires for increased trade in services has resulted in larger Canadian investments in U.S. services industries. Rugman is essentially saying that the emphasis on globalization has been misplaced, leading to misunderstanding. Business schools mislead students if they teach global strategy when business decisions largely are influenced by regional trends and changes. Policy makers mislead the public about the impacts of global policies if the more urgent needs for policy exist within multi-country triads. Policy makers create unrealistic hopes for global trade outcomes if the needs for such policies are exaggerated. Tracking trade shares and examining the behavior of the 100 or 500 largest companies by triad offers useful but insufficient evidence about the regionalization/globalization debate. Results depend very much on the definition of optimal distance. Rugman implicitly assumes that a triad is the right distance measure. But triads are very large both in terms of distance and culture (Uruguay and Quebec are in the same triad). My approach uses a much shorter measure of distance that derives from the location difference of the 15 states on the southern border of Canada relative to the other states.3 My paper focuses on the principal discriminating factor in the regional/global debate – distance or proximity. Regionalization makes sense because distance – whether measured by kilometers or culture – is a “barrier” to trade. It makes sense, therefore, that business find it easier and more profitable to engage in business closer to home. Doing global business – in all triads – would be a second best solution for many companies that seek the benefits of international trade. My work tests the importance of distance by putting it on equal footing with industry mix explanations of international trade. Underlying the idea of globalization are pre-existing industry linkages. While a company might want to trade closer to home, it might also want to take advantage of country, company, or industry relationships that have formed over time. The question we pose is that for 1996 to 2001, for changes in U.S. export sales to Canada, how important was the role of proximity relative to industry mix? If proximity is shown to be less important, then we consider this as weakening the case for regionalization.
SIMILARITY INDEX I use an industrial mix Similarity Index4 to identify cases where industry differences played strong roles in state export performance from 1996 to 2001.5
72
LARRY DAVIDSON
The range of this index is from zero, indicating complete dissimilarity, to 100, indicating the state’s sectoral distribution of exports is identical to the national distribution. I use industry export Similarity Indexes that compare export sales of the 15 border states to the remaining 35 non-border states. If, according to a Similarity Index, border state exports to a given foreign destination became less similar to non-border states exports over a given time period, that result would imply that industry mix was important to relative export performances of border states. If this turned out to be true for exports to Canada, then this would be evidence supporting the regionalization hypothesis. If this evidence of export industry dissimilarity turned out to be truer for Canada than for other export destinations, this would make the regionalization case even stronger. Because of distance and location, one would expect export industry dissimilarity between border and non-border states to be much more important for Canada than for other export destinations. Of course, if pre-existing industrial structure of export sales was more important than distance, then one might expect no change in industry-based similarity over time. To make this point more vivid, consider the following matrix: State/Destination
Border Not border
Canada
Rest of World (R)
Industry 1
Industry 2
Industry 1
Industry 2
0.7 0.6
0.3 0.4
0.5 0.5
0.5 0.5
Each cell contains export sales share data for each industry in a border or non-border state. For simplicity, in this example we assume only two industries comprise a state’s exports. Assume also that industry 1 in the border state specializes in Canada while industries 1 and 2 are roughly equal with respect to exports to the rest of the world. Consider the cell for border state exports to Canada which says that 70% of the border state’s X1 exports go to Canada. The remaining 30% of border state’s exports to Canada are from Industry X2. Now consider three types of changes that might affect these export sales. (1) Regional shock when distance is very important; (2) Regional shock when distance is not important; (3) Global shocks. Border states have a larger share of industry 1 exports going to Canada than do the non-border states. If distance is very important, one would expect that a regional
Regional Integration of U.S. Border States with Canada
73
shock that promotes more integration would favor border state industry1 more than non-border state industry 1.6 In that case the industry 1 border to non-border share gap should increase and the Similarity Index for Canada would decrease. But this would have no effect on the Similarity Index value for the rest of the world. If distance was not important, in contrast, there is no reason to expect the Similarity Index to decrease for either location. Global shocks could have two different kinds of impacts. Consider a global shock that favors border state industry 1’s exports to the world. For example, suppose that border state industry 1 has been negatively impacted by trade regulations that are now removed in countries where border states have heavily exported. This might decrease the Similarity Indexes for both Canada and/or the U.S. Thus, evidence that the Similarity Index for Canada decreased would not be sufficient evidence of regionalization. A second kind of global shock might negatively impact industry 2 in the border state. In that case a global shock would decrease the Similarity Index as global factors tended to disproportionately impact the minority share industry in the border vs. the non-border states. The upshot of these examples is that rising dissimilarity of industrial export distributions of border to non-border states could result from: Regional factors that promote increased trade and distance matters; Global industry or country factors that increase trade and favor high share industries; Global industry or country factors that decrease trade and penalize low share industries.
EXPORTS SALES AND EXPORT SHARES OF BORDER AND NON-BORDER STATES, 1996–2001 U.S. exports to the world were $622.8 billion in 1996 and subsequently increased to $731 billion in 2001 after peaking at $780.4 billion in the year before.7 Table 1 shows the situation of the border states in 1996. These 15 states sold $188 billion to the world, or about 30% of all U.S. exports. These states, however, accounted for almost exactly 50% of U.S. exports to Canada. In contrast the $7.6 billion sold to Mexico was only 13% of U.S. exports to that destination in 1996. With respect to individual states, there was considerable diversity in sales by export destination. New York was the largest exporter to the world from among the border states. Michigan, however, was the largest exporter to both Canada and Mexico. Michigan, Illinois, New York, and Ohio accounted for about two-thirds of the border state’s exports to Mexico and Canada. Washington, a major exporter
74
LARRY DAVIDSON
Table 1. Export Sales of Border States to the World, Canada, and Mexico, 1996 (in Billions of Dollars).
U.S. Border states New York Michigan Washington Illinois Ohio Massachusetts Indiana Wisconsin Minnesota Vermont Idaho New Hampshire Maine North Dakota Montana
Mexico
Canada
World
56.761 7.552 0.867 3.085 0.234 1.182 0.709 0.326 0.325 0.361 0.230 0.022 0.044 0.053 0.017 0.058 0.040
132.584 66.371 9.655 16.613 2.612 7.621 10.930 3.532 5.442 3.128 2.588 2.361 0.274 0.520 0.527 0.380 0.187
622.827 188.001 34.230 27.553 26.482 24.176 22.676 14.524 10.984 9.504 8.992 3.302 1.571 1.481 1.380 0.707 0.440
to the world, in comparison, was a relatively small part of the border state’s exports to either Mexico or Canada. Table 2 shows export growth between 1996 and 2001 from the U.S., the 15 border states, and the remaining non-border states to the World, Canada, and Mexico. U.S. exports to the world grew by 17% over that 6-year time period. Exports to Mexico and Canada grew faster, at 79 and 23%, respectively. Exports of border states to the world grew by 22.5%, faster than exports of the non-border states of 15.1%. The 15 states along the Canadian border, therefore, led the nation in international sales to the world. The same result holds true for exports to Mexico. The magnitude of the growth rate differences, however, are much larger for exports to Mexico. States with especially strong growth rates to Mexico were Indiana, Washington, Ohio, and New York. With respect to Canada, however, the story changes. The non-border states exports to Canada grew by 35%, almost three times the increase of the 13% increase for the border states. Michigan and New York, two of the larger exporters to Canada in 1996, had very weak growth in export sales to Canada during that time period. These facts alone warrant a closer look at the regionalization phenomenon since, at surface, they seem to contradict the importance of distance. The growth rate experience is evident in the shares of exports. Table 3 shows the shares of world exports as they were in 1996 and the changes in shares
Regional Integration of U.S. Border States with Canada
75
Table 2. Export Sales of Border and Non-Border States to the World, Canada, and Mexico, Percentage Change from 1996 to 2001 (in Percent).
U.S. Non-border states Border states New Hampshire Idaho Washington Maine Indiana Illinois New York Massachusetts Ohio Michigan Minnesota North Dakota Montana Wisconsin Vermont
Mexico
Canada
World
79 75 106 55 43 268 96 444 91 114 75 197 55 89 −34 −15 86 −17
23 35 13 14 29 5 60 14 50 1 −19 27 6 2 4 62 21 −41
17 15 23 62 35 32 31 31 26 23 20 19 17 17 14 11 10 −14
between 1996 and 2001. We see that 35% of the exports of the states that border Canada went to Canada in 1996. Mexico was a minor destination, receiving only about 4% of exports from these northern states. The non-border states were more apt to sell their exports to Asia (35%) or Europe (23%). These states were more likely than border states to sell to Mexico but less likely to sell to Canada than their northern counterparts. Some of these shares changed markedly between 1996 and 2001. The largest changes in shares for the border states were a 3% increase in share to Mexico and 3% decreases in shares to Canada and Japan. They also increased share by 2% to Europe. The non-border states, in contrast, had an increase share of goods going to Canada. The share going to Mexico rose by 6%. These results provide conflicting evidence for the regionalization hypothesis. First, U.S. neighbor Mexico clearly has become more important to both groups of states. Its share increases dominated all other world region share increases. For the non-border states, Mexico and Canada together saw a share increase of 9%, much larger than any of the other areas. But for the border states, export share to Canada fell by about the same amount as exports to Japan. The share of exports to Mexico rose, but not much faster than increases to the EU and to Australia/Oceana. For the country as a whole, therefore, the results are mixed and we look further at
76
LARRY DAVIDSON
Table 3. Export Sales Share of Border and Non-Border State Exports to World Destinations, 1996 and Change from 1996 to 2001. Destination
Canada EU15 Other Asia Japan Southeast Asia Mexico Other Western Hemisphere Rest of Europe Africa Australia/Oceania
Border States 96 Share (+ or − Change)
Non-Border States 96 Share (+ or − Change)
35 (−3) 21 (+2) 13 (–) 10 (−3) 5 (–) 4 (+3) 4 (–) 4 (–) 1 (+1) 2 (+2)
15 (+3) 20 (+1) 16 (−2) 11 (−3) 8 (−2) 11 (+6) 10 (–) 3 (–) 2 (–) 2 (–)
Note: In percent, shares do not add to 100 because of rounding errors.
industry performances to examine these apparent contradictions. In particular, a declining share of exports from border states to Canada is not sufficient reason to reject regionalization. The question revolves around the industrial composition of the decline – was it because key industries in the border states had a common experience with the same industries in non-border states (no increase in dissimilarity)? Or was it because border state industries with strong ties to Canada showed declines relative to the same industries in the non-border states (increase in dissimilarity). Thus, the Similarity Index will give us information that goes beyond the usual aggregated information about share changes.
SIMILARITY INDEX VALUES Introduction A Similarity Index is used to measure the similarity of industry exports of the border to the non-border states for their exports to the world in 1996 and for changes between 1996 and 2001. A Similarity Index sums the minimum share values for each industry, where the minimum refers to the lesser of the shares for the border and non-border states.8 Similarity Index Values by Export Destination Table 4 shows the Similarity Index values for the border states in 1996. The value for exports to the world in 1996 was 77.9, indicating that the industrial composition
Regional Integration of U.S. Border States with Canada
77
Table 4. Similarity Index Values by Export Destination in 1996 and Percentage Change from 1996 to 2001 (Index Values and Percent). Index Value
% Change
EU15 Canada World Other W. Hemis
79.2 79.1 77.9 77.6
5.4 −4.6 −2.3 −2.6
Other Asia Australia Japan
76.5 74.4 73.3
−12.2 −0.6 6.1
Africa
70.1
−26.2
Southeast Asia Mexico Other Europe
69.9 66.5 61.9
−13.2 −2.6 −4.3
of border state and non-border states exports to the world were quite similar in that year.9 Somewhat higher values existed in 1996 for exports to EU15 and Canada, respectively 79.2 and 79.1.The border states were least like the non-border states with respect to exports sales in 1996 to Other Europe and Mexico. The destinations roughly fell into three groups with respect to the Similarity Index: Highly similar: EU15, Canada, Other Western Hemisphere, Other Asia; Somewhat similar: Australia, Japan, Africa, Southeast Asia; Not very similar: Mexico, and Other Europe. Table 4 also shows that similarity declined between 1996 and 2001 for exports to the world and for exports to eight of the 10 regions. This means that for all destinations, except for EU15 and Japan, the structure of industry exports of the border states became less like the structure for the non-border states.10 The border states were becoming more distinctive, less like the non-border states for all destinations besides EU15 and Japan. The Similarity Index decreases ranged from about −0.6% for Australia to −26.2% for Africa. The value for Canada fell by nearly 5% about twice the decline for exports to the world.11 The single fact that the Similarity Index fell twice as much for Canada than for the world is evidence in favor of regionalization and the importance of distance. If distance is very important, then one would expect that the states closest to Canada would develop closer and stronger trade relationships with Canada. But several other facts suggest that the Canada Similarity Index change was not very special and question the importance of distance:
78
LARRY DAVIDSON
Three country destinations (Africa, Southeast Asia, and Other Asia) had very large negative Similarity Index changes indicating that the border states had significant share changes relative to the non-border states. Other Europe and Other Western Hemisphere had negative changes similar in size to Canada’s. The border states are considerably farther from Mexico than Canada, yet the change in Similarity Index was similar for exports to these two countries. Our examples of similarity change above suggest some answers to these questions. We listed three factors that might cause rising dissimilarity. Any of these would explain a decline in the Similarity Index (border states export share becoming more dissimilar with non-border states): Regional factors that promote increased trade and distance matters. Global industry or country factors that increase trade and favor high share industries. Global industry or country factors that decrease trade and penalize low share industries. Trade did increase over the time period, but the share of the border states exports to Canada decreased. Therefore, it looks like global rather than regional factors might have been influential towards explaining the decline in the Similarity Index. This requires a deeper look into the sources of changes in the Similarity Index. Similarity Index Changes by Industry The similarity changes reflect changes in the industry shares of exports to the world of each state. We can see why various border state exports become more or less the same as the non-border states by examining the role of the 98 industry sectors that comprise each state’s export sales. It would seem straight-forward to examine the contribution to the Similarity Index for each industry by state. However, because of the construction of the index and the interplay of the many industries, the changes in the minimum share values do not always convey this information. Instead, I calculated for each industry in each state for its exports, the value which I will call I-SIM,12 where I-SIM = Absolute value of the border state share gap in 2001 − Absolute value of the border state share gap in 1996 The border state share gap is defined as the difference between the industry share in the border states and the non-border states. I-SIM is always greater (less) than zero when the industry share’s are becoming more similar (dissimilar).
Regional Integration of U.S. Border States with Canada
79
The industries with the greatest I-SIM values for exports to the world and exports to Canada were: 27 (Mineral Fuel/Oil), 84 (Machinery), 85 (Electrical Machinery), 87 (Vehicles) and 88 (Aircraft/Spacecraft). These industries dominated the share changes between 1996 and 2001. Industries 84, 85, and 87 are the main export industries of the United States. These three industries in 1996 accounted for over 57% of the border state’s exports to the world and over 49% of non-border states exports. Whereas I-SIM values were used to choose these important industries, we discuss below the actual share changes by industry and export destination to interpret why they contributed to changes in similarity. Four cases are analyzed. First, the most intuitive case supporting regionalization is when the share gap for exports to Canada in 1996 was positive and then became more positive between 1996 and 2001. This result held for the following cases, suggesting that border states improved their position relative to non-border states with respect to exports to Canada: 84: the border state share rose more than the non-border state share 87: the border state share fell, but somewhat less than the non-border state share Second, this result is somewhat diminished by the fact that border states generally improved their share positions relative to non-border states with respect to exports to other locations in which the border states had no “distance” advantage: 88: border state share to the world increased while share of non-border states fell; 84, 87: border state exports to Mexico increased much more than non-border state exports to Mexico. Third, an intuitive case favoring regionalization exists when the share gap was negative in 1996 and then shrank (moved closer to zero from a negative value) between 1996 and 2001. In this case the border states would be converging in share on the non-border states from below. There were no cases of this happening with respect to Canada. But there were several cases of this for other destinations that go against this intuition since they imply that the states that border Canada were doing relatively better in Mexico and the EU than the non-border states: 85, 88: border states exports shares rose more than non-border states export shares to Mexico; 27: border states export share fell but by less than non-border states export shares to the EU15. Fourth, if the share gap in 1996 was positive and it declined between 1996 and 2001, this would indicate that the border state advantage to Canada was eroded by
80
LARRY DAVIDSON
the more distant non-border states. This would be evidence against regionalization. This was true for industry 27, wherein border states export share fell to Canada while non-border states share was rising. This result was also true for other destinations where the share gap was eroded between 1996 and 2001: industry 84 (to the world), industry 85 (to the EU15), and industry 87 (to the world and the EU15). If the share gap in 1996 was negative and became more negative this would indicate that the border states were becoming even less competitive with respect to export shares over time. One would not expect this result for exports to Canada if regionalization were strong within an industry. But this was true for industry 85 which saw the border share falling to Canada significantly more than the non-border share. For industry 88, the border state share rose, but the non-border state share increased even more. In these cases, the states became more dissimilar, but this was not evidence consistent with regionalization. This loss of relative share was not just true for exports to Canada. It was also true for industries 85 to the world and industry 27 to the world and to Mexico.
SUMMARY Did export sales to Canada of the border states differ from the non-border states? If yes, then this implies that distance and proximity were important. The Similarity Index decreases in value whenever the absolute value of the share gap between border and non-border industries increased. Another way of saying this is that the shares of border and non-border industries diverged. If the close proximity of the border states to Canada was meaningful, then the special relationship between these states and Canada should have been enhanced and revealed. The Similarity Index value between 1996 and 2001 for Canada fell by 5%. This implies increasing divergence – thus the border states were becoming less like the non-border states with respect to Canada. The Similarity Index fell for 10 of the 15 border states indicating that the increasing divergence was shared by most of the border states. The strongest divergence was exhibited by Idaho, Maine, Illinois, and Wisconsin. This contrasted with Similarity Index changes for exports to the world wherein only four of the 15 border states showed increasing divergence. This implies that proximity to Canada was important for the time period from 1996 to 2001. Thus, regionalization seems to be at work in North America. But there is other evidence that questions this conclusion. First, while border state exports were increasing over the 1996–2001 period, exports to Canada (13%) were growing slower than exports to the World (23%). As a result, the share of border state exports to Canada decreased by 3%.
Regional Integration of U.S. Border States with Canada
81
Meanwhile the shares of border state exports were rising to Mexico, EU15, Australia/Oceania, and Africa. The non-border states exports had the opposite experience with exports to the world growing at 15% and exports to Canada growing by 35%. The share of non-border state exports to both Mexico and Canada rose significantly – though the Mexico share rose twice as fast as the Canadian one. In short, while border states were becoming less like the non-border states, it was not because of increased intensity of trade with Canada. It was partly because border states were increasing share in other parts of the world while non-border states were doing relatively better in Canada. In a few cases, the increased relative share of border states exports was to Mexico. One might consider that as evidence of regionalization. But it is not in the sense defined in this article. Border states are further from Mexico than the non-border states. This suggests that something other than distance was at play in determining these changes.13 In effect, this industry information supported the idea that global shocks were impacting key industries and they were having similar effects on border state exports to various regions of the world. Second, the details of the Similarity Index declines were revealed by the I-SIM values for the contribution of each industry. The main contributors were industries 27 (Mineral Fuel/Oil), 84 (Machinery), 85 (Electrical Machinery), 87 (Vehicles) and 88 (Aircraft/Spacecraft). Consider their contributions. Analyzing 20 possible cases of industry share changes (for these five industries and four export destinations) between 1996 and 2001 finds only two strong indications of growing regionalization: industries 84 and 87 exports to Canada. The remaining cases did not support growing dissimilarity in the sense that border state industry exports to Canada rose relative to non-border state exports. In most cases there was evidence that suggested that differential export performance that had nothing to do with distance. In conclusion, my analysis of Similarity Indexes questions the impact of distance or proximity on trade between the U.S. and Canada between 1996 and 2001. During this time period, the relationship between the border states and Canada did become more special – industry export shares of the closer border states diverged from those of the more distant non-border states. While this special distinctiveness of the closer border states might indicate a special role for proximity, a closer inspection suggests that the special relationship is not what would be normally expected from regional integration. The share of border state exports to Canada declined over this time period. Furthermore, it appears that existing (in 1996) industry strengths or linkages had more to do with trade than proximity. The distinctiveness of the border states arose primarily because the border states were increasing trade with non-Canadian destinations while the more distant non-border states were focusing more on Canada.
82
LARRY DAVIDSON
Generalizing this result from the Canada/U.S. experience to the full North American Triad or other Triads is not warranted. Nevertheless, this paper suggests that digging deeper into industry mix and specific sub-triad country or region details is worth the effort. Further work along these lines could expand our understanding of the regional-global experiences of trade. We might corroborate Alan Rugman’s conclusion that few multinationals are global in the sense that they do not operate in all three triads. But we might also find that Rugman simply has not given the process long enough to unfold. If apparent regionalization comes more from industry mix and less from proximity, then the potential for rapid globalization may be stronger than Rugman admits.
NOTES 1. Howard J. Wall (2003) examined the impacts of NAFTA on exports for the time period from 1993 to 1997. He found that NAFTA did not uniformly increase trade within the pact’s members and explained why economies of scale and trade creation and diversion can and did lead to increased trade between pact and non-pact members. The trade experiences of various regions of Canada were quite distinctive. For example, whereas Western Canadian exports to Mexico rose by over 90%, those from Eastern Canada increased by less than 1%. 2. Rugman and Verbeke (2003b, p. 8). 3. Alaska is not included with the 15 border states. It is not part of a contiguous region of border states and its economy is quite different from the others. Alaska largely exports oil, fishing, and other mineral products. 4. Coughlin and Pollard (2001, p. 29). 5. We choose 1996–2001 for several reasons. First, the data we use with the rich industry detail is published with state detail only back to 1996. We wanted the data to be as long-term as possible, but no data exists for the pre-1996 time period, especially because of the timing of NAFTA. Second, data for full-year 2002 were not available when the research was done for this paper. Third, because there was very rapid export growth that peaked in 1999/2000, ending earlier than 2001 would have yielded a sample period with little cyclical variability. Containing 2000 and 2001 allows for a time period with both very strong increases and decreases in export sales. While some might point out that our beginning and ending dates are not equivalent time periods insofar as export strength is concerned, there is nevertheless, an expected upward trend over the time period as witnessed by the positive growth rates over that time period. 6. Lorraine Eden commented on an earlier draft of this paper that much of the impact of the Canadian-U.S. Free Trade Agreement would have been felt before 1996. NAFTA would have been a more dominant shock between 1996 and 2001. Thus, one might not have expected significant increases in trade integration between Canada and the border states. In this version I compare Similarity Index changes with respect to Canada and Mexico and find the surprising additional result that the Mexico-bound exports of states along the Canadian border were more affected than states closer to Mexico. This underscores the point of the
Regional Integration of U.S. Border States with Canada
83
article, that distance is not the paramount factor in determining export flows. This paper is not meant to be an event-shock study. Much was happening – beyond NAFTA and the unfolding impacts of the Canadian-U.S. Free Trade Agreement – in the 1996–2001 time period that might have impacted trade with the border and non-border state export destinations. This work simply compares the role of distance and industry mix over that time period. 7. The data used throughout this report is Harmonized System of Tariffs or Schedule B information supplied by the U.S. Census Bureau and distributed by Global Trade Information Services, Inc. For more information about this and other similar state export sales data, see Davidson (2003). 8. These sums are simple sums of minimum share values and are not weighted. This might give the impression that unimportant industries had as much influence on the totals as important industries (those with large shares). But the fact is that the sums and the changes in sums were dominated by the largest industries. This makes sense because the minimum share for most of the industries were very small numbers. Many were less than 1%. The share values and changes in share values for a few large industries were quite large in comparison – several in double digits. Inasmuch, the more important industries dominated the results. 9. The similarity values for the individual states ranged from 36 to 76, indicating that while some states were quite like the non-border states, some were not. Most similar to the non-border states were Illinois, New York, and Minnesota. Least like the non-border states were Vermont, Washington, North Dakota, and Montana. 10. This was primarily because of industry changes in Washington, Illinois, Ohio and Wisconsin. In contrast, the industrial structures of 11 border states became more like that of the non-border states, though to differing degrees. Having the largest changes toward similarity were Vermont, Massachusetts, Maine, Montana, and New York. 11. These conclusions about the degree of similarity or dissimilarity stem from intuitive judgments. I know of no standard statistical tests to determine statistically significant deviations. 12. I thank my colleague Mike Baye for helping me think through this issue and for suggesting I-SIM. 13. My colleague, Chuck Tryczinka, explains that the distance within the United States may not be significant and that the border and non-border states are “equally” close to Canada and Mexico. This suggests more inquiry into how to define the smallest unit of significant distance.
ACKNOWLEDGMENTS The author expresses his appreciation to the Canadian government and to the Center for International Business and Education for funding for this project. Valued comments from Alan Rugman, Charles Tryczinka, and Lorriane Eden led to significant improvements. The author also thanks Global Trade Information Services for their help in packaging the data in a highly usable form and to his graduate assistant, Diego Agudelo, for his database management expertise.
84
LARRY DAVIDSON
REFERENCES Coughlin, C. C., & Pollard, P. S. (2001). Comparing manufacturing export growth across states: What accounts for the differences? St. Louis Federal Reserve Bank Review (January/February), 25–40. Davidson, L. S. (2003). Subnational international business policy in an age of regional trade integration. In: A. M. Rugman (Ed.), Indiana as a Case Study. Leadership in International Education and Research, monograph (forthcoming). Research in Global Strategic Management (Vol. 9). Oxford: Elsevier (June). Rugman, A. M., & Brain, C. (2002). Intra-regional trade and foreign direct investment in North America, mimeo, Indiana University. Rugman, A. M., & Verbeke, A. (2003a, August). Regional multinationals and triad strategy. In: A. M. Rugman (Ed.), Research in Global Strategic Management (Vol. 8) (forthcoming). Rugman, A. M., & Verbeke, A. (2003b). Regional and global strategies of multinational enterprises. Journal of International Business Studies (forthcoming). Wall, H. J. (2003). NAFTA and the geography of North American Trade. St. Louis Federal Reserve Bank Review (March/April), 13–26.
PROTECTING CANADA AND THE U.S. AGAINST TERRORISM: A COMMON SECURITY PERIMETER? Georges A. Tanguay and Marie-Christine Therrien ABSTRACT We argue that national security is a public good and its production can be analyzed in a strategic context. We first present the context of the border between Canada and the United States. Next, we discuss the options of status quo and adoption of a common security perimeter relative to sovereignty and security. We show that efficient border policies could require cooperation among countries but motivating such collaboration may be difficult since joint border security policies may involve a prisoners’ dilemma problem. On the other hand, we show that the likelihood of joint increased security will be higher if there are country-specific benefits for a country improving security at its border. If this is the case, we demonstrate it is possible to reach optimal security using independent border policies.
1. INTRODUCTION Border security has often been an issue in many countries. There has been little concern, however, over the security of the Canada-U.S. border in modern times. This changed on September 11th 2001. The immediate response has been to increase the amount of resources devoted to the security of the Canada-U.S. North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 85–96 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10005-3
85
86
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
border (as well as other borders and ports). It is unclear, however, if this is the best policy response. In fact, it may be the case that the best policy response might be to make this border freer (i.e. less secure) and to focus on greater joint-security coordination between the United States and Canada. A possibility would be the adoption of a common security perimeter. Indeed, the concept of a common security perimeter has resurfaced, stronger than ever, in the news and in diplomatic spheres. Given that, we should assess the costs and benefits associated with potential options to changes in the border. In this paper, we present and study two potential options, discussing some of the possible gains in terms of security. The options are: (1) Maintain the status quo by continuing a coordination of Canadian and U.S. policies of different matters such as security, immigration, border management. (2) Adopt a common perimeter and eliminate the border. In Section 2, we present the context of the border between Canada and the United States. Next, in Section 3, we discuss the two options in terms of sovereignty, territory and security. In Section 4, we analyze both options and show that under plausible conditions, conserving independent border policies may be the best option.
2. CONTEXT OF THE BORDER Understanding the security issues related to a border has to be coupled with an understanding of the relationship between two neighboring countries. Therefore, in this section we present two important factors: trade and border policy management. These two factors bring the specificity of the Canada-U.S. border relationship into context. 2.1. Trade The Canada-U.S. border comprises 130 land border crossings on the longest unguarded border in the world: 8890 kilometers. There are 200 million border crossings between Canada and the U.S. each year and traffic is expected to increase at a rate of 10% each year over the next 10 years (www.dfait-maeci.can-am.gc.ca). This should increase pressures for security measures. In the light of September 11th, the concept of the border being a place to fend off contraband, criminals and terrorists might be reconsidered: Quoting Flynn (2002): developing the means to manage terrorists’ threats and other transnational muck that is contaminating the integrative process within the global community is essential, but we need to liberate
Protecting Canada and the U.S. Against Terrorism
87
ourselves from the notion that the border is the best place for accomplishing this. Indeed, an over-reliance on the border to regulate and police the flow of goods and people can contribute to the problem.
Easy flowing borders are important to the commercial flow between Canada and the United States. Trade has doubled between Canada and the United States since the signing of the first free trade agreement in 1989 and now, more than 1.3 billion dollars worth of goods cross the Canadian-U.S. border each day. This trade partnership is the most important in the world. More specifically, 87% of Canadian exports are destined to the U.S. and 25% of U.S. exports are for the Canadian market (www.dfait-maeci.gc.ca/can-am/). Canada is concerned with any potential disruptions in the flow of cross-border trade since Canadians are highly dependent on its exports to the United States. For instance, the terrorist attacks of 9/11 caused the United States to close its borders for approximately 24 hours, bringing commercial flows to an almost complete stop. This had a profound effect on people and businesses trading under NAFTA. As an example, Flynn (2002) reports that Ford shut down five plants for a week with estimated losses of $1 million per hour. But concerns over border issues do not only lie in trade flows. Divergent policies on visa issuance, asylum to refugees and immigration are at the center of security issues as we now explain. 2.2. Border Policy and Management On September 12th the U.S. General Attorney accused Canada of being an easy access for terrorists. Canada’s Prime Minister Jean Chr´etien responded that it was the responsibility of the U.S. Border Patrol to control U.S. border crossings not Canada’s. Since then announcements were made on moving towards a common security perimeter, although Canada has often stated some concerns with maintaining its sovereignty over the border issues. The U.S. wanting more harmonization of immigration and border policies has created tension in an otherwise very good relationship. Given that the U.S. is an undeniable superpower, this could be read as “do it the American way” or find a lower common denominator. Canadians want closer ties in trade, but they also wish to remain sovereign over their own country. Other concerns, concerning refugee status-seekers, were voiced when the alleged Algerian terrorist Ahmed Ressam crossed the border in Washington State with the intention of bombing the Los Angeles airport on New Year’s Eve of 2000. Also, the allegations that some of the September 11th terrorists had entered the U.S. through Canada, and that Canada is a safe haven for terrorists, has reintroduced debate concerning the establishment of a common security perimeter. Over the last 10 years, both countries have increasingly coordinated border management. Today, more than 50 agencies (Canada and the United-States combined)
88
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
are implicated in border management. This represents complex management issues about coordination. Even before September 11th, many initiatives were being discussed and put forward by both countries. Initiatives such as the Shared Border Accord, the Border Vision, the Cross-Border Crime Forum and more importantly the Canada-U.S. Partnership Forum (CUSP) (www.dfait-maeci.gc.ca) have addressed issues such as promoting trade, reducing costs in control, intelligence sharing and transnational crime. These initiatives were put forward after the first attacks on the World Trade Center in 1993, as concerns for security and terrorists’ threats were heightened at that moment. The need to create a balance between the easy flow of goods and security factors became a very high burden on agencies. In December 2001, as a direct result from the CUSP dialogues and the events of 9/11, Canada and the U.S. signed the Smart Border declaration, a 30-point plan to enhance security while facilitating the flow of people and goods (www.dfait-maeci.can-am.gc.ca). This declaration is seen by many as the operational application of moving towards a common security perimeter. The 30-point action plan coordinates many aspects such as risk management for trade and immigration policies. The plan should account for the paradoxes of globalization: more movements of goods and people to promote trade and commerce, and the difficulty of control of transnational threats such as terrorism. Trying to manage border dialectic has become increasingly difficult for policy makers and the agencies mandated to implement these policies: (. . .) Policy makers anxious about reigning in globalization’s dark side look to the border to fend off contrabands, criminals, illegal migrants, and terrorists (Flynn, 2002).
3. THE BORDER: SOVEREIGNTY, TERRITORY AND SECURITY Borders serve as operational and symbolic instruments. Since we have shown the importance of the border relationship and the need for the U.S. and Canada to tighten border control, we examine here two potential options of changing the border coupled with the concepts of sovereignty, territory and security.
3.1. Sovereignty Previous allegations made by the Attorney General and the media resulted in the popular belief that Canada was: (i) easy to enter for terrorists; and (ii) by extension,
Protecting Canada and the U.S. Against Terrorism
89
an easy country for refugee asylum seekers. These beliefs are at the heart of the sovereignty issues: the nation-state has the right to choose who enters the country. Therefore, the concept of sovereignty lies within the different immigration, refugee status and defense policies. These allegations were also fed by the two cases of Ahmed Ressam and Nabil al-Marabh. As noted by Macklin (2002): both men’s refugee claims were rejected. Nor is there evidence that any of the actual perpetrators of the September 11th hijacking/bombing entered the U.S. as refugee claimants, much less Canada. Each apparently arrived with fixed-term visitor or student visas and then overstayed their visas, thereby joining the ranks of an estimated 6 million other migrants residing in the U.S. without legal status.
Linking the 9/11 events with the refugee/immigration falsely accuses the refugees as the trigger for these events. Considering this, the link between sovereignty, security and border control does not lie with the accusation of refugee seekers. The explanation lies more with the increase and acceleration of movement of goods and people resulting from FTA and with the decrease in resources dealing with border transit. As mentioned, sovereignty is clearly defined as the right to choose who enters the country. This is why, in the status quo option, both countries remain completely sovereign over their own policies and coordinate to the level they want, rendering possible a step-by-step approach. On the other hand, keeping the status quo means that the U.S. might tighten border control policies leading to potential negative effect on trade and crossing flows. Also, if Canada’s reputation as being easily accessible to terrorists remains unchanged, it could require more investments on its part. In the common security perimeter option, benefits can clearly be seen for U.S. sovereignty but none for Canada’s sovereignty. As mentioned before, the higher dependence of Canada on the U.S. would allow the U.S. to impose its immigration, refugee and defense policies through budget and resource allocations. Since the U.S. defense system is more developed and geared towards protection, it would have the upper hand at imposing its views, whereas the Canadian defense system is more oriented towards peace keeping.
3.2. Territory and Security The security of the territory is ensured by immigration and defense policies.1 Countries such as Canada and the U.S. exercise control over their territories through different policy and administrative tools such as visas and inspection. For example, recent changes in Canadian law have rendered possible the prosecution
90
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
of non-citizens under criminal law. These changes have created a shift in the perception that the foreigner is a source of danger for the security of the nation. This destabilization has created a doubt that the border is the best place to ensure the security of the country. Therefore, the concept of security is linked to immigration policies. However, as Macklin (2002) mentions it could be difficult to prove terrorism-related offense under criminal prosecution. If this is the case, it will be easier to deport than to imprison and this will bring on another difficult task. As terrorism knows no border, the deportation of an alleged terrorist to his/her own country could prove difficult for the security of Canada: “It is not obvious that deporting a person who is determined to engage in acts of terror to, say, Afghanistan, Sudan or Iraq will deny him the opportunity to pursue his objective at some later date, perhaps through other means.” These future terrorist acts could be on the Canadian embassy in the country of deportation. But the person could also decide to retaliate in another country creating a potential for an international NIMBY (Not In My Backyard) problem. The concepts of territory and security are also linked to the economy. As mentioned previously, the Canadian and the American economies are highly, but asymmetrically, interdependent. Therefore, not all the security objectives are the same. Golob (2002) clearly shows that the U.S. objective is to develop its security through homeland security as this “will make the U.S. not only safer but stronger, both absolutely and relative to other countries.” She also argues that the Canadian notion of security “is coupled with social inclusiveness and protection which communicates the government’s commitment to traditional measures of state legitimation through social program [. . .].” Nonetheless, the dependence of Canada’s economy on its U.S. neighbor makes it more inclined to cooperate with its U.S. counterpart on border security issues. As a direct result of September 11th, both countries have increased appropriation of funds for increased security. Canada announced a 7 billion (CAD) investment in “security and intelligence agencies” (www.dfait-maeci.gc.ca). The creation of the Homeland Security Department with a 36 billion U.S.D budget is a significant investment in security and represents the largest change in U.S. public management history (www.dhs.gov). These numbers show the importance of increasing security measures for both countries. Most of actual discussions between the two countries, through the Smart Border action plan, have been directed at more coordinating mechanisms related to data and intelligence sharing or administrative changes to reduce cost for trade issues. In considering the status quo option, the notion of security and territory is being addressed by both countries. The actual costs of maintaining and developing security will require further investments in resources for all of the agencies. This will also require more developments in technologies related to border security.
Protecting Canada and the U.S. Against Terrorism
91
For the private sector, corporations’ border crossing wait times are costly, and a potential border closing in the event of another terrorist attack would lead to substantial losses (as the September 11th events were). On the other hand, many coordination/collaboration issues are already under way and prove to be effective, increasing the security of both territories. Moreover, with the development and sharing of new technological tools, each country’s R&D costs are lowered. This option also renders laws, policies and procedures more harmonious, bringing benefits to both countries. A common security perimeter option would shift all costs for both countries associated with the control agencies of the actual border to the defence of the common perimeter. The benefits of no border are quite high for companies engaged in trade as they would bear no cost linked to customs issues. The creation of a common security perimeter would confer, it seems, more benefits for trade. On the other hand, we will see in the next section, that it will not necessarily increase security efficiency.
4. BORDER POLICY AND COOPERATION2 Blocking entry to terrorists at the border is often seen as the best way to decrease or eliminate terrorism within a country. For example, we saw that the response to 9/11 has been to increase the amount of resources devoted to the security of the Canada-U.S. border. Given the costs and benefits of border security controls, we can assume there is an efficient level of security such that the marginal cost of increased border security is equal to its marginal benefit. But a possible obstacle in achieving the efficient level of border security against terrorists results from the public nature of border security. First, border security can be seen as a good which is non-excludable. For example, consider the case of two countries sharing a common border like Canada and the United States. If Canada takes measures (invest in border security) to decrease entries by potential terrorists, this could also reduce entries by terrorists into the U.S.3 This is especially true given that NAFTA facilitates border crossings between Canada and the U.S. Therefore, the U.S. will benefit from Canadian measures. We cannot exclude a country from the benefits (at least in parts) stemming from investment in higher border security by an adjacent country. Second, the benefits generated by increased border security are in parts non-rival over countries. For one, terrorism imposes varying costs on people in all countries regardless of where a terrorist event occurs. Also, if a lower number of terrorists enters one country, this should benefit a neighbor. As mentioned earlier, this is especially the case if border crossings between the two are numerous
92
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
and easy (for instance, Canada and the U.S. under NAFTA). Hence, the fact that one country benefits from taking measures to decrease entries by potential terrorists on its territory, does not reduce the benefits therefore available to other targets. Therefore, since reducing the general threat of terrorist entries through increased security generates a public good for two or more countries, efficient border policies could require cooperation among countries benefiting from increased border security. For instance, this could take the form of a common security perimeter or of an increased collaboration (e.g.: immigration rules) leading to higher security while favoring trade. It is in this vein that the Canadian Foreign Affairs and International Trade Committee and the Canadian Council of Chief Executives (representing 150 of Canada’s largest corporations) have called for a North American perimeter and for a customs union between Canada and the United States (National Post, 01/23/03). However, motivating collaboration may be difficult because joint border security policies may involve a “prisoners’ dilemma” problem. Given that increased border security provides public benefits, each country would hope, to some extent, to free-ride on the increased security of the other. Acting rationally neither country would increase security and that could lead to the worst possible outcome. In the Canadian-American case, assuming that both countries are potential targets, this means that while both countries are better off increasing border security, each may under-invest given the level of increased security expected by the other. It is then possible for countries to free-ride, trying to let the other pay for increased security. At first look, the incentive to free-ride would be greater for countries believing themselves to be secondary targets. For instance, if Canada estimates a very low probability that it would be hit by terrorist activities and that the U.S. is the only target, it is more likely that it will free-ride and under-invest in border security. It seems then that the negative effects of free-riding would be worst when countries are asymmetric in terms of being primary or secondary targets. On the other hand, a closer look at this problem suggests that the “prisoners’ dilemma” aspect may not be a significant obstacle to optimal border security. This is the case for two reasons. First, the likelihood of joint increased security will be higher if there are country-specific benefits for a country improving security at its border. A reduction in terrorism in general necessarily means fewer attacks against specific countries and their citizens. If the United States can discourage terrorist attacks within its borders and against its citizens by implementing border security measures then most of the country-specific benefits accrue to the USA increasing
Protecting Canada and the U.S. Against Terrorism
93
their border security. If this is the case, the benefits of cooperation with Canada are reduced since it is more likely that the U.S. will increase its border security regardless of what Canada does. The obvious question then will concern Canada’s incentives to free-ride since it doesn’t seem to have many country-specific benefits. This leads us to the second reason why both countries will increase border security and why the free-riding or prisoners’ dilemma aspect may not be a significant obstacle to optimal border security. If because of specific benefits the U.S. increases security to reduce terrorism it will generate positive externalities. Increased U.S. border security should reduce the overall level of terrorism in both Canada and in the U.S. However, increased American border security could also generate negative externalities for Canada. This could be the case for two reasons. First, this could disrupt trade flows by increased security, and in turn increase costs given that 87% of Canadian exports are destined for the U.S. (see Section 2.1). Second, since increasing U.S. border security changes the relative costs of conducting terrorist activities in both countries (the relative costs of hitting both targets). This change could then shift terrorist activities to Canada.4 This in turn could lead to Canada increasing its own border security. If this is the case, therefore the country-specific benefits the U.S. would receive from better border security would even be higher! Hence, it is possible that if the U.S. has the motivation to increase border security so will Canada because it is aware that not doing it would have negative effects on its exports to the U.S. and that terrorist might be more attracted to attacking Canada given the higher difficulty (costs) to hit the U.S. Hence, given plausible conditions, it would then be possible to reach optimal security using somehow independent border policies. Given the two options in the standard prisoners’ dilemma setting, cooperation (increase border security for both countries) or non-cooperation (status quo), we believe it is very likely that both countries will cooperate and that therefore optimality will be reached.5 Let us give an example using a simple and plausible matrix showing that a possible equilibrium would be for both countries to increase border security. In the payoff matrix the rows represent the decision to increase security (not increase) by the U.S. while the columns represent the decision to increase (not increase) border security by Canada. The first number in each cell represents the payoff for the U.S. while the second is the payoff for Canada. The numbers in the matrix reflect the fact that the U.S. is the main target and that Canada can be victimized based on the fact that it is a U.S. friend. Also, we assume that (Increase, Increase) is the optimal outcome. For example, that outcome could be the one we aim to attain with a common security perimeter (complete collaboration).
94
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
Canada
USA Increase Not increase
Increase
Not Increase
100, 80 70, 40
80, 50 50, 60
First, note the U.S. advantage is to increase border security regardless of what Canada does. This is because the U.S. has important specific gains from higher border security. In other words, increasing border security is a dominant strategy for the U.S. On the other hand, it is to Canada’s advantage to increase its border security only if the U.S. does the same. First, because of potential negative effects on trade, and second, because of a potential shift of terrorists to conduct their activities in Canada as it would become difficult to do so in the U.S. with increased security. Since the U.S. always decides to increase border security both countries will do so.6 The free-rider solution is never attractive to the U.S. and therefore will never be attractive to Canada. Therefore, independent border policies may lead to optimal security. The point is not that collaboration, taking the form of a common security perimeter or more common immigration rules, doesn’t allow attaining optimal security. Rather, we should ask ourselves if the costs of losing independent policy making (sovereignty over trade, immigration, security, etc.), because of the adoption of a common security perimeter, are justified since we may well reach an optimal security level given the incentives we have described.
5. CONCLUSION In this paper, we considered the case of two countries sharing a common border and argued that the “public nature” of border security may be an obstacle in achieving optimal security against terrorism. We highlighted this possibility using the example of Canada and the U.S. In this particular case, we demonstrate that there may indeed exist benefits from co-operative border policies. However, we showed that motivating collaboration may be difficult based on the fact that joint border security policies may involve a problem similar to the prisoners’ dilemma. That is, given the increased border security provides public benefits, it is then possible for countries to free-ride and try to let others pay for increased security. In the Canada-U.S. case, assuming that both countries are potential targets, this
Protecting Canada and the U.S. Against Terrorism
95
means that while both countries are better off increasing border security, each may under-invest given the level of increased security expected by the other. The consequences of that problem are enormous. For instance, this may call for a solution involving complete co-ordination of policies. For instance, this could take the form of a North American security perimeter leading to higher security while favouring trade. That would be associated with significant losses of sovereignty for Canada. On the other hand, we also showed that the “prisoners’ dilemma aspect” may not be a significant obstacle to optimal border security if there are country-specific benefits for a country to improve security at its border. For instance, if the U.S. can discourage terrorist attacks within its borders by implementing border security measures then most of the country-specific benefits accrue to the U.S. increasing their border security. If this is the case, the benefits of co-operation with Canada would be reduced and the U.S. would be more likely to increase its border security regardless of what Canada does. We showed that if this is indeed the case, Canada may also have an incentive to increase its security since higher U.S. security could have negative effects on Canadian exports to the U.S. and could shift terrorist activities to Canada. Therefore, given these incentives it would be possible to reach optimal security using somehow independent border policies. A common security perimeter would not be necessary. Although recent changes in co-ordination through the Smart Border plan talk of a “common security perimeter,” the actions put forward show more of an evolution in co-operation. Considering what we discussed in this paper, it seems that this may be a good alternative.
NOTES 1. For the purpose of this paper, we consider the territory of each country as the geographic location. Nevertheless, as noted by Macklin (2002), recent events have challenged our understanding of territory and border: “it becomes obvious that assaults on a given state do not actually require access to its territory.” When embassies are attacked or cyber crime is perpetrated territories become less clear. If terrorists do not respect borders, Macklin concludes that “neither do states in pursuit of border control.” Therefore the world, as a global village, is becoming more and more streamlined along the borders as nations exert their powers over more than their own citizens. Considering these changes in the concept of territory, the security perimeter concept becomes less obvious. 2. This section is inspired by Lee (1988), Enders and Sandler (2003) and Sandler and Arce (2003). 3. We assume only two neighbor countries in the analysis. It would be a simple task to extend the analysis to more than two countries. 4. For more details on that result see Sandler et al. (1983) and Im et al. (1987). Also, for a good review on substitution effects in transnational terrorism see Enders and Sandler (2003).
96
GEORGES A. TANGUAY AND MARIE-CHRISTINE THERRIEN
5. Lee mentions another possibility in which one country in effect “sells” the public good of reduced terrorism that is generated by the other country. He calls that possibility “paid-riding.” We abstract from this possibility. 6. This is the Nash equilibrium of this game.
ACKNOWLEDGMENTS We thank Jason Michaud and Peter Small for their research assistance. We also thank Michele Fratianni, Tom Duchesneau and Philip Trostel for helpful comments. We are responsible for any remaining errors. Corresponding Author: University of Maine, Department of Economics and Canadian-American Center, Orono, ME 04469–5774 USA. georges.tanguay @umit.maine.edu.
REFERENCES Enders, W., & Sandler, T. (2003). What do we know about the substitution effect in transnational terrorism? In: F. Cass (Ed.), Researching Terrorism Trends, Achievements and Failures, forthcoming. Flynn, S. (2002). Transforming border management in the post-September 11 World. In: Governance and Public Security (pp. 37–51). Campbell Public Affairs Institute, Maxwell School of Citizenship and Public Affairs, Syracuse University. Golob, S. R. (2002). North America beyond NAFTA? Sovereignty, identity. Canadian-American Public Policy, 52(December), 1–44. Im, E. I., Cauley, J., & Sandler, T. (1987). Cycles and substitutions in terrorist activities: A spectral approach. Kyklos, 2(40), 223–255. Lee, D. R. (1988). Free riding and paid riding in the fight against terrorism. American Economic Review, 78(2), 22–26. Macklin, A. (2002). Borderline Security. Public Law and Legal Theory Research Paper No. 02–03. University of Toronto, Faculty of Law. National Post (01/23/03). Manley rejects dismantling border. Sandler, T., & Arce, D. G. M. (2003). Terrorism and game theory. Simulation and Gaming, forthcoming. Sandler, T., Tschirhart, J., & Cauley, J. (1983). A theoretical analysis of transnational terrorism. American Political Science Review (March), 36–54.
FURTHER READING www.dfait-maeci.gc.ca www.dhs.gov
WAL-MART IN MEXICO Syed Tariq Anwar ABSTRACT This case study analyzes Wal-Mart’s market entry and expansion in the Mexican market. Today’s Wal-Mart has become the largest retailer in Mexico. The case discusses all the current issues of this debate within the perspectives of growing the international firm and other internationalization issues. The case also evaluates the global retailing industry and its changing competition. Within the current circumstances, it is expected that Wal-Mart will continue to grow in Mexico and the NAFTA region. On the other hand Wal-Mart may face heightened competition and will be challenged by Mexican and other international retailers. The significance of this work lies in its timeliness and relevance to the ongoing debate of NAFTA-related boom and internationalization in the retailing industry.
INTRODUCTION This illustrative case discusses and evaluates Wal-Mart Stores, Inc. (hereafter called Wal-Mart) and its Mexican expansion. Since the late eighties, Wal-Mart has transformed itself from virtually an unknown retailer into the largest company in the world. On an average, Wal-Mart opens a new supercenter every other week in the U.S. (Consumer Report, 2002). Some even categorize Wal-Mart as a global player because of its operations in ten countries (Govindarajan & Gupta, 1999). However, a closer look at the company reveals that Wal-Mart is not a global retailer and its international operations are highly region-specific. The bulk of the
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 97–113 © 2004 Published by Elsevier Ltd. ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10006-5
97
98
SYED TARIQ ANWAR
company’s revenues are generated from the North American market (Rugman & Girod, 2003). The company is a major force in bringing significant structural changes in the retailing industry and other areas of information technology (IT). In Mexico, Wal-Mart has grown to be the largest retailing entity. Wal-Mart’s customer base is so large that it is a household name in North America. The company is closely monitored by the Federal Reserve and other financial institutions regarding quarterly consumer prices and other household consumption patterns (Business Week, 2001, 2002a, b; Financial Times, 2001; The Wall Street Journal, 2002a, b). According to one estimate, Wal-Mart’s central database is the second largest after the Pentagon. The company has contributed a quarter of the productivity to the U.S. economy between 1995 and 1999 (Financial Times, 2003a). For many years, Wal-Mart has been rated one of the top ten “most admired companies” in the U.S. (Forbes, 2002; Fortune, 2003; Useem, 2003). Wal-Mart employs 1.3 million workers (the largest corporate employer in the world) and serves 100 million customers weekly. In 2002, the company’s revenues surpassed $219 billion with a profit of $6.6 billion (see Tables 1 and 2). Wal-Mart’s success can be judged from the fact that currently, it is ranked number four in market value ($240.9 billion) behind General Electric, Microsoft and Exxon Mobil. In 2003, Wal-Mart constitutes 2.3% of the U.S. economy. Retailers that can challenge Wal-Mart at the global level are Carrefour, Ahold, Tesco and others (Business Week, 2002a; Deloitte Touche Tohmatsu, 2002; Fortune, 2002a, b; see Table 2). Wal-Mart was established in 1962 when Sam Walton (the founder) opened a discount store in Arkansas. The company is composed of six retail divisions (see Table 1). Wal-Mart’s early growth has been attributed to its strong entrepreneurial and well-disciplined corporate culture, low prices and commitment to small towns and rural areas in the U.S. The company is highly competitive in North America and gets favorable reviews from the market. It is also known for its high-profile brand name, excellent logistics and distribution system, information technology, and low prices. Wal-Mart’s inventory management and distribution system is one of the best in the world. This helps the company to lower its costs and continue expanding in the U.S. and overseas. Wal-Mart’s early successes benefited from its niche-oriented strategies, changing demographics, and strong business and consumer demand of the eighties and nineties. Because of Wal-Mart’s marvelous growth and its entrepreneurial culture, it is well praised by the world media. The Economist (2001, p. 55) comments: The World’s largest retailer still thinks of itself as a small-town outfit. That may be its greatest strength.
Wal-Mart in Mexico
99
Table 1. Profile of Wal-Mart (August 2002). A: Financial data Consolidated revenues Profit Market capitalization Current assets Total assets Current liabilities Long-term debt Return on assets Return on shareholders’ equity B: Corporate data/key businesses Location Retail divisions
Number of Wal-Mart stores Number of supercenters Number of SAMS clubs Number of U.S. neighborhood markets International units Total employees Number of shareholders (March 31,02) C: Board of directors Chairman President & CEO Chairman of the executive committee of the board Directors
D: Wal-Mart’s operations in Mexico Total number of supercenters Total number of SAM’s clubs Total number of CIFRA units Total employees
$219.81 billion (net sales increase:14%; the largest company in the world) $6.67 billion 240.91 billion (ranked 4th in the world) $28.24 billion $83.45 billion $27.28 billion $15.68 billion 8.5% 20.1% Bentonville, Arkansas Wal-Mart Stores, Supercentres, SAM’s Club, Neighborhood Market, Wal-Mart International, and walmart.com. 1,647 1,066 500 31 1,170 1.38 million 324,000 S. Robson Walton H. Lee Scott David D. Glass James W. Breyer, John T. Chambers, Thomas M. Coughlin, Stephen Friedman, Stanlet C. Gault, David D. Glass, Ronald Hernandez, Dawn G. Lepore, J. Paul Reason, Elizabeth A. Sanders, H. Lee Scott, Jack C. Shewmaker, Donald D. Sonderquist, Jose Villarreal, John T. Walton, and S. Robson Walton. 27 34 420 70,720
Source: Financial Times; The Wall Street Journal (various issues); Company Annual Report <www.walmart.com>.
100
Table 2. Global Retailers Ranked by Sales (2002). Sales (2002) (Billions)
Profit (2002) (Billions)
Market Cap. (2002) (Billions)
1. Wal-Mart (U.S.) 10 2. Carrefour (France) 24 3. Ahold (Netherlands) 25 4. Home Depot (U.S.) 6 5. Kroger (U.S.) 1 6. Metro (Germany) 22 7. Sears Roebuck (U.S.) 4 8. Target (U.S.) 1 9. Albertson’s (U.S.) 1 10. Kmart (U.S.) 4 11. Costo (U.S.) 7 12. Rewe Gruppe (Germany) 11 13. Safeway (U.S.) 2 14. Tesco (U.K.) 10 15. J. C. Penney (U.S.) 4 16. ITM Enterprises (France) 8 17. Auchan (France) 14 18. Aldi Gruppe (Germany) 11 19. Edeka Gruppe (Germany) 7 20. Ito-Yokado (Japan) 15 21. Tengelmann (Germany) 16 22. Walgreen (U.S.) 3 23. J. Sainsbury (U.K.) 3 24. CVS (U.S.) 1 25. Otto Versand (Germany) 24
$219.81 billion 62.22 59.63 53.55 50.09 44.34 41.07 39.88 37.93 36.15 34.79 34.68a 34.30 33.91 32.55 30.68a 29.13a 28.79a 28.77a 26.82 25.14a 24.62 24.57 22.24 21.56a
$6.67 billion 1.13 1.89 3.04 1.04 0.402 0.735 1.36 0.501 −2.41 0.602 NA 1.25 1.19 0.980 0.423 0.282 NA NA 0.421 NA 0.885 0.521 0.413 NA
$240.91 billion 35.84 19.71 97.97 17.70 10.59 18.59 37.59 14.31 NA 17.83 NA 19.69 26.41 6.45 NA NA NA NA 22.27 NA 39.16 10.66 12.55 NA
Sector/Store Type
DIS/WHC CC/CS/HM/SM/DIS CC/CS/DIS/SM/HM/S HI/SR CS/DS/SR/SM DIV/DS/HM/MO/SM/WHC DS/GM/MO DIS/DRS DRS/SM DIS/SR WH CC/DS/DIS/HM/SM SM CS/SM/HM DS/MO CS/DIS/HM/RES/SR CS/DIS/SM/HM DIS CS/DIS/SM/HM CS/DS/DIS/HM/RES/SM DIS/HM/SM/SR DS CS/SM/HM DS/SR MO
(Fortune) Global 500 Rank 1 35 38 46 56 72 83 89 100 104 111 NA 113 114 124 NA NA NA NA 161 NA 183 184 215 NA
Notes on Sector/Store Type: CS: Convenience Store; CC: Cash & Carry; DIS: Discount Store; DIV: Diversified; DM/MO: Mail Order; DS: Department Store; DRS: Drug Store; F: Food; GM: General Merchandiser; HM: Hypermarket; HI: Home Improvement; SM: Supermarket; SR: Specialty Retailer; RES: Restaurant; WH: Warehouse; WHC: Warehouse Club; NA: Not available. Source: Business Week (2002b, pp. 58–84); Fortune (2002b), Chain Store Age (2002). Global Retailing (December), 90–92; Deloitte Touche Tohmatsu (2002). a Private/2001.
SYED TARIQ ANWAR
Rank/Company/Home Country/ No. of Countries Operating
Wal-Mart in Mexico
101
Wal-Mart’s strength is also rooted in its geographic reach in the NAFTA region, merchandising strategy, and overseas acquisitions. Deloitte Touche Tohmatsu (2002, p. G7) comments: Wal-Mart remains the undisputed leader in the retail world. Its sales are three times as large as those of Carrefour, the second largest retailer.
Since 1992, Wal-Mart has sought aggressive diversification by acquiring various retailing chains in Mexico, Canada, Germany and the U.K. This change in Wal-Mart’s core competence beyond the U.S. market signifies its future growth and expansion. Analysts believe that in the next five to eight years, Wal-Mart plans on becoming a visible retailer in Europe and Latin America. On the other hand, Wal-Mart may face heightened competition and will be challenged by established national retailers.
BRIEF HISTORY OF WAL-MART The first Wal-Mart store was opened in Rogers, Arkansas by Sam Walton in 1962. Sam by nature was a down-to-earth individual as well as an eccentric and a “complex personality” but often showed “humility” while dealing with his customers (Journal of Business Strategy, 1999, p. 37).1 By 1967, the company was operating 24 stores with total sales of $12.6 million. In 1968, Wal-Mart expanded beyond Arkansas and opened its first store in Sikeston, Missouri and Claremore, Oklahoma. On October 2, 1969, the company was incorporated as Wal-Mart Stores, Inc. During this time, operations of the company continued to grow. In 1970, Wal-Mart opened its first distribution center in Bentonville, Arkansas which aimed at seeking more efficiency in its retailing model. By 1971, Wal-Mart was operating in five states (Arkansas, Kansas, Louisiana, Missouri and Oklahoma) and had 38 stores with total sales of $44.2 million. In 1972, Wal-Mart’s stock was listed on the New York Stock Exchange and during the same year, its stock was split 100% in March at the market price of $47.50. By 1979, Wal-Mart’s sales reached $1 billion and had 276 stores in 11 states with 21,000 employees. During the 1980s, Wal-Mart continued to grow at a very fast pace. In 1983, the company opened its first SAM’s Club in Midwest City, Oklahoma. In 1984, David Glass was named Wal-Mart’s president. In 1988, 90% of Wal-Mart stores were using bar-code scanning which helped the company monitor its inventory and other supply chain activities. By the end of 1989, Wal-Mart was operating in 26 states and its sales had reached $20.6 billion. During the 1980s, the company had eight stock splits which improved employee relations and corporate visibility.
102
SYED TARIQ ANWAR
Beyond the 1980s, Wal-Mart continued to grow and in the 1990s became the nation’s number one retailer in sales ($25.8 billion), eventually surpassing Sears Roebuck. In 1991, the company entered into international markets and opened its first SAM’s club in Polanco (near Mexico City). In 1992, Sam died and S. Robson Walton was named chairman of the board. By the end of 1995, Wal-Mart had operations in all 50 states. In 1997, the company became the number one employer in the U.S. By the end of 1999, Wal-Mart’s total sales reached $137 billion and employed 910,000 workers worldwide. During the same period, the company opened stores and collaborated in distribution operations in Argentina, Brazil, Canada, China, Germany, Hong Kong, Japan, Mexico, South Korea, and the U.K. In 2000, H. Lee Scott was named President and CEO of Wal-Mart (The Wall Street Journal, 1999, 2001). In June 2002, the company became the largest corporation in the world with sales surpassing $219 billion. As of 2003, Wal-Mart employs 1.3 million workers and is the largest corporate employer in the world. Wal-Mart has 3,244 stores in the U.S. and 1,173 international outlets. The company has six retail divisions which include Wal-Mart Stores, Supercenters, SAM’s Club, Neighborhood Market, International, and Wal-Mart.com (Wal-Mart, 2002). The company’s online store (Wal-Mart.com) is not a visible player in the market. Amazon.com, Tesco.com, Peapod, and others are doing better in the online retailing industry (Business Week, 2000a, b). As a corporate donor, Wal-Mart is a major contributor in the U.S. and the Walton family is very active in various charitable activities. It is interesting to note that the company revenues grew from $11 billion in 1987 to $219 billion in 2002. This is an achievement which is unheard in today’s corporate world. For the last ten years, Wal-Mart has been rated by Fortune as one of the top ten “most admired companies” in America (Fortune, 2002a). Wal-Mart’s organizational system provides two lessons regarding its model and growth: (1) the company maintained a clear focus and did not deviate from its original organizational system; and (2) the company did not sacrifice the core elements and stayed with its sustained competitive advantage. By looking at Wal-Mart’s growth, it is evident that a well managed organizational system can be a good source of a competitive edge in the market.
ISSUES IN GLOBAL/REGIONAL STRATEGY AND RETAIL INTERNATIONALIZATION Companies formulate global strategies on the basis of their market coverage, competition and product lines. A truly global product based on standardized
Wal-Mart in Mexico
103
Table 3. Global vs. Regional Strategies: Issues and Concepts. Global Strategies (Yip, 2003)
Regional Strategies (Moore & Rugman, 2002; Rugman, 2001)
A. Global strategy levers can help bring benefits such as cost reduction/economies of scale, quality improvement, global branding and customer preference, and better competitive leverage. B. Global strategy levers include global market participation, global products, global location of activities, global marketing and global competitive moves. C. Globalization drivers include market drivers, cost drivers, government drivers and competitive drivers.
A. Regional markets are growing.
B. Most of the business activities are centered around city/region-specific clusters. C. Intraregional trade is growing. D. Four entities are important: city clusters, nations, regions and the globe. E. Issues of regions/industry clusters are critical when formulating strategies and other analysis.
Source: Moore and Rugman (2002), Rugman (2001), Yip (2003).
features and marketing mix strategies is a myth. Maximum standardization can be achieved around the core product whereas the secondary or peripheral elements change according to the markets and consumption patterns. Global products bring a multitude of benefits such as economies of scale, superior quality, customer preference and visibility (Yip, 2003). The area of global strategy is influenced by factors such as globalization drivers, consumer convergence, MNCs’ worldwide integration, standardized products, reduction in trade barriers, and advances in IT. Furthermore, globalization drivers encompass market drivers, cost drivers, government drivers and competitive drivers (Yip, 2003). On the other hand, there is a new school of thought which contradicts the main philosophy of global strategy and suggests that MNEs operate on the basis of regions and city clusters, leading to “don’t think global – think regional” approach (see Table 3). The regional strategy approach suggests that companies operate on the basis of city/region-specific clusters and most of the business activities and regions are becoming the central hubs leading to more intra-regional trade (Moore & Rugman, 2002). Rugman (2001, p. 11) further advocates this new thinking and states: To summarize, my theme is that a paradox of globalization is that it has never really occurred anyway; it is myth. Instead, the vast majority of MNE manufacturing and service activity is (and has always been) organized regionally, not globally; . . . the literature advocating
104
SYED TARIQ ANWAR
globalization is far too simplistic. While there are some economic drivers of globalization there are extremely strong cultural and political barriers preventing the development of a single world market.
Since 1990, retail internationalization has been on the rise because of worldwide demand and growing markets. Retailing is no longer a country-specific phenomenon and is influenced by many opportunities in the world (Burt, Mellahi, Jackson & Sparks, 2002; Vida, Reardon & Fairhurst, 2000). On the other hand, regional strategies suggest that most of the retailers’ sales come from their home/regional markets rather than global operations (MIT Sloan Management Review, 2003; Rugman & Girod, 2003). As in domestic markets, marketing mix elements (4Ps: Product, Price, Place and Promotion) play a critical role in the process of retail internationalization. Other elements (depending on consumption and country factors) include population density, store image/design, shelf space, distribution channels, promotion, pricing, customer service, location, and quality of store employees (Alexander, 1997; Simpson & Thorpe, 1995). In emerging markets and developing countries, large supermarkets and discount stores are a new occurrence (Financial Times, 2002a, b). According to Deloitte Touche Tohmatsu’s 2002 retailing survey, only five retailers had operations in 24 countries. Marks and Spencer was the most internationalized retailer having operations in 33 countries whereas IKEA, Toys R’us, Ahold, and Carrefour had operations in at least 24 countries (Deloitte Touche Tohmatsu, 2002). In the post-1945 period, retailers from the U.S. maintained significant international expansion. In the sixties and seventies, retailers from Europe started to show more interest beyond their national markets (Akehurst & Alexander, 1995). Retail internationalization deals with transfer of retail management know-how, technology, business operations, and other activities in international markets which may be different from a company’s home environment. Because of a wide variety of opportunities available in global markets, the retail internationalization process encourages companies to seek “multi-format strategies,” i.e. supermarket, hypermarket, category killers, discount stores, etc. (Colla & Dupuis, 2002). Since retail internationalization is so diverse, it leads to four types of retailers in the international markets: (1) cautious internationalists; (2) emboldened internationalists; (3) aggressive internationalists; and (4) world powers (Treadgold, 1988). Retailers may seek three types of strategies in the process of international expansion which include: (1) global formula-standardized strategies worldwide; (2) multinational formula-strategies adapted to the local taste and lifestyles; and (3) investment/acquisitions in the host country (Salmon & Tordjman, 1989). Other important issues in the international debate encompass
Wal-Mart in Mexico
105
mode of entry issues, global acquisitions, firm-specific advantages and other internationalization areas.2 There are various models available in the process of retail internationalization. These models can be used on the basis of need, market environment, and competition. In addition, the models’ applications are determined by a company’s internationalization styles, i.e. multinational, transnational and global. A conceptual model PLIN (product, lifestyle, image and niche) is a simplistic investigation of category classification, merchandise offerings, lifestyle patterns of consumers, a retail image based on positioning, and niche strategy within the area of differential advantage (Simpson & Thorpe, 1995). Like MNEs, retailers at the global level are often affected by boundaries such as political boundaries, regulatory boundaries (state, intra-state, and supra-state), economic boundaries, social boundaries, cultural boundaries, and retail structure boundaries (Alexander, 1997). Two factors (pull vs. pull) contribute to retail internationalization. Retailers seek international expansion because of their saturated domestic markets or other difficulties at home. International entry strategies may encompass methods such as licensing, concessions, franchising, joint ventures, acquisitions, and greenfield operations. Regarding internationalization style, companies normally adopt three forms when seeking expansion outside their national markets, i.e. multinational, transnational, and global (McGoldrick, 1995).
COMPETITION IN THE GLOBAL RETAILING INDUSTRY In the last ten years, the global retailing industry has expanded at a very fast pace because of surge in consumer demand and availability of new information technologies. The industry has witnessed mergers and consolidations in North America and Europe. In 2002, global retail sales stood at $7 trillion. Top ten retailers controlled 8% of the world retailing industry. Out of the first 200 global retailers, 78 originated from the U.S. Wal-Mart alone controlled 9.1% of the global retailing market. Tables 1 and 2 clearly present Wal-Mart’s dominance in the global retailing industry. The company is a towering giant in sales, profit, market capitalization and other areas. Companies like Carrefour, Ahold, Home Depot, Kroger, Metro, and Sears Roebuck stand in the shadow of Wal-Mart in each and every category. On the other hand, most of the U.S. retailers are not truly globalized and have stayed in North America. This could be attributed to the size of the U.S. market, overseas competition, regulatory barriers and regional opportunities (see Fig. 1).
106
SYED TARIQ ANWAR
Fig. 1. Major Global Retailers: Sales and Net Income (1997–2002). Source: Fortune; Chainstoreage; Company Annual Reports (various issues).
European retailers have branched out all over the EU market. In 2002, the top 100 European retailers had 74% of their operations in the EU, Eastern European and Scandinavian countries. Contrary to North America’s manufacturing MNEs, U.S. retailers have stayed home although this trend may change because of competition and saturated markets. In many countries, trends such as liberalized/free trade, market-based economies, advances in education and healthcare, and efficient supply-chains have brought opportunities to retailers. Other factors which have contributed to retail internationalization include
Wal-Mart in Mexico
107
improvement in quality of life, reduction in poverty and surge in population (Deloitte Touche Tohmatsu, 2002). Competition in the global retailing industry continues to heat up in the developed as well as emerging markets. In some regions, the retailing environment is further complicated by regulatory barriers and bureaucratic hurdles. The retailing industry is always impacted by the country-of-origin (COO) issues and national affiliations. In addition, from the consumer side, the industry has changed significantly regarding prices and demand. The internet is also playing an important role in changing the industry structure. In order to compete effectively, more and more retailers are seeking mergers and collaborative arrangements.
WAL-MART’S INTERNATIONALIZATION AND MARKET ENTRY IN MEXICO In 1991, Wal-Mart entered the international market and opened a store near Mexico City. The company started with a fifty-fifty joint venture with Cifra SA because of its knowledge and visibility in the Mexican market. By 1992, Wal-Mart had opened six SAM’s Clubs in Mexico. In its early years, Wal-Mart encountered a multitude of problems in Mexico which were typical of an emerging market. During the mid-nineties, Wal-Mart continued to face problems in the areas of merchandising, pricing, market research and the regulatory environment. Mexico’s higher tariffs, inadequate infrastructure, store locations, weak supplier network, and scattered distribution system created additional hurdles for the company which eventually led to high prices and negative publicity. Other problems culminated from government bureaucracy, inventory management, high transportation cost and the 1994 Peso crisis. The Peso crisis caused consumers to spend less in the market which temporarily halted Wal-Mart’s expansion (The Wall Street Journal, 2002b). In 1997, Wal-Mart paid $1.2 billion to acquire the rest of Cifra. This was a major FDI initiative by an outside retailer (The Wall Street Journal, 1997). Currently, Wal-Mart de Mexico operates 551 units in Mexico (443 discount stores, 62 superstores and 46 SAM’s Clubs) and is the largest retailer with sales over $13 billion (Wal-Mart, 2002). The company is also the third largest publicly traded company in Latin America after Petrobras and Telmex. Wal-Mart has capitalized on the opportunities which came with the NAFTA treaty. Although Wal-Mart initially encountered a multitude of problems in Mexico, its business is currently surging because of cost efficiencies, supplier network, and good inventory management. In the last ten years, Wal-Mart has learned from its experience in the Mexican market and made changes which are compatible with its North American expansion.3,4
108
SYED TARIQ ANWAR
In Mexico, Wal-Mart brand continues to receive good ratings because of its low prices and store locations.
WAL-MART’S STRATEGY AND PROBLEM AREAS IN MEXICO Global strategy of a company encompasses three major areas: (1) formulating the core strategy; (2) internationalizing the core strategy based on global expansion; and (3) globalizing the international strategy. In international markets, a company is surrounded by four types of globalization drivers which in one way or another effect its global strategy. The four globalization drivers include market drivers, cost drivers, government drivers, and competitive drivers (Yip, 2003). Of course, at each and every level of expansion, a company needs to standardize or adapt its operations and product policy areas to meet the local market. Regional strategies may be more appropriate when dealing with different demographics, lifestyles intra-regional trade. As stated earlier, Wal-Mart first applied its American retailing model in Mexico but ended up with low sales and bad publicity because of problems in the areas of its globalization drivers, supplier network, store locations and pricing. Later on, Wal-Mart re-evaluated its operations and gradually started to establish its brand name in Mexico. Of course, geographical proximity to the U.S. also helped the company to open additional stores after 1998. The following are Wal-Mart’s strategies and problem areas in Mexico: (1) At the time of its entry in Mexico, Wal-Mart’s main goal was to seek an established partner having the same type of core competence and core elements. For this reason, Wal-Mart made a fifty-fifty joint venture agreement with Cifra to materialize this goal. Being a cash-rich company, Wal-Mart had no problem seeking expansion in the Mexican market. In addition, Wal-Mart’s major foray in Mexico was the result of cheap labor, distribution system and inventory management (Colla & Dupuis, 2002). (2) A global strategy describes that most companies first internationalize their core business when seeking global expansion. In its early days, Wal-Mart used the same merchandising principle (everyday low prices) although it had difficulty managing its pricing structure. Currently, Wal-Mart is paying more attention to the local lifestyles and consumer tastes and has accumulated a good learning curve to be competitive on the basis of merchandising and cost structure. The company has made a network of local suppliers which help in the supply chain management. Wal-Mart keeps on changing its merchandising strategy and product offerings. Mexican lifestyles
Wal-Mart in Mexico
(3)
(4)
(5)
(6)
(7)
(8)
109
and local consumption patterns are given preferences over its American retail model. Mexico’s retailing industry is becoming more competitive because of competition from national retailers. Companies such as Commercial Mexicana, Grupo Gigante and Organizacion Soriana continue to capitalize on their national image, product offerings and commitment to the local market. Wal-Mart on the other hand, continues to benefit from its retail size and low prices and is paying more attention to the local consumers and suppliers (Business Week, 2002c). In any organizational system, managers and employees are equally part of the strategy process. Companies often share and communicate their plans at each and every level. This has been found to create harmony among the managers (Kim & Mauborgne, 2002). Wal-Mart is good in communicating with its employees and its organizational system is flexible and well focused. The retailing sector in Mexico continues to grow because of cheap labor and the NAFTA-related boom. Wal-Mart has centralized inventory management and IT systems that use its Arkansas facilities. All the purchases go through Wal-Mart’s international division. This has helped to reduce cost and bring efficiency (Financial Times, 2002b). The company’s expansion strategy and its Mexican retailing model have been so successful that it was copied by other Wal-Marts units in Argentina, Brazil and China (The Wall Street Journal, 2002b). Mexico has a large consumer market and is expected to grow in the coming years. Wal-Mart is well positioned to capitalize on the changing demographics and other trends in the Mexican market. As stated earlier, Wal-Mart is the largest retailer in Mexico and its sales are expected to reach $18 billion in the next five years although local retailers continue to remain strong in selected segments (Financial Times, 1999). Unlike Europe and Asia, Wal-Mart has become a household name in Mexico although local retailers still carry significant clout regarding consumer loyalty and product offerings. Since 1998, Wal-Mart’s simple pricing concept “everyday low prices” has been highly successful. The company has devised a retail format which targets five demographic segments and its multi-format approach and Bodega (low end stores including SAM’s Clubs) has been so successful that the model was copied by Wal-Mart in Brazil under the retailing format “Todo Dia” (Financial Times, 2002b). Retailing models practiced in emerging markets are somewhat different. U.S.-style of mass merchandising and discount stores may not be available. Although Mexico’s local chains (Commercial Mexicana, Grupo Gigante,
110
SYED TARIQ ANWAR
Organizacion Soriana and Puerto de Liverpool) maintain a strong presence in the market, they are unable to compete with Wal-Mart. For example, last year, Grupo Gigante saw its sales drop by 5% because of Wal-Mart’s growing presence in the Mexican market (Ratnesar, 2003).
WHAT LIES AHEAD IN THE MEXICAN MARKET? Within the present circumstances, the following changes are expected in the Mexican market: (1) Mexican retailing is expected to grow in the coming years because of changing demographics and consumer demand. As an emerging market, Mexico continues to grow because of low tariffs, younger population and NAFTA-related opportunities. According to industry analysts, Wal-Mart plans on making inroads in other parts of Mexico as well as South America. (2) Wal-Mart has a significant competitive advantage over its Mexican rivals because of efficient distribution system, low prices and strong brand name. In the last few years, Wal-Mart has launched its own private electronic exchange to bring additional efficiencies in its procurement process. Local retailers in Mexico do attract brand loyal customers but are losing ground to Wal-Mart’s “everyday low prices” and market size. Competition in the Mexican market is expected to heat up since Wal-Mart has plans for future expansion (Financial Times, 2003b). (3) Price wars are a normal feature in the Mexican retailing industry. Wal-Mart is able to outperform its competitors by providing low prices based on a network of Mexican suppliers and large purchase contracts. This is the core competence of Wal-Mart which brings savings and helps in the expansion process. (4) With all the growth and expansion in Mexico, Wal-Mart’s dealings with its suppliers have raised concerns. In 2002, the company was investigated by Mexico’s Federal Competition Commission (CFC) regarding its “aggressive and abusive practices” (Business Week, 2002c).
CONCLUDING COMMENTS As of 2003, Wal-Mart continues to enjoy its growth and expansion in Mexico. Apart from its early blunders, the company is on the right track and has become the largest retailer in Mexico. Like the U.S. and Canada, the company continues to accumulate an impressive market share since NAFTA brought a windfall because
Wal-Mart in Mexico
111
of low tariffs and prices. In 2002, Wal-Mart’s market capitalization surpassed $240 billion and was ranked fourth in the world after GM, Microsoft and Exxon-Mobil. According to industry analysts, the company plans on making inroads in other parts of Mexico and South America. In the U.S., Wal-Mart continues to expand in the areas of its core business but sales have started to saturate. Competition in the Mexican retailing industry may be heightened because of the availability of growing markets and surge in consumer demand. Many analysts see Wal-Mart as a unique retailer because of its sheer size and market share. Mexican retailers have a hard time creating the model which Wal-Mart established. Wal-Mart has the resources and the expertise to seek additional expansion in Mexico if economic conditions improve in North America. The company has reached to a point where it is difficult for competitors to imitate its model.
NOTES 1. For more discussion on Sam Walton, see: Trimble, Vance H. (1990). Sam Walton: The Inside Story of America’s Richest Man. New York: A Dutton Book; Fortune (1992). Sam Walton in His Own Words (June 29), 98–106; Walton, Sam and John Huey (1993). Sam Walton: Made in America – My Story. New York: Doubleday; Ortega, Bob (2000). In Sam We Trust: The Untold Story of Sam Walton and Wal-Mart, the World’s Most Powerful Retailer. New York: Times Business Books. 2. For more discussion in the area of international market entry stages and internationalization, see the works of the following researchers who have been at the forefront of this debate: Bartlett, Buckley, Casson, Caves, Cavusgil, Dunning, Ghoshal, Guisinger, Hymer, Johanson, Kindleberger, Kogut, Roth, Rugman, Stopford, Vahlne, Vernon, and Wells. 3. For more discussion, see The Wall Street Journal (2001), Financial Times (2002b), Financial Times (2002). Mexican Stores Wilt in the Face of Wal-Mart’s Onslaught (August 14), 15. 4. The material in the case is intended to be used as a basis for discussion rather than to illustrate either effective or ineffective handling of a managerial situation or business practices.
REFERENCES Akehurst, G., & Alexander, N. (1995). The internationalization process in retailing. In: G. Akehurst & N. Alexander (Eds), The Internationalization of Retailing (pp. 1–15). London, UK: Frank Cass. Alexander, N. (1997). International retailing. Oxford, UK: Blackwell. Burt, S. L., Mellahi, K., Jackson, T. P., & Sparks, L. (2002). Retail internationalization and retail failure: Issues from the case of Marks and Spencer. International Review of Retail, Distribution and Consumer Research, 12(2), 191–219.
112
SYED TARIQ ANWAR
Business Week (2000a). Will Walmart.com get it right this time? (November 6), 104–112. Business Week (2000b). This race isn’t even close (December 18), 208. Business Week (2001). The Marines learn new tactics – from Wal-Mart (December 24), 74. Business Week (2002a). Productivity: A retail link (June 10), 30. Business Week (2002b). The business week global 1000 (July 15), 62. Business Week (2002c). War of the superstores (September), 60. Colla, E., & Dupuis, M. (2002). Research and managerial issues on global retail competition: Carrefour/Wal-Mart. International Journal of Retail and Distribution Management, 30(2), 103–111. Consumer Report (2002). Where to buy (July), 11–15. Deloitte Touche Tohmatsu (2002). Global powers of retailing. Stores (January), 1–48. The Economist (2001). Wal around the world (December 8), 55. Financial Times (1999). Who’s afraid of Wal-Mart? (May 5), 15. Financial Times (2001). Forget IT, it was Wal-Mart behind that U.S. miracle (October 17), 9. Financial Times (2002a). An octopus in the shopping trolley (January 11), 9. Financial Times (2002b). Wal-Mart hopes to clone its Mexican cash cow (April 6/7), 10. Financial Times (2003a). As annual sales reach $240 billion, can Wal-Mart conquer markets outside the U.S.? (January 8), 11. Financial Times (2003b). Walmex plans expansion as sales rise 10% (July 3), 20. Forbes (2002). The shiniest reputations (March 4), 70–72. Fortune (2002a). The shiniest reputations in tarnished times (March 4), 70–72. Fortune (2002b). Fortune. Global 500 (July 22), F1–F13. Fortune (2003). America’s most admired companies (March 3), 81–94. Govindarajan, V., & Gupta, A. K. (1999). Taking Wal-Mart global, strategy and business (4th quarter), 14–25. Journal of Business Strategy (1999). Strategists of the century, 20(5), 37. Kim, W. C., & Mauborgne, R. (2002). Pursuing the holy grail of clear vision. Financial Times (August 6), 8. McGoldrick, P. J. (1995). Introduction to international retailing. In: P. J. McGoldrick & G. Davies (Eds), International Retailing: Trends and Strategies (pp. 1–14). London, UK: Pitman Publishing. MIT Sloan Management Review (2003). The myth of globalization? (Winter), 11. Moore, K., & Rugman, A. (2002). Don’t think global – think regional. Strategy & Business (3rd quarter), <www.strategy-business.com>. Ratnesar, R. (2003). Fresh from the border. Time Inside Business (April), 65–72. Rugman, A. (2001). The end of globalization. New York: AMACOM. Rugman, A., & Girod, S. (2003). Retail multinationals and globalization: The evidence is regional. European Management Journal, 21(1), 24–37. Salmon, W. J., & Tordjman, A. (1989). The internationalization of retailing. International Journal of Retailing, 4(2), 3–16. Simpson, E. M., & Thorpe, D. I. (1995). A conceptual model of strategic considerations for international retail expansion. The Service Industries Journal, 15(4), 16–24. Treadgold, A. (1988). Retailing without frontiers. Retail and Distribution Management, 16(6), 8–12. Useem, J. (2003). One nation under Wal-Mart. Fortune (March 3), 64–78. Vida, I., Reardon, J., & Fairhurst, A. (2000). Determinants of international retail involvement: The case of large U.S. retail chains. Journal of International Marketing, 8(4), 37–60. The Wall Street Journal (1997). Wal-Mart to buy majority stake in Cifra (June 4), A3.
Wal-Mart in Mexico
113
The Wall Street Journal (1999). Logistics whiz rises at Wal-Mart (March 11), B. The Wall Street Journal (2001). How Wal-Mart transfers power (March 27), B1 & B4. The Wall Street Journal (2002a). Wal-Mart’s ugly suit (May 30), A17. The Wall Street Journal (2002b). How NAFTA helped Wal-Mart dominate the Mexican market (August 31), A1 & A2. Wal-Mart (2002). <www.walmart.com>. Yip, G. S. (2003). Total global strategy II. New York: Prentice-Hall.
PART II: NORTH AMERICA FINANCIAL INTEGRATION AND A COMMON CURRENCY
ASSESSING THE BORDER: TESTS OF THE LAW OF ONE PRICE IN CANADA AND THE U.S. Janet Ceglowski ABSTRACT This paper investigates the role of the border in Canadian and U.S. prices, based on a sample of highly disaggregated city-level retail prices. It finds substantial short-run differences in cross-border prices. While most of these are eliminated over time, long-run differences in the cross-border prices remain. These long-run cross-border differences average just over 20%, compared to mean long-run intranational price gaps of 7–9%. Short-run price differences are eliminated at similar rates in the cross-border and intranational data. Evidence from national average prices suggests the gap between cross-border prices has not narrowed during the recent depreciation of the Canadian dollar.
INTRODUCTION The economic ties between Canada and the U.S. are extensive. Bilateral CanadaU.S. trade flows are the largest in the world. These trade ties have been bolstered by the Canada-U.S. Free Trade Agreement and NAFTA, which have reduced or eliminated many bilateral tariffs, and complemented by a history of foreign direct investment. Moreover, as Helliwell (1996) observes, Canada relies on U.S. exports for many fresh produce items that are either not grown in Canada North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 117–136 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10007-7
117
118
JANET CEGLOWSKI
(oranges) or only produced domestically during the warmer months (lettuce). The two countries are well situated for trade, sharing a 5,500 mile border with 130 land border crossing points. Indeed, the closest large market for some border cities lies across the border. Traffic across the border is heavy, with over 200 million border crossings a year (Canada-U.S. Partnership Forum, 2000). But do the two countries constitute a single, integrated product market? One way to address this question is to examine cross-border price behavior in the two countries. If Canada and the U.S. form a single market, prices of identical goods in the two countries should exhibit a tendency to converge or, at least, a stable relationship to one another. One price-based benchmark of market integration is the absolute version of the Law of One Price which posits parity between prices of identical goods in different markets. This paper examines cross-border price behavior by testing for price parity in a unique set of finely disaggregated retail prices sampled in a number of Canadian and U.S. cities. The analysis relies on actual price data, not price indexes, allowing for tests of absolute price parity. Such tests are relatively rare in empirical international economics owing to the dearth of international prices for comparable products. They are even less common for Canada and the U.S. However two recent papers have used price indexes to examine cross-border price behavior in Canada and the U.S. Engel and Rogers (1996) analyze disaggregated consumer price indexes in Canadian and U.S. cities, finding that cross-border relative price indexes are substantially more variable than intranational relative price indexes. Rogers and Jenkins (1995) perform a time-series analysis of relative Canada-U.S. price indexes for a set of disaggregated consumer goods and services, identifying only a handful that are stationary. Though their analysis examines price behavior for highly disaggregated products, it relies on national price indexes. In contrast, this paper adds a geographic dimension to the price convergence question, using actual retail prices identified by time, product, and city that allow for direct tests of absolute parity. Helliwell (1998) lists numerous reasons for prices to differ across locations: transportation costs, trading networks that are more easily accessible nearby or within the same country, differing tastes, non-traded goods in the underlying price indexes, currency fluctuations in combination with prices that are sticky in local-currency terms, and trade barriers. Of these, he notes that only currency fluctuations and possibly trade barriers are exclusive to cross-border prices. Thus, a variety of factors can contribute to intranational price differences. In theory, those same factors could apply internationally, generating cross-border price differences that have nothing to do with the border. For this reason, this analysis uses the behavior of intranational prices in Canada and the U.S. as a benchmark against which to evaluate cross-border price behavior. If the markets are fully integrated, price behavior across the border should resemble that within each
Assessing the Border
119
country. However evidence of cross-border deviations from the Law of One Price should be evaluated against price behavior within each country. In other words, the degree of product market integration between Canada and the U.S. is best gauged relative to the degree observed within each country.
THE DATA The prices used in this analysis consist of quarterly series for 21 consumer products from 25 Canadian cities and 24 U.S. cities over the period 1982:2–1993:4. The Canadian data are from Statistics Canada’s City Average Retail Price Survey. For each Canadian city in the survey, the prices are reported as weighted averages of prices from individual stores, where the weights correspond to the overall market share of each store in the relevant city.1 Statistics Canada phased out the survey in 1994 so this analysis ends in 1993. The U.S. data are from the American Chamber of Commerce Research Association’s (ACCRA) Cost of Living Index, which reports the results of a quarterly price survey conducted by volunteers. ACCRA publishes average retail prices for consumer goods and services from a sample of establishments for each reporting location. However, the reporting cities and their number vary from quarter to quarter. Consequently the U.S. price series contain frequent missing values. The U.S. cities included in this analysis were selected by eliminating all those with more than four missing values for any of the included products over the sample period.2 Because these surveys were designed to provide cost-of-living comparisons across cities, each reports actual retail prices for a narrowly-defined sample of consumer products. The Canadian survey consists of prices for food items and a few non-food consumer goods while the U.S. survey covers a wide range of consumer goods and services. Twenty-one items were priced in both surveys: babyfood, bacon, bananas, bread, chicken, cigarettes, coffee, cornflakes, detergent, eggs, hamburger, lettuce, milk, orange juice, potatoes, soda, steak, sugar, tissues, tomatoes, and toothpaste. The prices of these items form the data set used in this analysis and are described in greater detail in the Appendix. The data were collected during the same weeks in both surveys, facilitating cross-survey comparisons. In some cases, the reported sizes of the products differed in the two surveys so adjustment factors were applied in order that prices for similar sizes would be compared.
SHORT-RUN ANALYSIS The absolute version of the Law of One Price predicts parity between the prices of identical products located in different markets. When the comparison is between
120
JANET CEGLOWSKI
prices in two different countries, absolute price parity would mean that for good k and countries i and j, e t p ikt /p jkt = 1 where e is the bilateral exchange rate. In log form, absolute parity would imply that rpijkt = ln(e t p ikt /p jkt ) = 0. Do the crossborder prices exhibit absolute parity? Some initial insights can be obtained by examining the logged cross-border relative prices, rpijkt , for each of the 21 sample products. For this purpose, a benchmark city was selected for each country: Toronto and St Louis. Toronto is an obvious choice for an economic center for Canada while St Louis was chosen because of the relative completeness of the data for that city. Cross-border relative price series were formed by pairing prices of good k in each U.S. city with Toronto prices for the same good at time t. Likewise, prices in each Canadian city were paired with St Louis prices for the same good. For most goods, this produced 48 cross-border relative price series. The weekly average bilateral exchange rate (e), was used to convert the prices to a common currency: Canada-St Louis:
rpikt = ln(et pikt /pSt Louis,k,t )
U.S.-Toronto:
rpikt = ln(pikt /et pToronto,k,t )
where i = Canadian city i, i = Toronto where i = U.S. city i
Because the relative price series are in logs, they provide a gauge of the size of the deviations from absolute price parity. Table 1 presents the minimum and maximum absolute cross-border price differences for each of the sample products. In most cases, the minimum is zero, implying that absolute price parity holds for at least one period in the sample. The maximum absolute differentials are substantial, exceeding 100% for several products. The average absolute crossborder differences are considerably smaller, with mean and median values of 30 and 25%, respectively. Simple averages of the cross-border differences are similar in magnitude to the average absolute differences but their signs indicate a tendency for U.S. prices to be lower than Canadian prices when expressed in a common currency. Are these cross-border price deviations large? A comparison with other cross-country price deviations can provide one perspective. Parsley and Wei (2001) report an average absolute deviation of 108% and an average deviation of −87% between disaggregated U.S. and Japanese prices. Relative to those statistics, the average deviations in Table 1 are small. Another perspective can be obtained by comparing the magnitude of the cross-border price deviations to similar intranational measures for Canada and the U.S. For this purpose, two sets of intranational prices were formed from the log ratio of the price in city i relative to the price in the relevant benchmark city: Canada-Toronto:
rpikt = ln(pikt /pToronto,k,t )
U.S.-St Louis:
rpikt = ln(pikt /pSt Louis,k,t )
where i = Canadian city i, i = Toronto where i = U.S. city i, i = St Louis
Assessing the Border
121
Table 1. Cross-Border Price Differences (%). Product
Babyfood Bacon Bananas Bread Chicken Cigarettes Coffee Cornflakes Detergent Eggs Hamburger Lettuce Milk Orange juice Potatoes Soda Steak Sugar Tissues Tomatoes Toothpaste Mean Median
Minimum Absolute Difference
Maximum Absolute Difference
CanadaSt Louis
U.S.Toronto
CanadaSt Louis
U.S.Toronto
0 0 0 2 13 1 0 0 10 0 0 0 0 2 0 0 0 0 0 0 0
0 0 0 0 2 16 0 0 24 0 0 0 0 0 0 1 0 0 0 0 0
73 71 73 136 113 117 59 59 107 90 75 128 95 63 120 106 63 103 42 74 73
74 94 77 98 128 108 62 52 133 79 60 89 101 96 97 95 59 65 52 58 74
Average Absolute Difference CanadaSt Louis
U.S.Toronto
36 25 17 53 70 54 15 21 65 35 16 26 29 37 24 40 15 24 13 24 16
30 33 17 39 66 53 19 13 70 28 14 18 50 40 23 28 15 21 16 14 25
28.5 25
27.5 25
Average Difference Canada- U.S.St Louis Toronto 36 24 5 53 69 54 8 −19 64 35 −11 18 28 32 2 38 −12 −17 9 −23 11 17.6 18
−30 −32 −10 −38 −66 −53 −13 3 −70 −24 −9 −2 −50 −40 −4 −24 10 20 −14 9 −23 −20 −23
Notes: For the Canada-St Louis data, price deviations are measured as rpijkt = ln(e t p ikt /p St Louis,kt ), i = Toronto where e is the U.S.$/Canadian $ exchange rate, pik is the price of k in Canadian city i and pStLouis,k is the price in St Louis. For the U.S.-Toronto data, price deviations are measured as rpijkt = ln(p ikt /e t p Toronto,kt ) where e is the U.S.$/Canadian $ exchange rate, pik is the price of k in U.S. city i and pToronto,k is the price in Toronto.
The minimum and maximum absolute values of each of the intranational series are shown in Table 2. As in the case of most cross-border relative prices, the minimum is zero for each intranational series. The maximum absolute differences vary widely by product though all suggest short-run deviations from absolute parity in the intranational series. Notably, for several individual products the intranational maximums do not differ appreciably from the cross-border maximums. However the average and average absolute intranational differences are relatively small and, in most cases, less than the mean cross-border differences. In fact, the mean and
122
JANET CEGLOWSKI
Table 2. Intranational Price Differences (%). Product
Babyfood Bacon Bananas Bread Chicken Cigarettes Coffee Cornflakes Detergent Eggs Hamburger Lettuce Milk Orange juice Potatoes Soda Steak Sugar Tissues Tomatoes Toothpaste Mean Median
Minimum Absolute Difference
Maximum Absolute Difference
CanadaToronto
U.S.St Louis
CanadaToronto
U.S.St Louis
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
29 32 69 54 61 53 43 59 49 65 91 73 67 39 111 73 56 83 43 50 64
49 67 67 112 76 29 60 52 64 105 59 87 35 50 76 74 41 62 32 55 66
Average Absolute Difference
Average Difference
CanadaU.S.Canada- U.S.Toronto St Louis Toronto St Louis 7 9 14 12 13 7 9 15 11 11 22 16 24 10 16 14 12 16 13 9 13
11 14 15 22 19 7 11 10 11 16 11 21 10 12 19 16 11 14 8 12 12
11.9 13
12.3 12
3 −6 −2 8 −8 0 −6 −8 −5 3 −20 4 −24 −8 −3 5 −4 −7 −10 −6 −10 −4.5 −6
3 −3 −3 7 12 0 1 −7 −1 8 1 12 2 −1 1 9 3 10 4 −8 −2 2.1 1
Notes: For the intranational data, price deviations are measured as rpijkt = ln(p ikt /p jkt ) where pik is the price of good k in city i and pjk is the price of good k in Toronto (Canada-Toronto) or St Louis (U.S.-St Louis).
median absolute cross-border differences are two to three times their intranational counterparts, suggesting a tendency for price gaps in the cross-border data to be larger than those in the intranational data. The gaps between the average cross-border and intranational differences are much larger, in accordance with the evidence of consistently lower U.S. prices in Table 1. Two principal findings emerge from Tables 1 and 2. First, short-run deviations from absolute parity characterize both the intranational and cross-border prices. This is hardly surprising in that information and transportation costs make instantaneous price equalization across geographically-distant markets unlikely within a nation’s borders, much less across international boundaries. Second, the
Assessing the Border
123
cross-border price differences are larger, on average, than the intranational price differences. These observations relate to short-run price behavior; the statistics in Tables 1 and 2 say little about the long-run properties of the relative price series. In particular, they are silent on whether the cross-border prices are stationary, tending toward a constant (or zero) long-run differential, or wander indefinitely apart. Clearly, the latter would be incompatible with cross-border price parity and market integration. Thus the analysis turns next to the long-run properties of the relative price series.
LONG-RUN PROPERTIES This section investigates whether long-run parity holds for the Canada-U.S. price series, using the relative cross-border prices formed from the St Louis and Toronto benchmarks. Parity could be absolute, implying no long-run difference in the cross-border prices when expressed in a common currency. Alternatively, parity could be relative if there is a constant long-run proportionality between the cross-border prices. In both cases, long-run parity also requires that the relative prices are stationary. Unit root tests can be used to determine whether parity is a long-run property of the cross-border prices. This involves a test of a unit root null and an alternative of a mean-stationary process in the following equation: rpikt = ␣ + 0 rpikt−1 +
n
s rpikt−s + e ikt
(1)
s=1
Including a constant term in: (1) introduces the possibility of convergence to a non-zero mean, allowing for long-run relative price parity. Persistent differentials between Canadian and U.S. prices could arise from differences in taxes, local non-traded input costs, or income. They could also result from arbitrage costs associated with transporting and selling goods across the border; the differentials could persist so long as they fell short of the arbitrage costs, forming a band of no arbitrage as in O’Connell and Wei (2002). In this sample, differences in product sizes or other characteristics could contribute to persistent price differentials.3 The results of augmented Dickey-Fuller (ADF) tests applied to the cross-border relative price series are summarized in Table 3. In these regressions the lag length, n, was determined for each series in a sequential testing-down procedure; beginning with a maximum of 8 lags, Eq. (1) was estimated by sequentially reducing the number of lags until the coefficient on the last included lag was statistically significant at the 5% error level. Evidence of mean-stationarity is found in a substantial number of the univariate regressions; the hypothesis of a
124
Table 3. Univariate ADF Tests for Cross-Border Prices. Canada-St Louis (n = 24)a
Product # That Reject Unit Root at 10%
1 19 20 8 22 0 4 3 8 20 20 15 8 11 11 2 13 12 15
1 21 21 9 23 0 5 7 15 20 21 19 12 12 14 2 13 14 18
−0.136 −0.586 −0.738 −0.386 −0.625 −0.147 −0.225 −0.493 −0.630 −0.493 −0.649 −0.891 −0.229 −0.391 −0.541 −0.118 −0.426 −0.349 −0.490
# of ␣ Signif at 5%
1 23 9 11 24 1 7 22 13 16 14 3 11 6 19 10 11 0 9
Long-Run Gap (%)
33 21 6 50 71 101 4 16 62 38 12 14 28 30 8 63 14 17 9
# That Reject Unit Root at 5%
10%
5 15 17 3 5 0 0 3 3 17 13 18 0 11 8 3 7 6 12
6 16 17 5 7 0 1 4 5 17 17 18 0 16 8 7 10 8 14
0
# of ␣ Signif at 5%
Long-Run Gap (%)
−0.183 −0.555 −0.677 −0.251 −0.224 −0.130 −0.298 −0.489 −0.513 −0.934 −0.068 −0.763 −0.343 −0.269 −0.314 −0.453 −0.251 −0.024 −0.296
9 18 12 12 13 1 0 20 13 13 7 4 6 10 3 17 9 0 7
27 34 12 44 70 92 12 6 66 28 10 7 69 39 10 29 13 23 17
JANET CEGLOWSKI
Babyfood Bacon Bananas Bread Chicken Cigarettes Coffee Cornflakes Detergent Eggs Hamburger Lettuce Milk Orange juice Potatoes Soda Steak Sugar Tissues
5%
0
U.S.-Toronto (n = 24)
Assessing the Border
Tomatoes Toothpaste
1 12
1 15
Median % of total
46
54
−0.178 −0.508
22 11
28 13
4 8
11 8
31
39
21
−0.490 50
−0.248 −0.331
9 18
9 19 23
−0.298 40
Notes : n refers to the number of relative price series for each product. The reported results refer to the following univariate ADF equation: rpikt = ␣ + 0 rpikt−1 +
n
s rpikt−s + e ikt
s=1
where the lag length n was determined for each series using the sequential testing-down procedure described in the text and a maximum lag of eight. The reported values of 0 are the median estimates from the univariate ADF equations. For each product, the long-run price gap is calculated as the median of |−␣/0 |. a n = 12 for orange juice and n = 22 for soda.
125
126
JANET CEGLOWSKI
unit root can be rejected at the 10% level in 54% of the Canada-St Louis price pairs and 39% of the U.S.-Toronto pairs. These results indicate that long-run crossborder convergence is a defining characteristic of many of the relative prices in the sample.4 In order to gauge the rate at which the cross-border prices converge to their long-run values, half lives for temporary price differences can be computed as −ln(2)/ln(1 + 0 ). Using the median values of the estimates for 0 produces relatively short half lives of one quarter for the Canada-St Louis series and two quarters for the U.S.-Toronto series (Table 3). A number of the intercept terms for Eq. (1) are statistically significant, indicating that for those cross-border prices that are stationary, reversion is frequently to a non-zero constant. Estimates of the implied long-run deviations from absolute price parity reveal median differentials of 21–23%.5 Among those products for which evidence of stationarity is found, the implied long-run differentials vary widely from a low of under 10% (bananas, coffee, cornflakes) to a high of 70% (chicken). The largest long-run differentials are associated with products for which either trade restrictions exist (chicken), cross-border size differences could be significant (detergent, soda), or the unit root null could not be rejected (cigarettes, milk). As before, these findings can be compared to similar evidence for the intranational price series. Table 4 presents the results of univariate unit root tests for mean-stationarity applied to the relative intranational prices for the Toronto and St Louis benchmarks. Evidence of mean-stationarity is found with slightly greater frequency in the intranational series: the unit root null can be rejected at the 10% error level for 56% of the Canada-Toronto series and 64% of the U.S.-St Louis series, compared to 54% of the Canada-St Louis series and 39% of the U.S.-Toronto series. In addition, there are only small differences in the median estimates of 0 across the four sets of ADF estimates, indicating minor differences in the average rates at which short-run price differences are eliminated in the cross-border and intranational series.6 In fact, the principal dimension in which the cross-border results differ from the intranational results is in the estimates of the intercept and the implied long-run price differentials. The estimates imply median long-run intranational differentials of 7–9%, indicating the persistence of intercity price differences, albeit small ones, within a nation’s borders. However for the cross-border estimates, the median longrun price differentials are on the order of 20%, two to three times the size of the longrun intranational differentials. Thus the evidence indicates there is little difference in the incidence of stationarity in the cross-border and intranational relative prices. The same is true for the rate of convergence for a number of the sample products. In these dimensions, the border does not impact price behavior to a large degree. However cross-border prices are distinguished by the relative size of their long-run
Canada-Toronto (n = 24)a
Product # That Reject Unit Root at
Babyfood Bacon Bananas Bread Chicken Cigarettes Coffee Cornflakes Detergent Eggs Hamburger Lettuce Milk Orange juice Potatoes Soda Steak Sugar Tissues
5%
10%
8 16 14 10 10 10 18 3 16 15 17 22 1 3 14 18 12 12 4
10 16 15 13 14 12 21 3 18 16 17 22 1 3 15 18 13 14 6
0
−0.363 −0.615 −0.530 −0.437 −0.341 −0.288 −0.543 −0.169 −0.390 −0.429 −0.520 −0.803 −0.021 −0.173 −0.840 −0.685 −0.595 −0.406 −0.273
# of ␣ Signif at 5%
2 2 9 4 5 4 11 7 9 10 3 11 9 7 11 10 6 4 7
U.S.-St Louis (n = 23) Long-Run Gap (%)
3 9 7 9 11 1 6 7 9 10 18 7 76 14 9 10 6 11 16
# That Reject Unit Root at 5%
10%
3 14 20 10 10 3 12 16 13 16 20 18 5 16 17 12 15 10 14
5 16 20 13 11 6 14 19 19 18 22 18 9 20 18 12 17 12 15
0
# of ␣ Signif at 5%
Long-Run Gap (%)
−0.149 −0.505 −0.899 −0.325 −0.333 −0.191 −0.457 −0.486 −0.506 −0.644 −0.673 −0.793 −0.234 −0.454 −0.798 −0.427 −0.585 −0.394 −0.446
8 14 5 9 16 15 3 15 18 11 4 13 7 9 12 12 15 2 15
8 5 7 9 11 5 6 5 5 6 4 12 7 3 10 9 5 11 7
Assessing the Border
Table 4. Univariate ADF Tests for Intranational Prices.
127
128
Table 4. (Continued ) Canada-Toronto (n = 24)a
Product # That Reject Unit Root at 5%
10%
Tomatoes Toothpaste
10 16
13 17
Median % of total
51
56
0
−0.317 −0.485
# of ␣ Signif at 5%
U.S.-St Louis (n = 23) Long-Run Gap (%)
8 5
7 10
# That Reject Unit Root at 5%
10%
13 8
17 10
55
64
9
−0.429
0
# of ␣ Signif at 5%
Long-Run Gap (%)
−0.385 −0.324
14 6
10 10 7
−0.454
29
46
Notes: n refers to the number of relative price series for each product. The reported results refer to the following univariate ADF equation: rpikt = ␣ + 0 rpikt−1 +
n
s rpikt−s + e ikt
s=1
JANET CEGLOWSKI
where the lag length n was determined for each series using the sequential testing-down procedure described in the text and a maximum lag of eight. The reported values of 0 are the median values from the univariate ADF equations. For each product, the long-run price gap is calculated as the median of |−␣/0 |. a n = 13 for orange juice and n = 23 for soda.
Assessing the Border
129
differences. Even then, though, the median cross-border price gap is less than 25% and only two to three times the median long-run intranational price gap. The sample used here consists of prices of only 21 products, many of which are food items. Cross-border prices of more differentiated consumer goods or non-traded products like some services might exhibit less evidence of long-run convergence or larger long-run price differences. On the other hand, the fact that this sample involves final retail prices might be expected to work against finding evidence of convergence. In other words, cross-border price parity could be more widespread for wholesale prices of relatively homogeneous primary and intermediate goods.7 Clearly additional research is needed to determine the degree to which these results extend to other cross-border prices.
MORE RECENT EVIDENCE Because the preceding analysis ends in 1993, it doesn’t directly address the recent effects of the free trade agreements between Canada and the U.S. or the sustained undervaluation of the Canadian dollar. However some initial observations about the recent behavior of cross-border prices can be inferred from national average retail prices for a subset of the 21 products analyzed above. In 1995 Statistics Canada began to publish Canadian average retail prices for many of the same goods that were previously included in its City Average Retail Price Survey. For purposes of comparison, these Canadian averages can be matched to the mean U.S. city prices from the ACCRA survey.8 Prices for 18 of the 21 sample products were available from these two sources. However there are potential timing differences between the ACCRA survey and the survey used to construct the Canadian averages. For this reason, Table 5 simply presents the changes in the local-currency price ratios, pU.S.,k,t /pCanada,k,t , and the monthly average bilateral exchange rate between January (first quarter) 1997 and October (fourth quarter) 2002, a period of persistent depreciation of the Canadian dollar.9 The Canadian dollar depreciated by nearly 16% during this period. The mean and median changes in the local-currency relative price ratios were −2 and 2% respectively, indicating that the changes in the relative prices were small both absolutely and relative to the size of the Canadian dollar depreciation. These central measures obscure substantial variations in the relative price behavior of the individual products. For a few items (babyfood, cigarettes, tomatoes, and toothpaste), the relative price changes were of the same sign and roughly the same magnitude as the exchange rate change. In these cases, there was no significant increase or decrease in the gap between the common-currency cross-border prices. For others (bananas, bread, coffee, cornflakes, soda, and
130
JANET CEGLOWSKI
Table 5. National Average Cross-Border Prices (% Change). Product Babyfood Bananas Bread Chicken Cigarettes Coffee Cornflakes Eggs Hamburger Milk Orange juice Potatoes Soda Steak Sugar Tissues Tomatoes Toothpaste
% Change (January 1997 to October 2002) 10.8 2.0 0.3 −9.2 15.7 3.4 4.2 −23.8 −23.0 −5.7 −51.0 −6.4 2.5 −9.9 −18.8 0.4 14.3 12.6
Mean Median
−2.4 2.2
Exchange rate (C$/U.S.$)
15.7
Note: The cross-border price ratios are measured in local-currency terms as pU.S.,k,t /pCanada,k,t . The bilateral exchange rate is the monthly average. Sources: Author’s calculations based on data from Statistics Canada, Average retail prices for food and other selected items, CANSIM II Table 326–0012; ACCRA, Cost of Living Index; and the Pacific Exchange Rate Service, http://pacific.commerce.ubc.ca.
tissues) U.S. prices increased relative to Canadian prices but at rates substantially below the rate of depreciation. For the rest, U.S. prices fell relative to Canadian prices, substantially so in the case of eggs, hamburger, orange juice, and sugar. Combined with the decline in the Canadian dollar, these latter two categories of relative price changes indicate a widening of any gap between the cross-border prices. For none of the products in Table 5 did an increase in the relative U.S. price outweigh the fall in the Canadian dollar. In this limited sample, therefore, movements in the local-currency relative prices and the exchange rate did not reduce but, in most cases, widened the cross-border price differences during the recent period of Canadian dollar depreciation.10 While suggestive, these findings should be considered preliminary pending additional research into the behavior of cross-border prices during the sustained undervaluation of the Canadian dollar.
Assessing the Border
131
CONCLUSIONS This study analyzes the behavior of cross-border prices in a three-dimensional sample of retail prices for 21 consumer goods in 25 Canadian cities and 24 U.S. cities. It finds short-run differences in the cross-border prices, some of which are substantial. However it also finds evidence of long-run cross-border price parity for a number of the sample goods, ranging from simple produce items like bananas to manufactured and differentiated products like toothpaste. Moreover, the estimates indicate that temporary cross-border price differences are eliminated quickly. In fact, the incidence of stationarity and, in many cases, the average rate of price convergence in the cross-border sample are comparable to those for the intranational sample, indicating that in these two ways there is little difference in the behavior of the cross-border and intranational relative prices. However the cross-border prices exhibit larger long-run differences. The estimates imply median long-run cross-border price differences of 21–23%, compared to median long-run price differences of 7–9% in the intranational estimates. These findings suggest that the principal dimension in which the border affects the long-run behavior of relative prices is the size of the long-run price differences. Even so, most of the long-run cross-border price differences are modest both absolutely and relative to their intranational counterparts. More recent evidence from national average retail prices suggests the gap between cross-border prices did not narrow and may have widened during the recent period of Canadian dollar depreciation.
NOTES 1. The cities are Calgary, Charlottetown, Chicoutimi, Edmonton, Halifax, Hamilton, Hull, London, Moncton, Montreal, Ottawa, Quebec City, Regina, St John, St John’s, Saskatoon, Sherbrooke, Sudbury, Sydney, Thunder Bay, Toronto, Trois Rivieres, Vancouver, Victoria, and Winnipeg. The reported prices include sales taxes when applicable. See Statistics Canada (January/March 1988, Appendix 4) for a full description of the data and collection methods. 2. The reported prices are averages, excluding sales tax, from samples of generally between 5 and 10 establishments within the urbanized area of each location. See ACCRA Cost of Living Manual (2000) for additional details on the data collection. The cities used in this analysis are Abilene TX, Appleton WI, Benton Harbor MI, Birmingham AL, Blythe CA, Casper WY, Cedar Rapids IA, Denver CO, Fort Wayne IN, Hastings NE, Houston TX, Indianapolis IN, Lexington KY, Lubbock TX, Louisville KY, Mobile AL, New Orleans LA, Omaha NE, Palm Springs CA, Rapid City SD, Reno/Sparks NV, St Louis MO, Salt Lake City UT, Vermillion SD.
132
JANET CEGLOWSKI
3. When the reported sizes differed in the two surveys, prices for all but one product were adjusted to account for those differences. Even in these cases, however, there is room for persistent price differentials. For example, a 2-liter bottle of cola in Canada is probably not 2.67 times the price of a 750 ml bottle. Slight differences in the products themselves could be another source of price differentials. For instance, frozen orange juice (U.S. survey) and bottled orange juice (Canadian survey) may not be viewed as exact substitutes. Detergent prices were not size-adjusted because of both a fundamental difference in the units used in each survey (volume in Canada, weight in the U.S.) and several changes in the reported sizes and densities of the product over the sample period. 4. Eleven of the products in this sample also appear in Rogers and Jenkins’ (1995) analysis of relative Canada-U.S. price indexes. Among this subset of eleven, the results in Table 3 reveal considerably more evidence of stationarity than do Rogers and Jenkins’ univariate ADF tests. This could be attributable to the added geographic dimension of the data used here; instead of a national average price index for each good as in Rogers and Jenkins, this analysis is based on city-level prices. That added dimension could include significant local price differences across both cities and time that need not show up in a national average. To the extent that it introduced more price variability, it could translate into more volatile relative cross-border prices and thereby contribute to stronger evidence of stationarity. 5. For each product, the absolute value of the implied long-run differential is calculated as the median of |−␣/0 | based on the univariate ADF estimates. 6. This is consistent with Crucini and Shintani (2002), a multi-country study that reports no difference in convergence rates for intranational and international prices. 7. However, Dunn (1970) estimates ordinary least squares regressions for 1950–1962, finding little evidence of a statistically-significant relationship between changes in the bilateral exchange rate and changes in relative Canada-U.S. wholesale prices for six intermediate and commodity goods. Richardson (1978) analyzes disaggregated Canadian and U.S. wholesale price indexes over the period 1965–1974, finding that arbitrage falls short of the absolute law of one price. 8. The recent ACCRA data sometimes includes prices for a limited number of Canadian cities. I excluded the Canadian data from the calculations of the U.S. means. 9. Note that while the data are not seasonally adjusted, it is unlikely that seasonal factors would differentially affect the Canadian and U.S. price series. Nonetheless, an analysis for the period October 1996 to October 2002 confirms the findings detailed below. 10. Statistics Canada (October/December 1995, Appendix 2) cautions that the quality, brands, and outlets in their average price survey vary, which limits the usefulness of the resulting prices as metrics of pure price changes over time. However the disaggregated consumer price indexes constructed by Statistics Canada are not subject to the same limitations. Though not shown, findings similar to those in Table 5 follow when the Canadian average prices are replaced by the comparable disaggregated consumer price indexes. This suggests that the relative price movements in Table 5 are not driven by commodity changes in the Statistics Canada average price survey.
ACKNOWLEDGMENT I would like to thank Michele Fratiani for insightful comments.
Assessing the Border
133
REFERENCES American Chamber of Commerce Researchers Association (2000). ACCRA cost of living manual. www.accra.org. American Chamber of Commerce Researchers Association. Cost of living index. Various issues for 1982–2002. Canada-U.S. Partnership Forum (2000). Building a border for the 21st century. Ottawa: Department of Foreign Affairs and International Trade. Crucini, M., & Shintani, M. (2002). Persistence in law-of-one-price deviations: Evidence from Microdata. Working Paper 02-W22, Department of Economics, Vanderbilt University. Dunn, R. (1970). Flexible exchange rates and oligopoly pricing: A study of Canadian markets. Journal of Political Economy, 78, 140–151. Engel, C., & Rogers, J. (1996). How wide is the border? American Economic Review, 86, 1112– 1125. Helliwell, J. (1996). Do national borders matter for Quebec’s trade? Canadian Journal of Economics, 29, 507–522. Helliwell, J. (1998). How much do national borders matter? DC: Brookings Institution. O’Connell, P., & Wei, S. J. (2002). The bigger they are, the harder they fall: Retail price differences across U.S. cities. Journal of International Economics, 56, 21–53. Parsley, D., & Wei, S. J. (1996). Convergence to the law of one price without trade barriers or currency fluctuations. Quarterly Journal of Economics, 111, 1211–1236. Parsley, D., & Wei, S. J. (2001). Explaining the border effect: The role of exchange rate variability, shipping costs, and geography. Journal of International Economics, 55, 87– 105. Richardson, J. D. (1978). Some empirical evidence on commodity arbitrage and the law of one price. Journal of International Economics, 8, 341–351. Rogers, J., & Jenkins, M. (1995). Haircuts or hysteresis? Sources of movements in real exchange rates. Journal of International Economics, 38, 339–360. Statistics Canada. Consumer prices and price indexes. Various issues for 1982–1995.
APPENDIX: DATA SOURCES AND DEFINITIONS The U.S. price data are from American Chamber of Commerce Researchers Association (ACCRA), Cost of Living Index. David Parsley kindly provided the U.S. price data used in Parsley and Wei (1996). The U.S. data used here differ from that in Parsley and Wei in two ways: missing values are not used to indicate reported size changes and the sample period covers 1976–1993. The Canadian price data were provided electronically by Statistics Canada but are published in Statistics Canada’s Consumer Prices and Price Indexes. The weekly average exchange rates are from the Pacific Exchange Rate Service, http://pacific.commerce.ubc.ca. Distances were calculated as the shortest road route in miles from www.freetrip.com. The product definitions are taken from the above data sources:
134
JANET CEGLOWSKI
Product
Canada
U.S.
Babyfood
Generally available manufacturers’ brands, strained or vegetable mixture in 4.5 oz glass jar or can. Store brands excluded
4–4.5 oz jar, strained vegetables, lowest price
Bacon
Generally available manufacturers’ brands, cooked side bacon, sliced, rindless, vacuum packed, packers first line, 500 g. Store brands excluded
1 lb package. Lowest price 1982:2–1985:4, Oscar Mayer, Hormel Black Label, Armour, or Field’s 1986:1–1991:3. Series ends in 1991:3
Bananas
First grade, price per lb or kg
Price per lb
Bread
Manufacturers’ brands, pan style standard loaf, sliced and wrapped, prepared from white wheat flour, vitamin enriched, 16–32 oz
24 oz loaf of white bread, lowest price
Chicken
Eviscerated fresh or frozen, broiler or fryer, whole, Grade A, 1 or 2 kg, price per lb or kg
Whole fryer, Grade A, lowest price, price per lb
Cigarettes
Generally available popular brands, king size, filter tip, carton of 200
Carton of king-size Winstons. Prices include all state and Federal taxes
Coffee
Generally available manufacturers’ brands of roasted whole beans, regular, or all purpose ground coffee, good quality. Store brands excluded. 454 g bag 1982:–1983:1, 369 g bag 1983:2–1993:4
Vacuum-packed can, Maxwell House, Hills Brothers, or Folgers. 1 lb 1982:2–1988:3, 13 oz 1988:4–1993:4
Cornflakes
Generally available manufacturers’ brands, 675 g package. Store brands excluded
Kellogg’s or Post Toasties, 18 oz
Assessing the Border
135
Product
Canada
U.S.
Detergent
Generally available manufacturers’ brands, all purpose, granulated or powdered. Store brands excluded. 6 liters 1982:–1991:3, 4 liters 1991:4–1993:4
Tide, Bold, or Cheer. 49 oz 1982:2–1988:4, 42 oz 1989:1–1990:3, 39 oz 1990:4–1991:3, 42 oz “ultra” 1991:4–1993:4
Eggs
Large, grade A, in a carton of 1 dozen
Large, grade A, 1 dozen
Hamburger
Regular with 24–30% fat content, fresh, 1 lb or kg
Price per lb, lowest price
Lettuce
Fresh, green, good quality, average size head, price per lb or kg
Iceberg, price per head (approximately 1.25 lbs)
Milk
Fresh fluid whole dairy milk, pasteurized, may be homogenized or contain vitamin D. Disposable 1 liter carton
Whole milk. Half-gallon carton
Orange juice
Generally available manufacturers’ brands, unsweetened or 5% sugar added, 48 oz. Store brands excluded
Frozen. 6 oz can, lowest price 1982:2–1985:4, 12 oz can Minute Maid brand 1986:1–1993:4
Potatoes
Canada No. 1 white table potatoes, 10 lbs
10 lb bag, white or red
Soda
Generally available manufacturers’ brands, cola flavor. Returnable 750 ml bottle, excluding deposit. Store brands excluded
2 liter Coca Cola, excluding deposit
136
JANET CEGLOWSKI
Product
Canada
U.S.
Steak
Sirloin cut from section of hind quarter next to short loin, bone-in or bone-out, fresh, Canada A or B brand, 1 lb or kg
T-bone USDA Choice, price per lb
Sugar
White granulated, cane or beet, 2 kg
White, cane or beet, lowest price. 5 lb 1982:2–1992:3, 4 lb 1992:4–1993:4
Tissues
Generally available manufacturers’ brands, package of 200 double tissues, approximately 20.6 cm × 23.9 cm. Store brands excluded
Kleenex brand. 200-count box 1982:2–1982:4, 175-count box 1983:1–1993:4
Tomatoes
Generally available manufacturers’ brands of canned field tomatoes, choice grade, 28 oz
Hunt’s or Del Monte canned tomatoes. 15–17 oz 1982:2–1985:4, 14.5 oz 1986:1–1993:4
Toothpaste
Generally available manufacturers’ brands of standard dentifrice, may contain fluoride, 100 ml. Store brands excluded
6–7 oz tube, Crest or Colgate
THE IMPACT OF NAFTA ON THE INTEGRATION OF THE CANADIAN, MEXICAN, AND U.S. EQUITY MARKETS Claire G. Gilmore and Ginette M. McManus ABSTRACT This paper tests for the existence of a long-run co-movement between the three North American stock markets of Canada, Mexico, and the U.S. and examines whether or not the implementation of the North American Free Trade Agreement (NAFTA) has led to more integrated equity markets. Application of the Johansen and Juselius (1990) cointegration procedure indicates a long-term relationship. A vector error-correction (VEC) model establishes that all three markets are involved in the long-run adjustment toward equilibrium. Overall, the results suggest that the implementation of NAFTA has promoted greater economic integration between the three North American countries.
1. INTRODUCTION The North American Free Trade Agreement (NAFTA) was initiated in 1990 when the U.S. began negotiating with Mexico for a free-trade agreement and
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 137–151 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10008-9
137
138
CLAIRE G. GILMORE AND GINETTE M. McMANUS
Canada requested a three-way treaty. NAFTA was signed in 1992, but later side agreements on labor and environmental issues were added and the new trade regime for North America took effect on January 1, 1994. Economically, NAFTA eliminates trade and capital restrictions between North American countries. The agreement provides for immediate or phased-out elimination of tariffs and other trade barriers on products and services traded, as well as promotes capital flows between its members. Although NAFTA specifically addresses the real asset markets of the three North American nations, the financial assets traded on the North American stock markets represent claims on these assets and are therefore affected as well. An important empirical question is whether the implementation of NAFTA has promoted greater economic integration or less segmentation in both real and financial sectors of the North American economies. The reductions in trade barriers, the development of greater economic ties between North American countries, and the financial deregulation inherent in NAFTA, such as the relaxation of restrictions on cross-country investing and foreign stock ownership, would be expected to contribute to greater integration of Canadian, Mexican, and U.S. stock markets. Countries experiencing greater economic integration or interdependence through liberalization of trade and capital flows across borders should also experience greater integration or degree of co-movement in their stock markets. While the hypothesis of economic and equity market integration has been examined in several studies, relatively little attention has been directed toward the investigation of possible long-run co-movement between the stock markets of Canada, Mexico, and the U.S. This paper uses cointegration techniques to test the long-run relationship between the equity markets of these North American countries over the 1994–2002 period. In addition, we examine the short-run dynamics between these markets. Our results, based on weekly data over the 1994–2002 period, indicate that the stock markets of Canada, Mexico, and the U.S. are cointegrated, with all three participating in the long-term adjustment. Furthermore, there is some evidence of short-run interaction between the markets. The results suggest that NAFTA has succeeded in liberalizing the equity markets across the three North American countries. The remainder of the paper is organized as follows. Section 2 discusses the relevant literature and explains the contribution made by the study. In Section 3 the methodology is briefly presented and methodological details are provided in the Appendix. We discuss the data in Section 4. Section 5 reports the empirical results and Section 6 contains the concluding remarks.
The Impact of NAFTA on the Integration
139
2. BACKGROUND AND LITERATURE REVIEW Many studies have tested for the presence of cointegration between various stock markets, both within and across regions, including Eun and Shim (1989), Jeon and Von Furstenberg (1990), Kasa (1992), Chan et al. (1992, 1997), Smith et al. (1993), Byers and Peel (1993), Arshanapalli and Doukas (1993), Chowdhury (1994), Rogers (1994), Blackman et al. (1994), Allen and MacDonald (1995), Kwan et al. (1995), Richards (1995), Monadjemi and Perry (1996), Karolyi and Stulz (1996), DeFusco et al. (1996), Hassan and Naka (1996), Choudhry (1994, 1997), Chaudhuri (1997a, b), and Kanas (1998). While these studies have examined capital market integration across developed and emerging economies of Europe, Asia, and Latin America, few studies have investigated the possible existence of long-run co-movements between the Canadian, Mexican, and U.S. stock markets since the implementation of NAFTA in 1994. Ewing et al. (1999) examine the long-run co-movement of the North American stock markets using cointegration tests over the 1987–1997 period. They detect no evidence of cointegration in these markets, either before or after the passage of NAFTA. On the other hand, Clinebell et al. (1998) perform similar tests over the 1989–1997 period and conclude that while the stock markets in the three members countries were not cointegrated before NAFTA, they have become so since the agreement was implemented. Both studies use monthly observations over a relatively short time span for the post-NAFTA period (1994–1997). These limitations may very well explain why these papers reach opposite conclusions on the issue of long-term co-movements between the stock markets of Canada, Mexico, and the United States. Darrat and Zhong (2001) formulate several tests to assess the possible impact of NAFTA on the degree of integration between the U.S., Mexican, and Canadian equity markets using weekly observations over the period from June 1, 1989, to December 31, 1999. While they also found evidence against cointegration in the pre-NAFTA sub-period, their results suggest strong interdependence of the markets in the post-NAFTA sub-period (from January 1, 1994, to December 31, 1999). In this paper we contribute to this literature by analyzing the nature of both the long-term and the short-term relationships between the equity markets of Canada, Mexico, and the U.S. since the implementation of NAFTA in 1994. We examine data for the 1994–2002 period, thus utilizing a sufficiently long time period to allow for more reliable findings on the long-term as well as short-term properties of the North American equity markets. In addition, both weekly and monthly observations will be utilized to facilitate the comparisons of our results with those of the prior studies on this important issue.
140
CLAIRE G. GILMORE AND GINETTE M. McMANUS
3. METHODOLOGY To study the effects of NAFTA on the behavior of the stock markets of its three members we use the methodology of cointegration analysis. Details of the methodology are discussed in the appendix. Cointegration tests allow us to determine whether stock prices of different national markets share a common long-run stochastic trend (i.e. have a long-term relationship), while allowing for the possibility of short-run divergence. The first step in the analysis is to test each index series for the presence of unit roots, which would show whether the series are non-stationary. Non-stationarity is a precondition for cointegration; additionally, all the series must be integrated of the same order. The Augmented Dickey-Fuller (ADF) test, an extension of the Dickey and Fuller (1981) method, is used. However, this test assumes that the errors are statistically independent and have a constant variance. To circumvent these limiting assumptions, Phillips and Perron (1988) developed a generalization of the Dickey-Fuller test, which is also applied here. Once the non-stationarity requirements are met, two basic approaches can be employed to determine whether the time series are cointegrated: the EngleGranger methodology (1987) and the Johansen (1988) procedure. Engle-Granger uses ordinary least-squares (OLS) regression to estimate the long-run equilibrium relationship and then tests the resulting error term for stationarity. Because of limitations in the Engle-Granger approach the Johansen test is employed instead. The latter determines the rank of the coefficient matrix of a vector autogression (VAR) of the series, with the rank indicating whether there is contegration, as well as the number of cointegrating relationships. If cointegration is found, vectorerror-correction (VEC) analysis can be used to model the long-term relationship between the North American equity markets. The VEC model will also provide information about short-run Granger causality between the markets. In addition, variance decomposition can be applied to shed further light on the nature of the system.
4. DATA The sample covers the period January 5, 1994, through December 31, 2002, and the data consist of daily closing equity price indices for the stock markets of Canada, Mexico, and the U.S. The TSE 300 index is used to represent the Canadian stock market, the IPC index for Mexico, and the S&P 500 index proxies the U.S. equity market. We convert the data to weekly (and monthly) observations using Wednesday figures to avoid problems of non-trading, non-synchronous,
The Impact of NAFTA on the Integration
141
Fig. 1. Standardized Stock Price Indexes of United States, Canada, and Mexico.
and bid/ask spreads in daily data (Hung & Cheung, 1995; Lo & MacKinlay, 1988). All data are in U.S. dollar terms and converted to natural logarithms (the tests were also run in local currency terms and produced essentially similar but weaker results). Figure 1 plots weekly stock price indexes for the three North American equity markets since the implementation of NAFTA in 1994. The weekly level series are standardized by subtracting the mean from each observation and then dividing by the standard deviation. Price movements in stock markets of North America appear to trend together over time, have similar turning points, and exhibit a relatively close co-movement since NAFTA began in January 1994. Table 1 provides the relevant descriptive statistics for the weekly returns (firstdifferences of stock prices) series for the 1994–2002 period. As expected, the Canadian and Mexican stock markets have the highest return volatility, with a weekly standard deviation of 5.55% and 5.47%, respectively. In comparison, the standard deviation for the U.S. is 2.39%. However, the average weekly return for both Canada and Mexico is lower than the mean return for the U.S. Table 2 shows the correlations for the full period as well as for two sub-periods of equal length. For the full period the correlations are strong, particularly for
142
CLAIRE G. GILMORE AND GINETTE M. McMANUS
Table 1. Summary Statistics of Weekly Index Returns Series. Statistic
Country
N Mean Median Maximum Minimum Standard deviation Skewness Kurtosis Jarque-Bera Probability
United States
Canada
Mexico
470 0.0014 0.0030 0.1018 −0.0904 0.0239 −0.2014 4.5321 49.0371 0.0000
470 0.0005 0.0017 0.0847 −0.1115 0.0555 −0.5115 4.7716 81.7824 0.0000
470 −0.0008 −0.0008 0.2047 −0.2555 0.0547 −0.2238 5.8981 168.0396 0.0000
Note: Index returns are estimated as the log-relative of weekly prices for January 5, 1994, to December 31, 2002, using the S&P 500 for the U.S., the TSE 300 for Canada, and the IPC for Mexico. The Jarque-Bera statistic tests the null hypothesis of a normal distribution and is distributed as a 2 with 2 degrees of freedom.
the U.S. and Canadian equity markets. Furthermore, the evidence indicates that correlations have increased in the latter sub-period. We report below the results of cointegration tests to assess whether the U.S., Canadian, and Mexican stock markets move together over the long-run.
Table 2. Correlation Coefficients of Weekly Stock Return Series. Market
United States
Canada
Mexico
Panel A: Entire period (January 5, 1994 to December 31, 2002) United States 1 Canada 0.7490 1 Mexico 0.4407 0.4651
1
Panel B: First sub-period (January 5, 1994 to July 1, 1998) United States 1 Canada 0.6788 1 Mexico 0.3539 0.3562
1
Panel C: Second sub-period (July 8, 1998 to December 31, 2002) United States 1 Canada 0.7732 1 Mexico 0.5506 0.5838
1
The Impact of NAFTA on the Integration
143
5. EMPIRICAL RESULTS Following a procedure suggested in Donaldo et al. (1990) for the ADF test when the form of the data-generating process is unknown, we started with the least restrictive model, including both intercept and trend. This approach uses F-statistics provided by Dickey and Fuller (1981) to test joint hypotheses on the coefficients of the various forms of the model. In each case the appropriate model contained an intercept but no trend term. Appropriate lag lengths were selected according to the Akaike information criterion (AIC). In addition, the Box-Ljung Q test was applied to the residuals to establish a lack of serial correlation in each case. F tests and t tests were also used to determine lag lengths and produced somewhat longer lag structures, but the qualitative results of the unit root tests were unchanged. Both the ADF and PP tests were applied to the levels and first differences of each series. The results using the AIC criterion are presented in Table 3. For the level series, Table 3 shows that the null hypothesis of a unit root cannot be rejected for any series at the 5% confidence level. All three of the first-differenced series reject the null hypothesis, indicating that they are stationary. Consequently, they are integrated of order one, that is, I(1). The Johansen (1988) procedure was then applied to determine whether any of the three North America equity markets are pairwise cointegrated. Since the stock market indexes exhibit a trend, a constant was included in the models. Lag structures were chosen according to the AIC and the minimum lag length which provides white-noise residuals; in each case five lags were used. Both the trace test and the maximum eigenvalue test produced evidence of bilateral cointegration between the U.S. and Mexico. However, there is no evidence of bilateral cointegration between the U.S. and Canadian equity markets or between the Mexican and Canadian stock markets (Table 4).
Table 3. Unit Root Tests for Weekly Stock Indices. Country Index
United States (1) Canada (2) Mexico (2)
Index Level
First Differences
ADF
PP
ADF
PP
−1.676 −1.207 −1.900
−1.752 −1.229 −1.900
−9.653∗ −8.207∗ −8.449∗
23.789∗ 14.381∗ 15.249∗
Note: Unit root tests are conducted using the Augmented Dickey-Fuller (ADF) and the Phillips-Perron (PP) tests. Lag lengths, in parentheses, were chosen according to the Akaike information criterion (AIC). The critical values are based on MacKinnon (1991); an asterisk indicates significant at the 5% level or better.
144
CLAIRE G. GILMORE AND GINETTE M. McMANUS
Table 4. Bilateral Johansen Cointegration Tests Results. Hypotheses H0
HA
5% Critical Values
r=0 r≤1
r=1 r=2
15.67 9.24
r=0 r≤1
r≥1 r≥2
19.96 9.24
Conclusion
United StatesCanada (5)
United StatesMexico (5)
Canada-Mexico (5)
Maximum eigenvalue test 9.04 18.15∗ 5.70 8.86
12.36 4.78
14.74 5.70
Trace test 27.01∗ 8.86
17.14 4.78
r=0
r=1
r=0
Note: The Johansen (1988) test is used to test bivariate cointegration between the North American equity markets. The r denotes the maximum number of cointegrating vectors. Two forms of the Johansen test were used. The 5% critical values are provided by Osterwald-Lenum (1992). The results indicate cointegration for the U.S. and Mexican markets.
Although the U.S. and Canadian markets as well as the Mexican and Canadian markets are not pairwise cointegrated, it is possible that these markets, as a group, could be cointegrated. This is done to take into account the possibility of omitted variable problems and untested interaction effects in the bivariate case, which can cause bivariate and multivariate cointegration tests to differ. Therefore, a multilateral Johansen test was applied with a lag length of seven according to the AIC and the minimum lag length which provides white-noise residuals. The results are reported in Table 5. The hypothesis of zero cointegrating vectors against the alternative of one or more cointegrating vectors is rejected at the 5% level of Table 5. Multilateral Johansen Cointegration Test Results. Hypotheses H0 r=0 r≤1 r≤2 Conclusion
Maximum Eigenvalue Test
Trace Test
27.89∗ 8.11 6.24
42.24∗ 14.36 6.24
r=1
r=1
HA r=1 r=2 r=3
Note: The Johansen (1988) test is used to test multilateral cointegration of the three North American stock markets as a group. The r denotes the maximum number of cointegrating vectors. Two forms of the Johansen test were used, the trace test and the maximum eigenvalue test: trace (r) = −T ln(1 − i ) and max (r, r + 1) = −T ln(1 − r+1 ). Seven (7) lags were used. The 5% critical values provided by Osterwald-Lenum (1992) indicate that there is one cointegrating vector.
The Impact of NAFTA on the Integration
145
significance. The hypothesis of at most one cointegrating vector is not rejected. This implies that there are two common stochastic trends, indicating a degree of capital market integration. The cointegrating vector, whose coefficients are normalized on the U.S. market, is Z = 1S&P + 27.011IPC − 3.55TSE − 149.73. We also performed bilateral and multilateral cointegration tests of U.S., Canada, and Mexico equity markets using monthly data. The empirical results, not reported to conserve space but available upon request, are less consistent and offer weaker evidence of multilateral cointegration than those based on weekly data. Since the evidence points toward cointegration of the three equity markets, the next step is to estimate a VEC model, with one cointegrating vector. The results are shown in Table 6. The error-correction model provides information about both the long-run and the short-run dynamics of the adjustment process between the national market indexes. The significance and size of the error-correction terms, referred to as Z(−1) in Table 6, capture the single-period response of the dependent variable to departures from the long-term equilibrium. The results indicate that all three markets contribute to the adjustment toward long-term equilibrium. These Z(−1) coefficients also represent the speed of adjustment of the markets in the long run. As they are small, they indicate a slow adjustment toward long-term equilibrium. Short-term interactions are shown by the coefficients on the lagged differenced terms. Overall, there is limited short-term interaction between the three markets in terms of t statistics. Results in Table 6 indicate that the U.S. market adjusts primarily to itself while Mexico is influenced from the U.S. as well. Surprisingly, we do not find significant short-term feedback from the U.S. or Mexico on the Table 6. Vector Error-Correction Model Results.
S&P (−1) S&P (−2) S&P (−3) S&P (−7) TSE (−1) TSE (−2) TSE (−3) TSE (−7) IPC (−1) IPC (−2) IPC (−3) IPC (−7) Z (−1)
S&P
TSE
IPC
−0.1983∗ (−2.748) 0.0989 (1.358) −0.0200 (−0.271) −0.2144∗ (−2.961) 0.0919 (1.367) −0.1475∗ (−2.196) 0.0869 (1.275) 0.0081 (0.120) 0.0266 (1.038) 0.0078 (0.333) −0.0227 (−0.973) 0.0140 (0.560) −0.0007∗ (−4.492)
−0.0761 (−0.961) 0.0789 (0.987) 0.0315 (0.388) −0.1349 (−1.698) 0.0764 (1.034) −0.0706 (−0.958) 0.0317 (0.423) −0.0150 (−0.203) 0.0152 (0.592) 0.0088 (0.344) 0.0070 (0.273) 0.0338 (1.320) −0.0005∗ (−2.595)
−0.2794 (−1.681) −0.4189∗ (−2.495) −0.2995 (−1.758) −0.6034∗ (−3.619) 0.1098 (0.709) 0.2942 (1.901) 0.1209 (0.770) 0.3041 (1.962) 0.0819 (1.524) 0.0875 (1.626) 0.1291∗ (2.399) 0.0527 (0.179) −0.0014∗ (−3.826)
Note: t -Statistics are shown in parentheses.
146
CLAIRE G. GILMORE AND GINETTE M. McMANUS
Table 7. Variance Decomposition of Forecast Errors Results. Weeks
Relative Variance of
United States
Mexico
Canada
1 13 26
United States
100.00 95.09 80.74
0.00 2.99 18.09
0.00 1.91 1.17
1 13 26
Mexico
19.08 7.27 6.98
80.92 84.81 81.87
0.00 7.92 11.16
1 13 26
Canada
56.62 47.75 45.14
2.22 1.63 1.22
41.16 50.62 53.64
Canadian market. In addition, insignificant results are also found in testing the lagged variables as a group using F tests. Variance decompositions are used to further explore the nature of the short-term relationship of the three NAFTA equity markets. Variance decompositions of the forecast errors provide a measure of the overall relative importance of the markets in producing fluctuations in stock returns in their own and other markets. Table 7 shows a decomposition of 1-, 13-, and 26-week ahead forecast error variances into fractions accounted for by innovations in the different markets, based on the ordering S&P 500, IPC, TSE 300. The price movements of the S&P 500 have a strong effect on the variance of the forecast errors in the Canadian and Mexican markets but not the reverse. Specifically, the forecast errors of S&P 500 returns are accounted for mostly by U.S. innovations and are not explained substantially by innovations in the other markets. At 26 weeks the U.S. market accounts for 80.74% of its own variance and explains 45.15% of the TSE 300 index variance and 6.98% of the IPC variance. Neither of the other markets explains a sizeable percentage of the S&P 500 variance. It is important to note that the ordering of the variables does have an impact on the results, particularly if the error terms are highly correlated, as is the case here. Consequently, we also tested alternative orderings. They indicate a lesser effect of the S&P 500 on the variance of the forecast errors for the TSE 300 and the IPC. However, the relatively limited impact of the other markets on the S&P 500 variance is not altered.
6. CONCLUSIONS Given the lack of solid evidence in the literature on the impact of NAFTA on the interrelationships between the equity markets of its members, this study examines
The Impact of NAFTA on the Integration
147
whether the implementation of NAFTA has led to more integrated equity markets. Our results, based on weekly data over the 1994–2002 period, indicate that the stock markets of the U.S., Canada, and Mexico are cointegrated. Our results contradict those of Ewing et al. (1999) but support those of Clinebell et al. (1998) and Darrat and Zhong (2001). While differences in findings may be attributed to different time periods and frequency of observations, our study with a much longer time period provides evidence that NAFTA has indeed had an impact on the integration of its members’ equity markets. Investors, corporate managers, and policy makers ought to be very interested in the consequences of the integration between the North American countries in the post-NAFTA period. First, cointegrated equity markets imply reduced long-term diversification benefits available to market participants. More specifically, shortterm correlations overstate the benefits of diversification for long-term investors when there is cointegration. Second, cointegrated equity markets imply that corporations seeking external financing in North American equity markets have access to capital on equal terms but face an increased competition for financing. Finally, cointegrated equity markets imply some predictability in returns, which may not be compatible with the efficient market hypothesis (EMH), according to Granger (1986). However, if market efficiency is viewed as implying absence of arbitrage opportunities, then cointegrated equity markets does not necessarily mean market inefficiency (see Fama, 1991). Whether NAFTA has affected the efficiency of the North American equity markets is, however, a matter of further empirical research since evidence of predictability in returns may be statistically significant but economically insignificant. NAFTA has been fully implemented only since the year 2000 and an extended study may provide additional insights into the consequences of cointegrated equity markets.
REFERENCES Allen, D. E., & MacDonald, G. (1995). The long-run gains from international equity diversification: Australian evidence from cointegration tests. Applied Financial Economics, 5, 33–42. Arshanapalli, B., & Doukas, J. (1993). International stock market linkages: Evidence from the preand post-October 1987 period. Journal of Banking and Finance, 17, 193–208. Blackman, S. C., Holden, K., & Thomas, W. A. (1994). Long-term relationships between international share prices. Applied Financial Economics, 4, 297–304. Byers, J. D., & Peel, D. A. (1993). Some evidence on the interdependence of national stock markets and the gains from international portfolio diversification. Applied Financial Economics, 3, 239–242. Chan, K. C., Gup, B. E., & Pan, M. S. (1992). An empirical analysis of stock prices in major Asian markets and the U.S. Financial Review, 27, 289–307. Chan, K. C., Gup, B. E., & Pan, M. S. (1997). International stock market efficiency and integration: A study of eighteen nations. Journal of Business Finance and Accounting, 24, 803–813.
148
CLAIRE G. GILMORE AND GINETTE M. McMANUS
Chaudhuri, K. (1997a). Cointegration, error correction, and Granger Causality: An application with Latin American stock markets. Applied Economics and Literature, 4, 469–471. Chaudhuri, K. (1997b). Stock returns in emerging markets: A common trend analysis. Applied Economics and Literature, 4, 105–108. Choudhry, T. (1994). Stochastic trends and stock prices: An international inquiry. Applied Financial Economics, 4, 383–390. Choudhry, T. (1997). Stochastic trends and stock prices: Evidence from Latin American markets. Journal of Macroeconomics, 19, 285–304. Chowdhury, A. R. (1994). Stock market interdependencies: Evidence from the Asian NIEs. Journal of Macroeconomics, 16, 629–651. Clinebell, J. M., Kahl, D. R., & Stevens, J. L. (1998). Cointegration and Granger Causality relationships between the United States, Canada, and Mexico before and after the North American free trade agreement. Unpublished manuscript. Darrat, A. F., & Zhong, M. (2001). Equity market integration and multinational trade accords: The case of NAFTA. Unpublished manuscript. DeFusco, R. A., Geppert, J. M., & Tsetsekos, G. P. (1996). Long-run diversification potential in emerging stock markets. Financial Review, 31, 343–363. Dickey, D. A., & Fuller, W. A. (1981). Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica, 49, 1057–1072. Donaldo, J., Jenkinson, T., & Sosvila-Rivero, S. (1990). Cointegration and unit roots. Journal of Economic Surveys, 4, 249–273. Engle, R. F., & Granger, C. W. J. (1987). Cointegration and error correction: Representation, estimation, and testing. Econometrica, 55, 2511–2576. Eun, C. S., & Shim, S. (1989). International transmission of stock market movements. Journal of Financial and Quantitative Analysis, 24, 241–256. Ewing, B. T., Payne, J. E., & Sowell, C. (1999). NAFTA and North American stock market linkages: An empirical note. North American Journal of Economics and Finance, 10, 443–451. Fama, E. (1991). Efficient capital markets: II. Journal of Finance, 46, 1575–1617. Granger, C. W. J. (1986). Developments in the study of cointegrated economic variables. Oxford Bulletin of Economics and Statistics, 48, 213–228. Hassan, M. K., & Naka, A. (1996). Short-run and long-run dynamic linkages among international stock markets. Global Finance Journal, 5, 387–405. Hung, B., & Cheung, Y. (1995). Interdependence of Asian emerging equity markets. Journal of Business, Finance and Accounting, 22, 281–288. Jeon, B. N., & Von Furstenberg, G. M. (1990). Growing international co-movements in stock price indexes. Quarterly Review of Economics and Business, 30, 15–30. Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control, 12, 231–234. Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on cointegration: With applications to the demand for money. Oxford Bulletin of Economics and Statistics, 52, 169–210. Kanas, A. (1998). Linkages between the U.S. and European equity markets: Further evidence from cointegration tests. Applied Financial Economics, 8, 607–614. Karolyi, G. A., & Stulz, R. M. (1996). Why do markets move together: An investigation of U.S.-Japan stock return co-movements. Journal of Finance, 5, 951–986. Kasa, K. (1992). Common stochastic trends in international stock markets. Journal of Monetary Economics, 29, 95–124.
The Impact of NAFTA on the Integration
149
Kwan, A., Sim, A., & Cotsomitis, J. (1995). The causal relationships between equity indices on world exchanges. Applied Economics, 27, 467–471. Lo, A. W., & MacKinlay, A. C. (1988). Stock market prices do follow random walks: Evidence from a simple specification test. Review of Financial Studies, 1, 41–66. Monadjemi, M. S., & Perry, L. (1996). Share price movements: A study of four OECD countries. Applied Economics and Literature, 3, 135–137. Newey, W. K., & West, K. C. (1987). A simple positive definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica, 55, 703–708. Osterwald-Lenum, M. (1992). A note with quantiles of the asymptotic distribution of the maximum likelihood cointegration rank test statistic. Oxford Bulletin of Economics and Statistics, 54, 461–472. Phillips, P., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75, 335–346. Richards, A. J. (1995). Co-movements in national stock market returns: Evidence of predictability, but no integration. Journal of Monetary Economics, 36, 631–654. Rogers, J. H. (1994). Entry barriers and price movements between major emerging stock markets. Journal of Macroeconomics, 16, 221–241. Smith, K. L., Brocato, J., & Rogers, J. E. (1993). Regularities in the data between major equity markets: Evidence from Granger Causality tests. Applied Financial Economics, 3, 55–60.
APPENDIX This appendix presents some of the mathematical details of cointegration theory. If there are two variables, xt and yt , which are both nonstationary in levels but stationary in first differences, then xt and yt are integrated of order one, I(1), and their linear combination having the form z t = x t − ay t
(1)
is generally also I(1). However, if there is an (a) such that zt is integrated of order zero, I(0), the linear combination of xt and yt is stationary and the two variables are said to be cointegrated (Engle & Granger, 1987). If two variables are cointegrated, there is an underlying long-run relationship between them. In the short run the series may drift apart, but if they are cointegrated, they will move toward long-run equilibrium through the error-correction mechanism. The first step in the analysis is to test each series for the presence of unit roots. This can be done by means of the Augmented Dickey Fuller (ADF) test, an extension of the Dickey and Fuller (1981) method. The ADF test uses a regression of the first differences of the series against the series lagged once, and lagged difference terms, with optional constant and time trend terms: y t = a 0 + a 1 t + ␥y t−1 + b i y t−i+1 + e t
(2)
150
CLAIRE G. GILMORE AND GINETTE M. McMANUS
In the equation is the first difference operator, a0 is an intercept, a1 t is a linear time trend, et is an error term, and i is the number of lagged first-differenced terms such that et is white noise. The test for a unit root has the null hypothesis that ␥ = 0. If the coefficient is significantly different from zero, the hypothesis that yt contains a unit root is rejected. If the test on the level series fails to reject, the ADF procedure is then applied to the first-differences of the series. Rejection leads to the conclusion that the series is integrated of order one, I(1). A limitation of the Dickey-Fuller test is that it assumes that the errors are statistically independent and have a constant variance. In 1988, Phillips and Perron (PP) generalized the ADF test: T y t = b 0 + b 1 y t−1 + b 2 t − + t (3) 2 T is the number of observations and the disturbance term t is such that E(t ) = 0, but there is no requirement that the disturbance term is serially uncorrelated or homogeneous. The equation is estimated by ordinary least squares and the t-statistic of the b1 coefficient is corrected for serial correlation in t using the Newey-West (1987) procedure for adjusting the standard errors. The Johansen (1988) approach to testing for cointegration relies on the relationship between the rank of a matrix and its characteristic roots, or eigenvalues. Let Xt be a vector of n time series variables, each of which is integrated of order (1), and assume that Xt can be modeled by a vector autoregression (VAR): X t = A 1 X t−1 + · · · + A p X t−p + t
(4)
x t = x t−1 + Ŵx t−i + t
(5)
Rewrite the VAR as
where = A i − I, Ŵi = −A i . If the coefficient matrix has reduced rank r < k, there exist k × r matrices ␣ and  each with rank r such that = ␣′ and ′ xt is stationary. The number of cointegrating relations is given by r, and each column of  is a cointegrating vector. Three cases are possible. First, if is of full rank, all elements of X are stationary, and none of the series has a unit root. Second, if the rank of = 0, there are no combinations which are stationary and there are no cointegrating vectors. Third, if the rank of is r such that 0 < r < k, then the X variables are cointegrated and there exist r cointegrating vectors. Equation (8) can be modified to allow for an intercept and a linear trend. The number of distinct cointegrating vectors can be obtained by determining the significance of the characteristic roots of . To identify the number of characteristic roots that are not different from unity we use two statistics, the trace test and the
The Impact of NAFTA on the Integration
151
maximum eigenvalue test: trace (r) = −T ln(1 − i )
(6)
max (r, r + 1) = −T ln(1 − r+1 )
(7)
and
where i = the estimated values of the characteristic roots (eigenvalues) obtained from the estimated matrix, r is the number of cointegrating vectors, and T = the number of usable observations. The trace test evaluates the null hypothesis that the number of distinct cointegrating vectors is less than or equal to r against a general alternative. The maximum eigenvalue test examines the number of cointegrating vectors vs. that number plus one. If the variables in Xt are not cointegrated, the rank of is zero and all the characteristic roots are zero. Since ln(1) = 0, each of the expressions ln(1 − i ) will equal zero in that case. Critical values for the test are provided by Johansen and Juselius (1990) and by Osterwald-Lenum (1992).
A SINGLE CURRENCY FOR NAFTA? Patrick M. Crowley ABSTRACT This paper attempts to evaluate whether the set of NAFTA countries (the U.S., Canada and Mexico) should adopt the same currency. The theoretical basis for the paper is the optimal currency area theory which suggests that countries or regions that experience similar business cycles can gain advantages in adopting the same currency. The statistical methodology used in the paper to evaluate whether states or provinces have similar business cycle correlations is model-based cluster analysis, a recently-developed method to group data in the applied statistics literature.
1. INTRODUCTION The launch of the euro currency was officially completed as of January 1, 2002, marking the final stage of monetary union in the European Union (EU). European Economic and Monetary Union (EMU) was historically unprecedented, in that it adopted a time-tabled process, included strict economic convergence criteria, and is not counterbalanced at the European Union (EU) level with a government that implements pan-EU fiscal policy. Economic theory, following the pioneering work of Mundell (1961) and others, suggests that regions that are subject to similar economic cycles and shocks could be subject to the same monetary policy and so use the same currency – hence the notion of optimal currency areas (OCAs). The drafters of the Maastricht Treaty clearly did not have the optimal currency area literature in mind when they specified five economic criteria for membership of EMU which related to inflation North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 153–173 Published by Elsevier Ltd. ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10009-0
153
154
PATRICK M. CROWLEY
rates, interest rates, budget deficits, public debt, and maintenance of exchange rates in the Exchange Rate Mechanism (ERM) of the European Monetary System (EMS). Much ink has been spilled over both optimal currency areas (OCA) and the economic convergence criteria (see Buiter, Corsetti & Roubini, 1993; Crowley, 1996; Eichengreen, 1993; Fratianni, von Hagen & Waller 1992), but little work has been done on the applicability of both the OCA theory and economic convergence criteria to other regional integration agreements. In Canada, given the weak Canadian dollar, a vociferous debate has taken place over the last couple of years on North American Monetary Union (NAMU). The debate first focused on whether the North American Free Trade Agreement (NAFTA) might give way to a common currency for the trading bloc, in a similar way in which the single market in the EU justified a single currency. The likelihood of the adoption of an entirely new currency appeared to most Canadians as an anathema, as the U.S. would be unlikely to substitute the U.S. currency for something completely new, and the lack of institutional development in NAFTA would also dictate against this option. The President of Mexico, Vicente Fox, also recently brought up the issue of a single market and single currency for NAFTA, and so it now appears that Mexico also seems keen to progress to a deeper level of integration with the U.S. This paper seeks to evaluate whether NAFTA in its current form forms an optimal currency area, in the sense that: (i) synchronicity of business cycles is achieved; and (ii) states and provinces have a similar experience with movements in inflation and unemployment. The paper is divided into six sections. Section 2 evaluates the theoretical and empirical literature on optimal currency areas and business cycle synchronicity, while Section 3 reviews the methodology used in this paper. Section 4 outlines the data and data preparation. Section 5 presents the empirical results of the analysis of business cycle correlations and other associated variables using model-based cluster analysis. Section 6 concludes.
2. OCAS AND BUSINESS CYCLES 2.1. The Optimal Currency Area Literature Two seminal papers on optimal currency areas (McKinnon, 1963; Mundell, 1961) outlined the conditions under which several administrative jurisdictions might be suitable to be subject to the same monetary policy. Further refinements of this approach were subsequently made by Kenen (1969). Bayoumi (1994) also offered a formal model of optimal currency areas (OCAs) with microeconomic foundations to underscore Mundell’s original thesis. The conditions for an OCA
A Single Currency for NAFTA
155
are that members of the currency union should, for the most part, experience symmetric shocks and that economic cycles should be synchronous. If countries experience asymmetric shocks or have asynchronous business cycles then the costs of being subject to a single monetary policy may be significant, and may outweigh the costs. To offset asymmetric shocks or asymmetric business cycles, then certain currency area characteristics may ameliorate costs, notably: (i) a significant degree of labour mobility; (ii) fiscal transfers through a “federal” level of government; and (iii) flexible wages and prices. The empirical time-series literature on OCAs can be divided into three strands – a strand that uses basic regional data (from a sub-national level) to evaluate whether countries use exchange rates to offset shocks, with the implication that similar exchange rate volatilities would imply similar shock magnitudes, while at the same time evaluating whether participants possess the three offsetting characterstics (De Grauwe & Vanhaverbeke, 1993); a strand that uses structural vector autoregression (SVAR) time series methodology (Blanchard & Quah, 1989) to identify demand and supply shocks (see Bayoumi & Eichengreen, 1994a for many parts of the world, Bayoumi & Eichengreen, 1994b for North America and De Serres & Lalonde, 1994; Lalonde & St-Amant, 1993 for Canada) and then looks at the correlation of these shocks across countries or regions. Lastly, another strand of the literature evaluates the synchronicity of business cycles across prospective currency union members (Artis & Zhang, 1997a; Baxter & Stockman, 1989). The first strand of OCA empirical research has been criticised for being largely descriptive, while the second (SVAR) methodology has been criticised (Buiter, 1998) for being arbitrary in terms of the restrictions that are required for identification of monetary and real shocks (usually the assumption that shocks that are neutral in the long run are monetary shocks). The third strand of research also responds to another criticism of the VAR methodology: that a shock approach ignores long run business cycle synchronicity – the synchronicity approach compares cyclical components in GDP and then uses correlations in business cycles to draw out implications about suitability as constituents of an OCA. The obvious drawback here is that this approach completely ignores the incidence of temporary shocks and does not consider the ability of exchange rates to also compensate for shocks. A recent development in the OCA literature has been recognition that ex-ante evaluations of which countries constitute an OCA might ignore the Lucas critique, in that new members of an OCA might (a) modify policy to be better suited to an OCA (Tavlas, 1993) or (b) be more suited to being in an OCA ex-post (Frankel & Rose, 1997). The latter approach takes into consideration factors which usually do not appear in the ex-ante OCA approach, such as trade intensity, real interest rate cycles, and fiscal policy coordination.1
156
PATRICK M. CROWLEY
2.2. Synchronicity of Business Cycles Following Gerlach (1988) and Baxter and Stockman (1989) on business cycle correlations, there has been considerable research devoted to the propagation of business cycles, and the existence of a world business cycle in the pre- and post-Bretton Woods periods. Recent research on business cycles has focussed on the effects of trade in propogating business cycles (Imbs, 1999), comparisons of business cycles in monetary unions (Wynne & Koo, 2000), and on new measures of co-movement (Croux, Forni & Reichlin, 1999) of output data for different regions or countries. Artis and Zhang (1997a) explored the idea of group-specific business cycles after the inception of the ERM of the EMS in 1979, positing a distinctly European business cycle.2 In this study, cyclical components of industrial production were obtained using several de-trending methods,3 and then the cross-correlations of the cyclical components of these series with the U.S. series and the German series were calculated. A European business cycle was confirmed, but the cycle was confined to members of the ERM of the EMS, as might have been expected. Here we employ the same similar methodology, with two differences. First, in the European context, Artis and Zhang (1997a) justified using the cyclical component of the German series as a basis for evaluating whether a European business cycle existed, predicated on other research which clearly showed Germany to be the largest and most influential economy in the EU, and the Bundesbank to be a “leader” in terms of the setting of monetary policy in the ERM of the EMS. In the NAFTA context, as the U.S. is already a country with a single monetary authority, the U.S. national aggregates are used as the appropriate “target” variables for the purposes of calculating cross-correlations, both for individual U.S. states, as well as Canadian provinces and Mexico. Second, as research on Canada (Crowley, 2000) showed that some lagged effects on regional GDP took place, also a lagged business cycle correlation is calculated. The analysis was undertaken by using a Hodrick-Prescott filter; a HodrickPrescott filter is based on minimizing the following expression with respect to gt : N N−1 2 2 [(g t+1 − g t ) − (g t − g t−1 )] (1) (y t − g t ) + min t=1
t=2
where yt is the raw data, gt is the growth component, and so (y t − g t ) is the cyclical component. is a dampening parameter whose value is extremely sensitive to the data being used. Following recent research by Pedersen (2000), which calculated optimal values of using the estimated spectral shape of real U.S. GDP in both monthly, quarterly, and annual series, a value of = 40 was used.
A Single Currency for NAFTA
157
2.3. Data and Data Sources To use model-based cluster analysis for NAFTA following the optimal currency area theory, data is needed that corroborates the degree of synchronicity in business cycles plus the degree in flexibility in labor markets. State and provincial data in the U.S. and Canada, respectively, is available on a limited basis, so variables are selected or constructed to attempt to best characterise what the optimal currency area literature suggests should be important in subjecting regions to a common monetary policy. Mexico posed a special challenge, as not only did language difficulties impede the data collection process, but also Inegi, the national statistical agency, has only recently started to regularize data collection by state for certain economic indicators.4 Partly because of these problems, only national Mexican data is used, which probably accords more closely to the current size of the GDP of Mexico when compared to both U.S. states and Canadian provinces. Appendix B documents all data sources. Here we use both the contemporaneous and lagged cyclical business cycle cross-correlations from Section 2. In addition to these two variables, we include: (i) unemployment rates – this is used to proxy both the degree of labor market flexibility compared to the U.S., plus it can also be thought of as a measure of labor migration inertia. The OCA literature suggests that countries with more flexible labor markets are better suited to join single currency areas; (ii) inflation rates – this is the CPI by state or province and is used to capture commonality in consumer price inflation experience. Unfortunately, no other variables were available consistently across the NAFTA region – indeed the OCA literature would imply that a measure of openness and a measure of labor mobility should also be included. The data used to construct the variables above were sourced annually from national statistical agencies in all three countries. Each annual series was correlated against its U.S. national counterpart, using the available data. Appendix C gives details of the periods over which correlations were calculated. The correlations clearly show a wide variation between U.S. states, and also highlight how different Mexico appears to be from the rest of the U.S., with the exception of Hawaii (HI). Cluster analysis should also be expected to highlight some of these differences and determine the number of groupings there are among states and provinces in the NAFTA bloc. Even at this stage in the analysis, it appears apparent that Mexico has had a rather different experience from the rest of North America over the period in question, so might be expected to fall into a category (cluster) of its own.
158
PATRICK M. CROWLEY
For preparation for cluster analysis, all correlation coefficients were normalized so that their range is no longer (−1, +1). This is to ensure equal weighting to the four variables. Appendix D tabulates the means and standard deviations for correlations for each variable.
3. METHODOLOGY 3.1. Cluster Analysis In economics cluster analysis has been applied to EU data by several authors, notably Jacquemin and Sapir (1995) and Artis and Zhang (1997b, 1998a, b), with interesting results. The cluster analysis done on the EU has largely corroborated the evidence on suitability for membership of EMU gained from the aforementioned empirical methods used in the OCA literature. The methodology has also started to infiltrate into the economics profession in North America, with Crone (1999), Galbraith and Jiaquing (1999), Honohan (2000) and Boreiko (2002) using the technique – also Maharaj and Inder (1999) is another application using cluster analysis to forecast time series in economics. In other disciplines, cluster analysis is frequently used – applications range from astrophysics (Mukerjee et al., 1998) to microbiology (van Ooyen, 2001). Cluster analysis aims to determine the intrinsic structure of data when no information other than the observed values is available – the data is to be partitioned into meaningful subgroups. This approach should be put in contrast with discriminant analysis, in which known groupings of some observations are used to categorize others and infer the structure of the data as a whole. Clustering methods range from those that are largely heuristic to more formal procedures based on statistical models, and they are hierarchical or based on allocating observations among tentative clusters (such as k-means clustering). Hierarchical methods proceed by stages, partitioning or combining the data at each stage. Hierarchical methods fall into two categories: “agglomerative” and “divisive” – with agglomerative denoting the merging of clusters at each stage and divisive denoting the splitting of clusters at each stage – in most cases agglomerative and divisive methods give similar clusterings. At each stage some criterion is optimized and used to determine which clusters should be combined or split – most methods use single link (nearest neighbor), complete link (farthest neighbor) or sum of squares. In model methods, however, usually a maximum likelihood based on specific distributional assumptions is used to merge or divide groups. Useful references for these heuristic clustering methods are Anderberg (1993), Kaufman and Rousseeuw (1990) and Hartigan (1975).
A Single Currency for NAFTA
159
Unfortunately, although these clustering methods are appealing, none of them addresses the issue of how many clusters there should be. Various strategies have been put forward to choose the number of clusters, but up until recently none of these methods has been satisfactory from a computational point of view, or from a methodological point of view (see Bock, 1996 for a survey of this issue and related research). The alternative that has been presented by Fraley and Raftery (1998, 1999) is computationally relatively straightforward, and is intuitively appealing given the emphasis on optimization and statistical technique in economics.
3.2. Model-Based Cluster Analysis In probability based clustering, each observation is assumed to be generated by a mixture of underlying probability distributions where each component in the mixture represents a different cluster. Given a set of observations x = (x 1 , . . . , x n ), then the density of an observation xi from the kth component in a total number of G components, is f k (x i |k ), where k are the parameters. In most cases, f k (x i |k ) is assumed to be multivariate normal (Gaussian), so in this instance the parameters k consist of a mean vector k and a covariance matrix k . The clusters will then be ellipsoidal, with center at k , and the covariance matrix will determine the other characteristics. The mixture likelihood approach then maximizes the criterion: ℓM (1 , . . . , G : 1 , . . . , G |x) =
n G
k f k (x i |k )
(2)
i=1 k=1
where k is the probability that an observation belongs to the kth component. Banfield and Raftery (1993) developed a model-based framework for clustering by expressing the covariance matrix in terms of its eigenvalue decomposition, which is of the form = k D k A k D Tk (3) k
where Dk is the orthogonal matrix of eigenvectors, Ak is a diagonal matrix where the elements of the diagonals are proportional to the eigenvalues of k , and k is a scalar. This leads to a geometric interpretation of the ellipsoidal clusters – Dk determines the orientation, Ak determines the shape of the density contours and k specifies the volume. These characteristics can then be allowed to vary between clusters, or constrained to be the same for all clusters. This approach actually subsumes many previous approaches at model-based clustering – more details can
160
PATRICK M. CROWLEY
Table 1. Parameterizations of the Covariance Matrix. Model
ID
Distribution
Volume
Shape
Orientation
I k I DADT k D k A k D Tk D k AD Tk k D k AD Tk
EI VI EEE VVV EEV VEV
Spherical Spherical Ellipsoidal Ellipsoidal Ellipsoidal Ellipsoidal
Equal Variable Equal Variable Equal Variable
Equal Equal Equal Variable Equal Equal
NA NA Equal Variable Variable Variable
Source: Banfield and Raftery (1993).
be located in Fraley and Raftery (1998). The range of models used here is limited, given the limitations of the SPLUS software and the library MCLUST which was used for estimation. In the approach taken here, the parameterizations of the covariance matrix are detailed in Table 1. Given the different model parameterizations above, agglomerative hierarchical clustering can be used by merging clusters so as to maximize the resulting likelihood as specified in Eq. (2) above. 3.3. Clustering Algorithms The algorithm used for maximizing the likelihood function here is the EM (Expectation-Maximization) algorithm (see McLachlan & Krishnan, 1997). EM iterates between an “E” step, which computes a matrix z such that zik is an estimate of the conditional probability that observation i belongs to group k given the current parameter estimates, and an “M” step, which computes maximum likelihood parameter estimates given z. In the limit, under certain conditions the parameters usually converge to the maximum likelihood values for the Gaussian mixture model and the sums of the columns of z converge to n times the mixing proportions k , where n is the number of observations. The EM algorithm is not without its problems though. Fraley and Raftery (1998) detail several problems notably: (i) a slow rate of convergence; (ii) the number of conditional probabilities associated with each observation equals the number of components in the mixture, so that the EM algorithm may not be suitable for large datasets; and (iii) when the covariance matrix becomes singular or nearly singular (otherwise known as “ill-conditioned”) the EM algorithm breaks down. The latter problem was an issue in this study – usually relates to clusters which only contain a few observations or if the observations contained are co-linear, and in this study the former is the suspected problem.
A Single Currency for NAFTA
161
3.4. Model Selection The mixture model approach allows the use of approximate Bayes factors to compare models (see Kass & Raftery, 1995). The Bayes factor is the posterior odds for one model against the other assuming neither is favored a priori. With the EM algorithm twice the log Bayes factor is used to determine the number of clusters in hierarchical clustering based on the mixture likelihood – this measure is also known as the Bayesian Information Criterion (BIC) and is specified as: 2log p(x|M) + const ≈ 2ℓM log(n) − m M log(n) ≡ BIC
(4)
where p(x|M) is the likelihood of the data for the model M, ℓM (x|) is the maximized mixture log likelihood for the model and mM is the number of independent parameters to be estimated in the model. The larger the value of the BIC, the stronger the evidence for the model.5 A standard convention for calibrating BIC differences is that differences of less than 2 correspond to weak evidence, differences between 2 and 6 to positive evidence, differences between 6 and 10 to strong evidence, and differences greater than 10 to very strong evidence.
3.5. Clustering Strategy The general strategy adopted here is similar to that of Fraley and Raftery (1998) and is detailed for SPLUS library MCLUST in Fraley and Raftery (1999). The steps of strategy are as follows: (i) Determine a maximum number of clusters to consider, and a set of candidate parameterizations of the model to use. (ii) Use agglomerative hierarchical clustering for the unconstrained Gaussian model, to obtain classifications for up to M groups. (iii) Do EM for each parameterization and each number of clusters, starting with the classification from hierarchical clustering. (iv) Compute the BIC for the one cluster model for each parameterization and for the mixture likelihood with optimal parameters from EM for other clusters. (v) Plot the BIC – this should hopefully indicate a local maximum and a specific model. This strategy was followed for the research presented here.
162
PATRICK M. CROWLEY
4. EMPIRICAL RESULTS As Section 4 (e) detailed, cluster analysis methodology is adopted here. In all cases the EM algorithm was initialized using hierarchical clustering using the unconstrained model (EI) detailed in Table 1 above. The maximum number of clusters was chosen to be M = 10. From this point BIC values were calculated from an initial parameterization for all other possible models presented in Table 1. Table 2 gives the BIC for each of the candidate models for each of the 10 cluster groups specified. Some BIC estimates were not available, as the covariance matrix associated with one or more of the mixture components is ill-conditioned, so that the log likelihood and hence the BIC cannot be computed. Figure 1 below then graphs the BIC plots by numbers of clusters. Figure 1 shows that the best BIC value is obtained for four clusters under EEV (ellipsoidal, constant shape, equal volume), with the second highest value for the same model with six clusters. The difference between the two BIC values was far greater than 10, so this indicates very strong evidence in favour of four clusters of data points in the sample. The third highest BIC value was obtained for model VEV (ellipsoidal, constant shape, varying volume) with five clusters, and using the rule of thumb for BIC differences the difference between the two was less than two, indicating only weak evidence for EEV with six clusters. Now that the optimum number of clusters has been determined, model based clustering can be implemented with four clusters under the EEV model using the EM algorithm. The results of the clustering are presented in Table 3 and graphically in Fig. 2. The uncertainty in the classification associated with the EM algorithm gives a measure of the quality of the classification. In this case, using examples given in Fraley and Raftery (1999), the majority of observations are well classified. Uncertain classifications result when groups intersect, giving overlapping regions and therefore some uncertainty as to which cluster to allocate to. The clusters illustrated in Fig. 2(a) and (b) clearly show the clustering for each grouping, with some of the clusters intersecting given the scatterplots are only in two dimensions. Using the coding detailed in Appendix A, the table shows which provinces or states fall into each cluster, with the numbering indicating the order in which each cluster formed according to the agglomeration procedure. The results clearly imply several empirical facts, given the data and time periods used in the analysis: (i) the U.S. largely constitutes an OCA, although peripheral states may not be asynchronous with the majority of states;
A Single Currency for NAFTA
Table 2. BIC Values by Number of Clusters Using the EM Algorithm. ID EI VI EEE VVV EEV VEV
1
2
3
4
5
6
7
8
9
10
−712.99 −712.99 −662.68 −662.68 −662.68 −662.68
−690.29 −638.94 −658.59 −627.53 −612.61 −604.01
−604.71 −599.72 −586.43 NA −584.05 −601.92
−612.23 −582.31 −637.44 NA −505.93 −561.34
−625.55 −577.25 −656.46 NA −559.27 −536.2
–591.49 −561.52 −659 NA −534.64 −581.2
−598.13 −566.09 −682.66 NA −542.65 −602.67
–588.44 NA −562.15 NA −553.32 NA
−595.52 NA −599.4 NA −557.8 NA
−576.14 NA −551.92 NA −596.46 NA
Note: NA = ill-conditioned matrix. Initialized using the EI model.
163
164
PATRICK M. CROWLEY
Fig. 1. BIC by Numbers of Clusters. Key: 1 = EI, 2 = VI, 3 = EEE, 4 = VVV, 5 = EEV, 6 = VEV.
(ii) Canada is clearly not an OCA (these results corroborate those in Crowley (2000) which found, using heuristic methods and more data, that Canada can be split into three clusters), as there appear to be two distinct groups of provinces, essentially the maritimes and the rest of the country. (iii) Mexico clearly forms a separate cluster along with Hawaii and Wyoming, so therefore does not form an OCA with the USA. It is interesting to note though that Mexico is not classified in a cluster by itself. Table 3. Cluster Membership Using EEV Model and EM Algorithm. Cluster
Provinces or States
1 2 3 4
AB, BC, MAN, PEI, PQ, SASK, FL, WA NB, NF, NS, ONT, AK, CA, DE Rest of USA HI, WY, MEX
Uncertainty (Quantiles) 0% 0
25% 0
50% 1.36E-12
75% 1.8E-5
100% 0.149
A Single Currency for NAFTA
165
Fig. 2. 4 Cluster Classification from EM Using EEV Model. (a) GDP vs. Unemployment. (b) GDP vs. Lagged GDP.
166
PATRICK M. CROWLEY
The implications from this analysis are that NAFTA does not form an optimal currency area. Of course the argument still remains that if NAFTA adopted NAMU then perhaps NAMU might be an ex-post OCA, but obviously it would be difficult to test this a priori. Also, given the results above, as both the USA and Canada cannot be classified as OCAs, it implies that some other “glue” binds them together (e.g. fiscal transfers, history or political motives). As the lagged GDP cyclical correlations did not appear to give high correlation coefficients for the most part, the exercise was repeated using only three variables and dropping the lagged GDP correlations. Although the model changes to VVV, the optimal number of clusters remains at four. A further modification was undertaken to check the robustness of the results. A casual inspection of correlations reveals that Mexico’s correlations for inflation and unemployment are definite outliers from the group, so the exercise was repeated deleting Mexico from the data set. The optimal number of clusters remained at four, with some memberships changing, and the model once again changed to VVV. Clearly the number of clusters is robust to perturbations in the data, which demonstrates that the results are not anomalous. The issue of Mexico as an outlier,6 though, does raise a problem of how to deal with the effects of NAFTA and the economic recovery of Mexico throughout the 1990s. Given the data limitations here, it would be difficult to assess whether Mexico’s business cycle is now more attuned to the U.S. cycle – anecdotal evidence does appear to suggest that in the late 1990s and early 2000s this has been the case.
5. CONCLUSIONS This paper asked the question should there be “A single currency for NAFTA?” The answer is plainly no, not yet, unless there is some political imperative to do so. The paper used NAFTA data obtained from Canadian, the U.S. and Mexican statistical agencies to determine whether NAFTA can be characterized as an OCA. The results clearly indicated that although the U.S. can largely be characterized as an OCA, it is clear that Canada cannot, and moreover that Mexico does not clearly fall into groupings with either of its North American partners. The research used a model-based cluster analysis approach using a new strategy using Bayesian techniques to choose the optimal number of clusters and the model to be used for the clustering. Further, the clustering algorithm used, the EM algorithm, permits allocation of observations based on a maximum likelihood procedure. The limitations to this approach should be noted. First, the OCA theory would also suggest usage of variables such as labor mobility indicators, real wage
A Single Currency for NAFTA
167
indicators and fiscal transfers. Such variables are not uniformly available over the NAFTA area. Second, as has been noted in the literature on Europe, adoption of a single currency need not be optimal ex-ante, as it may become optimal ex-post through trade linkages and a common monetary policy. Many avenues for future research are apparent. Also, when and if Mexican state data becomes available it would be informative to see if the states bordering the U.S. would be grouped in with U.S. states, given the “maquilladora” plants. Moreover, separate analyses for the U.S. and Canada would clearly be informative (and are forthcoming), given the results obtained here. Lastly, the technique introduced here clearly has many other possible applications in economics.
NOTES 1. Neumeyer (1998) also considers the notion that political shocks could be incorporated as another variable contributing to the factors which might suggest an optimal currency area. 2. Further research by Artis, Krolzig and Toro (1999) has analysed the phasing of the European business cycle. 3. The methods used were phase-average trend (PAT) detrending, a linear trend and a Hodrick-Prescott filter. 4. Data is often available over different time periods by Mexican state, and also the collection methods often differ by Mexican state. This makes data collection and interpretation a challenge that will hopefully be addressed in a future paper. 5. This is because the number of clusters is not considered independent for calculating the BIC, and hence if each model is equally likely, the posterior probability p(x|M) should be higher, and hence the BIC should be higher. 6. George von Furstenburg should be acknowledged here as a discussant who raised the issue of Mexico as an outlier in terms of economic policy.
ACKNOWLEDGMENTS I would like to thank my research assistants Josh Osborne and Mike Cozart at Texas A&M University for excellent assistance over the past two years and Dean Abdelsamad for continuing to fund my research. I would also like to thank Chris Fraley and Adrian Raftery of the Department of Statistics at University of Washington for making the MCLUST model based software available for SPLUS, and Chris Fraley in particular for advice on clustering models. I would like to acknowledge George von Furstenberg’s (Indiana University) comments as a discussant for this paper, and also the helpful comments I have received over several years from Barry Eichengreen (University of California – Berkeley) on my research on optimal currency area issues.
168
PATRICK M. CROWLEY
REFERENCES Anderberg, M. (1993). Cluster analysis for applications. New York, USA: Academic Press. Artis, M., Krolzig, H. M., & Toro, J. (1999). The European Business Cycle. CEPR Discussion Paper 2242, CEPR, London, UK. Artis, M., & Zhang, W. (1997a). International business cycles and the ERM: Is there a European business cycle? International Journal of Finance and Economics, 2(January), 1–16. Artis, M., & Zhang, W. (1997b). On identifying the core of EMU: An exploration of some empirical criteria. CEPR Discussion Paper 1689, CEPR, London, UK. Artis, M., & Zhang, W. (1998a). Core and periphery in EMU: A cluster analysis. EUI Working Paper RSC No. 98/37. Artis, M., & Zhang, W. (1998b). Membership of EMU: A fuzzy clustering analysis of alternative criteria. EUI Working Paper RSC No. 98/52. Banfield, J., & Raftery, A. (1993). Model-based Gaussian and non-Gaussian clustering. Biometrics, 49, 803–821. Baxter, M., & Stockman, A. (1989). Business cycles and the exchange-rate regime. Journal of Monetary Economics, 27, 377–400. Bayoumi, T. (1994). A formal model of optimum currency areas. IMF Staff Papers, 41(4), 537– 554. Bayoumi, T., & Eichengreen, B. (1994a). One money or many?: Analyzing the prospects for monetary unification in various parts of the world. In: Princeton Studies in International Finance (Vol. 76). Princeton, USA: Princeton University. Bayoumi, T., & Eichengreen, B. (1994b). Monetary and exchange rate arrangements for NAFTA. Journal of Development Economics, 43, 125–165. Blanchard, O., & Quah, D. (1989). The dynamic effects of aggregate demand and supply disturbances. American Economic Review, 79, 655–673. Bock, H. (1996). Probability models and hypothesis testing in partitioning cluster analysis. In: P. Arabie, L. Hubert & G. DeSorte (Eds), Clustering and Classification (pp. 377–453). River Edge, NJ: World Science Publishers. Boreiko, D. (2002, March). EMU and accession countries: Fuzzy cluster analysis of membership, Oesterreichische Nationalbank, economics working paper series 71. (Also available as a National Bank of Chile Working Paper.) Buiter, W. (1998). The economic case for monetary union in the European Union. In: C. Deissenberg, R. F. Owen & D. Ulph (Eds), European Economic Integration, Supplement to the Review of International Economics (Vol. 5(4), pp. 10–35) (1997). Buiter, W., Corsetti, G., & Roubini, N. (1993). Excessive deficits: Sense and nonsense in the treaty of maastricht. Economic Policy, 16, 58–60. Crone, T. (1999). Using state indexes to define economic regions in the US. Working Paper 99–19, Federal Reserve Bank of Philadelphia, USA. Croux, C., Forni, M., & Reichlin, L. (1999). A measure of co-movement for economic variables: Theory and empirics. CEPR Discussion Paper 2339, London, UK. Crowley, P. (1996). EMU, Maastricht and the 1996 IGC. Contemporary Economic Policy, XIV(2), 41–55. Crowley, P. (2000). Is Canada an optimal currency area? An exploration of some empirical criteria using provincial data. Paper presented at Western Economic Association meetings in Vancouver, BC, Canada.
A Single Currency for NAFTA
169
De Grauwe, P., & Vanhavebeke, P. (1993). Is Europe an optimal currency area: Evidence from regional data. In: P. Masson & M. Taylor (Eds), Policy Issues in the Operation of Currency Unions (pp. 111–129). Cambridge, UK: Cambridge University Press. De Serres, A., & Lalonde, R. (1994). Sym´etrie des chocs touchant les r´egions Canadiennes et choix d’un r´egime de change, Document de travail, Banque du Canada, pp. 94–99. Eichengreen, B. (1993). European monetary unification. Journal of Economic Literature, 31, 1321–1357. Fraley, C., & Raftery, A. (1998). How many clusters? Which clustering method? Answers via model-based cluster analysis. Computer Journal, 41, 578–588. Fraley, C., & Raftery, A. (1999). MCLUST: Software for model-based cluster and discriminant analysis. Journal of Classification, 16, 297–306. Frankel, J., & Rose, A. (1997). Is EMU more justifiable ex-post than ex-ante? European Economic Review, 41, 753–760. Fratianni, M., von Hagen, J., & Waller, C. (1992). The Maastricht Way to EMU. Essays in International Finance 187. New Jersey, USA: Princeton University. Galbraith, J., & Jiaqing, L. (1999). Cluster and discriminant analysis on time-series as a research tool. UTIP Working Paper 6, LBJ School of Public Affairs, University of Texas at Austin, USA. Gerlach, H. (1988). World business cycles under fixed and flexible exchange rates. Journal of Money, Credit, and Banking, 20(4), 621–632. Hartigan, J. (1975). Clustering algorithms. NY, USA: Wiley. Honohan, P. (2000). Miniblocs and Fringe Currencies of the EMU. Journal of Economic Integration, 15(1), 47–75. Imbs, J. (1999). Co-fluctuations. CEPR Discussion Paper 2267, London, UK. Jacquemin, A., & Sapir, A. (1995). Is a European hard core credible? A statistical analysis, CEPR Discussion Paper 1242. Kass, R., & Raftery, A. (1995). Bayes factors. Journal of the American Statistical Association, 90, 773–795. Kaufman, L., & Rousseeuw, P. (1990). Finding groups in data: An introduction to cluster analysis. New York, USA: Wiley. Kenen, P. (1969). The theory of optimal currency areas: An eclectic view. In: R. Mundell & A. Swoboda (Eds), Monetary Problems of the International Economy. Chicago, USA: University of Chicago Press. Lalonde, R., & St-Amant, P. (1993). Zones mon´etaires optimales: Le cas du Mexique et des Etats Unis. Document du travail, Banque du Canada, Ottawa, Canada, pp. 93–112. Maharaj, E., & Inder, B. (1999). Forecasting time series from clusters. Working Paper 6/99, Department of Econometrics and Business Statistics, Monash University, Australia 33. McKinnon, R. (1963). Optimum currency areas. American Economic Review, 53, 717–755. McLachlan, G., & Krishnan, T. (1997). The EM algorithm and extensions. New York, NY: Wiley. Mukerjee, S., Feigelson, E., Babu, G., Murtagh, F., Fraley, C., & Raftery, A. (1998). Three types of gamma ray bursts. The Astrophysical Journal, 508, 314–327. Mundell, R. (1961). A theory of optimum currency areas. American Economic Review, 51, 657– 675. Pedersen, T. (2000). The Hodrick-Prescott filter, the Slutzky effect, and the distortionary effects of filters. Journal of Economic Dynamics and Control (forthcoming). Tavlas, G. (1993). The ‘New’ theory of optimum currency areas. The World Economy, 16, 663– 685.
170
PATRICK M. CROWLEY
van Ooyen, A. (2001). Theoretical aspects of pattern analysis. In: L. Dijkshoorn, K. Tower & M. Struelens (Eds), New Approaches for the Generation and Analysis of Microbial Fingerprints. Amsterdam, Netherlands: Elsevier. Wynne, M., & Koo, J. (2000). Business cycles under monetary union: A comparison of the EU and U.S. Economica, 67(267), 347–374.
APPENDIX A Labelling Conventions
Canada AB = Alberta BC = British Columbia MAN = Manitoba NB = New Brunswick NF = Newfoundland NS = Nova Scotia ONT = Ontario PEI = Prince Edward Island PQ = Quebec SASK = Saskatchewan U.S. AL = Alabama AK = Alaska AZ = Arizona AR = Arkansas CA = California CO = Colorado CN = Connecticut DE = Delaware FL = Florida GE = Georgia HI = Hawaii ID = Idaho IL = Illinois IN = Indiana
A Single Currency for NAFTA
IO = Iowa KA = Kansas KN = Kentucky LO = Louisiana ME = Maine MD = Maryland MA = Massachusetts MI = Michigan MN = Minnesota MS = Mississippi MO = Missouri MT = Montana NE = Nebraska NV = Nevada NH = New Hampshire NJ = New Jersey NM = New Mexico NY = New York NC = North Carolina ND = North Dakota OH = Ohio OK = Oklahoma OR = Oregon PA = Pennsylvania RI = Rhode Island SC = South Carolina SD = South Dakota TN = Tennessee TX = Texas UT = Utah VT = Vermont VI = Virginia WA = Washington WV = West Virginia WI = Wisconsin WY = Wyoming MEX = Mexico
171
172
PATRICK M. CROWLEY
APPENDIX B Data Sources
Data Sources
Mexico
Canada
United States
Gross domestic product
1980–1991; INEGI, Estadisticas Historicas de Mexico, CD-ROM 1992–1999; IMF, International Financial Statictics, September 2000 INEGI. National survey of Urban Use. http://dgcnesyp. inegi.gob.mx/cgiwin/bdi.exe Bank of Mexico. Indices de Precios. http://dgcnesyp. inegi.gob.mx/ BDINE/L10/ L100013.HTM
All of the provincial data was taken from CANSIM∗
Bureau of Economic Analysis, U.S. Department of Commerce. http://www.bea. doc.gov/bea/ regional/gsp/ action.cfm
All of the provincial data was taken from CANSIM
Bureau of Labor Statistics
All of the provincial data was taken from CANSIM
See Herman B. Leonard and Jay H. Walder, “The Federal Budget and the States, FY1999,” Appendix B for a discussion of the methodology used in constructing these indices.
Unemployment
Consumer price index
A Single Currency for NAFTA
173
APPENDIX C Correlation details
Variable
Mexico
Canada
U.S.
Mean
Standard Deviation
Cyclical component of real GDP Unemployment Inflation
1982–1997
1982–1996
1982–1997
0.509
0.396
1992–2000 1981–1999
1992–1998 1980–1999
1978–2000 1980–1999
0.793 0.939
0.178 0.132
NORTH AMERICAN INTEGRATION AND THE SINGLE CURRENCY? Martin Coiteux ABSTRACT The objective of this paper is two-fold. First, through the use of two indices that correct for the bias introduced by the relative size of the U.S. economy, it describes the degree of regional dependence reached in the NAFTA area as compared to Europe. Second, it asks whether the stage is set for the successful adoption of a single currency, namely the dollar, within the area. In an attempt to answer this question, the paper identifies the one critical asymmetry that should be addressed before considering the complete monetary integration of the area.
INTRODUCTION Despite all the talk about globalization, the world economy remains largely segmented along regional lines. Not only the small and medium-sized firms but also the largest corporations are likely to produce and sell in the same regional market (Rugman, 2001). For obvious reasons, this is particularly true of the regions that display the largest market size. A firm located in the territory of the European Union can serve a market that roughly represents 30% of world GDP and the same is true of a firm located in the North American geographical space. This microeconomic reality translates into a strong regional bias in the trade relationship of the countries that are members of the largest integration areas. For instance, Belgium and Spain are conducting more than 70% of their trade with North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 175–191 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10010-7
175
176
MARTIN COITEUX
other European Union countries. For Canada and Mexico, the comparable figure within the NAFTA area stands above 80%. The consequences of this trade bias are not only static. As the economies of Europe and North America are growing more open, their macroeconomic dependence on the regional market is becoming ever larger. Once a critical degree of dependence is reached, large changes in bilateral exchange rates may turn out to be destabilizing instead of stabilizing, as standard economic theory would take for granted. In Europe, there is little doubt that the destabilizing consequences of the large exchange rate changes experienced in 1992 greatly helped the political case in favour of a monetary union. This brings the question of whether the stage is set for the successful adoption of a single currency in North America. After all, when it comes to trade and investment matters, North America is perhaps a more integrated area than Europe. This point is well understood in Canada and Mexico as well as near the northern and southern borders of the United States. In Canada, this has led some prominent business leaders and academics to go public and advocate the adoption of the dollar as the single currency of North America (see in particular Harris, 2001; Tellier, 2001). So far, their case has not found much political echo in Ottawa and clearly, the issue of a single currency still remains off the radar screen in Washington. However, this should not be taken to imply that the political agenda would never change in the future. The mobilization of the business elites in the second half of the 1980s did a lot to convince the Canadian public opinion of the benefits of a free trade agreement with the United States and, in a meeting of the House of Commons Financial Committee held in May 2001, even Bank of Canada’s governor David Dodge admitted that the preference for a distinct and floating Canadian currency could change within ten years. The objective of this paper is two-fold. Taking Europe as a benchmark, it first describes the extent of regional dependence reached in North America by looking at two trade-related indices that correct for the bias caused by the relative size of the U.S. economy. The paper then goes on to ask whether the stage is set for the successful adoption of the dollar as the single currency of North America. In an attempt to answer this question, this paper identifies one critical, albeit generally overlooked asymmetry, which should be addressed prior to moving to the complete monetary integration of the area. Some prospective thoughts about the future are laid down at the end.
MEASURING REGIONAL DEPENDENCE There is no unique way of measuring the degree of regional dependence. The health of the Canadian and Mexican economies obviously depends very much on how
North American Integration and the Single Currency
177
well the U.S. economy is doing. When asking whether Canadian regions should adopt the U.S. dollar, Beine and Coulombe (2001) are looking at the correlation between the business cycles of the Canadian provinces on the one hand, and the U.S. business cycle on the other hand. On this account, the economies of the two most populous and industrialized provinces, Ontario and Quebec, are particularly dependent on the U.S. market and this translates into a strong overall Canadian dependence on the U.S. economy. Obviously, behind this high overall economic dependence lies a particularly strong level of trade dependence. The United States absorb over 85% of Canada’s exports, a very high figure indeed when one considers that the European Union absorbs less than 75% of the combined exports of Belgium and Luxembourg, two relatively small countries within the Union. Now, it is well known that the relative sizes of the trading partners impact very much in the result of calculations such as these. Unless the trade partners are all of equal size, some form of normalization is called for (Frankel, 1997). The “trade intensity index” (TII) is a particular form of normalization that has been used in numerous studies. One of the earliest references to this index is Kojima (1964) while some more recent published examples are Drysdale (1988), Eden and Appel Molot (1992), Hufbauer and Schott (1993) and Yeats (1998). When applied to a country’s total exports, the index is computed as follows: TII =
X ij /X i M j /M w
(1)
where X ij /X i represents the share of country i’s total exports that goes to region j and M j /M w represents region j’s share of world imports (Mw ). If region j is large in the sense of accounting for a large share of world imports, country i should ship a large share of its exports to that particular region. A computed ratio of 1 indicates neutrality in the trade relationship while a computed ratio larger than 1 reveals a regional bias or dependence. Although it is most common to find the TII computed for exports only, it can also be computed for exports and imports altogether (see Rajan, 1996). An alternative proposed by Blank and Coiteux (2002) consists in taking the ratio of two total trade intensity indices. This new index, which could be called the “relative trade intensity index” (RTII), is computed as follows: RTII =
(X + M)ij /(X + M)i,W−j (X + M)j − (X + M)ij /(X + M)W−j − (X + M)i,W−j
(2)
The numerator of this new index represents the ratio of country i’s total trade with region j to its trade with the rest of the world (W−j). The result is normalized by the ratio of partner j’s trade with the rest of the world (netting out trade with i) to the rest of the world trade with partners other than country i. If partner j is a relatively
178
MARTIN COITEUX
large world trader (in which case the denominator will be relatively large), country i should make a large part of its trade with this single partner. Again, a computed ratio of 1 indicates neutrality and a computed ratio larger than 1 reveals a regional bias (or dependence) in the trade relationship. The relative trade intensity index (RTII) defines the size of a particular region in terms of its relative importance in world trade. An alternative would be to define the size of the same region in relation to its relative GDP. This is the approach favoured by Summers (1991). The so-called gravity approach to international trade explicitly links the intensity of a trade relationship to the size of the partner’s GDP. In this spirit, Coiteux (1998) proposes the following “gravity based trade intensity index” (GBII): GBII =
(X + M)ij /(X + M)i,W−j GDPj /GDPW−i−j
(3)
Indices (2) and (3) are fairly similar except that in the latter case, it is partner j’s relative GDP (with respect to the rest of the world W−i−j) that defines economic size. Both indices have their merits. The first one focuses on the global trade importance of the partners; the more they trade globally (relative to the rest of the world), the more likely it is that a particular country will trade with them. The second one focuses on their sheer market size. The larger a partner’s GDP (relative to the rest of the world), the more likely it is that a particular country will trade with it. The “combined intensity index” (CII) proposed by Blank and Coiteux (2002) gives an equal weight to both definitions of economic size: √ CII = RTII GBII (4) In Table 1, the computed combined intensity indices for the three North American countries and a representative sample of European Union countries are presented for a selected number of years starting in 1984. The required data came from the United Nations Direction of Trade Statistics Division and the World Bank. At the moment of computing the indices, the complete data were available up to 1999. The results are striking in two respects. First, considering that a computed value of 1 would imply neutrality in the trade relationship, North American and European Union countries are highly dependent on their region. No country included in the sample approaches in any way this theoretical neutrality. Even the less dependent countries (Italy and the U.K.) still trade with their region more than three times what size would justify. But even more striking is the fact that the three North American countries are also the three most regionally oriented countries and that the United States is topping the list in trading with Canada and Mexico more than 20 times what their combined size would theoretically justify.
North American Integration and the Single Currency
179
Table 1. Dependence on Regional Trade as Measured by the Combined Intensity Index.
NAFTA Canada Mexico USA Europe Belgium-Luxembourg France Germany Italy Netherlands Spain U.K.
1984
1987
1990
1993
1996
1999
9.5 5.6 11.8
10.6 9.4 13.9
12.6 8.8 14.1
15.2 12.5 16.6
17.7 17.3 21.6
16.7 15.2 21.6
6.5 3.9 4.5 3.1 5.4 2.4 3.0
6.3 4.3 4.1 3.6 4.8 3.8 3.1
6.8 4.8 5.9 4.3 5.5 5.1 3.6
6.8 5.0 5.9 4.1 5.2 5.9 3.2
6.3 5.1 5.7 3.8 5.2 5.9 3.4
6.5 5.3 5.2 3.9 5.0 6.0 3.4
Now, one may wonder whether this is supposed to mean that the United States depend more on its NAFTA partners than vice versa. After all, the share of exports and imports in GDP is much lower for the United States (below 20%) than it is for Mexico (around 50%) and Canada (above 70%), not even talking about Belgium and Luxembourg (around 130%). The same combined intensity index cannot mean the same thing for a country whose trade represents 20% of GDP than for a country where it represents 130%. There is a logical way to correct for this particular bias and it simply consists in multiplying the combined intensity index (CII) by the country’s openness ratio (OR), computed as the sum of the country’s exports and imports, divided by its GDP. The new index, which could be called the “regional dependence index” (RDI), computed for the countries of the sample, is presented in Table 2. Several interesting results can be drawn from this simple calculation. First and most obvious, the United States no longer appears to be the most regionally dependent country of the sample but still, with a regional dependence index of 4, they are more regionally dependent than Europe’s two largest economies, France and Germany, whose regional dependence indices stand between 2 and 2.5. Based on this particular index, the United States looks more similar to the Netherlands, a country whose relative size in Europe is roughly similar to Canada’s in North America. Second, except for Spain, which upon its inclusion in the European Community, started converging to the dependence levels reached in France and Germany, Europe’s regional dependence indices have been relatively stable over the period. On the contrary, they exploded in North America, as a result of
180
MARTIN COITEUX
Table 2. Regional Dependence Index. 1984
1987
1990
1993
1996
1999
NAFTA Canada Mexico USA
4.3 1.3 1.7
4.7 3.1 2.0
5.4 2.3 2.2
7.6 3.4 2.8
11.3 8.8 4.2
11.8 7.9 4.0
Europe Belgium-Luxembourg France Germany Italy Netherlands Spain U.K.
8.6 1.6 2.3 1.1 6.0 0.8 1.4
7.2 1.5 1.9 1.1 4.4 1.1 1.2
8.1 1.8 2.9 1.3 5.4 1.4 1.4
7.1 1.6 2.2 1.3 4.4 1.7 1.3
7.3 1.8 2.3 1.4 5.0 2.2 1.6
8.5 2.2 2.4 1.4 5.4 2.6 1.4
the opening of the Mexican economy after 1984, the signature of a free trade agreement between Canada and the United States in 1988 and then NAFTA in 1994. Canada is now the real stand out of the sample with a dependence index approaching 12. With a computed regional dependence index close to 8, Mexico looks as regionally dependent on the North American market as Belgium and Luxembourg are on the European Union market. Clearly, the Mexican reforms and the consolidation of the free trade area have created a state of interdependence in North America probably not even seen in Europe. If trade and economic dependence plays an important role in tilting the balance of monetary preferences in favour of a single currency, one may ask why, 10 years after NAFTA, North American capitals still show little interest in the idea. Canada depends more on the regional market than the smallest European countries and the United States is more regionally dependent than France and Germany. Of course, the geo-political realities of the two continents differ. Power is more evenly distributed in Europe than it is in North America. Fear of the “elephantine embrace” remains a serious issue for Canada, especially in the post-September 11 environment, characterized by more self-assertiveness and a decisively more unilateral approach to international relations on the part of the United States. But as the free trade debate of the 1980s witnessed, there are powerful economic forces at work, which overtime, can radically change the terms of the debate. Moreover, the Canadian authorities’ main line of defence for the floating Canadian dollar remains an essentially economic one. In trying to answer the question of whether a North American currency makes sense in the present conditions,
North American Integration and the Single Currency
181
the rest of the paper thus continues to stress the economic over the political factors.
A SINGLE CURRENCY FOR NORTH AMERICA: SOME THEORETICAL CONSIDERATIONS In principle, the extensive body of literature known as the theory of optimum currency areas could be used in order to address the question of a single currency for North America. However, this literature has produced so wide an array of criteria to consider (see Tavlas, 1993; Tower & Willett, 1976) that it is hardly operational. This is why most empirically oriented authors working in the area have reduced the dimension of the problem to a question of symmetry and asymmetry of shocks between the countries that are candidates to a monetary union. In this approach, fixed exchange rates and by extension a single currency are considered to be desirable only when the shocks are found to be symmetric. Following the seminal work of Blanchard and Quah (1989), a series of studies associated with Bayoumi has approached this question of asymmetry in an aggregate supply – aggregate demand setting, using the methodology of structural VAR decomposition to identify the nature of the shocks and their degree of symmetry. This approach has been specifically applied to North America in Bayoumi and Eichengreen (1994) and in a series of studies produced at the Bank of Canada (see for instance DeSerres & Lalonde, 1994; Dupasquier, Lalonde & St-Amant, 1998; Lalonde & St-Amant, 1993). In general, these studies have tended to support the current preference for separate and floating currencies in North America although the most recent is a little more cautious: the cost for Canada of pegging the value of its currency to the U.S. dollar would not differ much from the cost for European countries of fixing their exchange rates together. The structural VAR decomposition approach has its merits but it also suffers from a number of shortcomings. One problem is the arbitrariness by which demand and supply shocks are identified. The most serious problem however, is probably the fact that the approach considers the exchange rate only very indirectly, at best. The shocks considered have output and prices consequences but apparently no exchange rate consequences. This is a bit curious since the ultimate objective is to decide whether it makes sense to renounce to the option of floating the exchange rate. An approach that focuses directly on the exchange rate is that of Coiteux and Geandet (2002). In very simple terms, it consists in looking at the symmetry or asymmetry of the determinants of the equilibrium real exchange rate; the latter
182
MARTIN COITEUX
being defined as the level of the real exchange rate that is compatible with internal (full-employment) and external (balance of payments) equilibriums. Let us consider the bilateral real exchange rate (BRER) between Canada and the United States: BRER =
EP CAN , P USA
(5)
where E designates the value of the Canadian dollar expressed in U.S. dollars and the two P’s represent the respective price levels (GDP deflators) of Canada and the United States. Let us consider the shock most often cited by the Bank of Canada in illustrating the benefits of a floating exchange rate. Canada is a net exporter of commodities while the United States are a net importer. When the price of commodities fall in international markets, the value of Canadian exports tend to fall and this brings in turn a fall in profits (with ripple effects on wages and tax income) and a downturn in economic activity and employment. In order to offset these recessive pressures, keeping constant the overall investment-saving balance, Canada needs to increase the real value of its net exports. This in turn requires the operation of the substitution effects that a fall in the real exchange rate normally produces. In principle, this could be achieved by letting Canadian prices fall relative to U.S. prices at an unchanged level of the exchange rate. But Canadian monetary policy works along the lines of an explicit inflation target band. Moreover, on the basis of both the stated objectives and the recent experience, the Bank of Canada and the Federal Reserve seem happy to deliver basically the same inflation rate. This means that a fall in the value of the Canadian dollar is required when the international price of commodities falls. As the Bank of Canada keeps repeating, this is exactly what happens under the current floating regime. Leaving aside for the moment the fact that commodities are weighting ever less in the vector of Canadian exports, there are obviously other shocks that are susceptible of changing the value of the equilibrium exchange rate. Even if economists have not been particularly successful in predicting the timing and magnitude of exchange rate changes (see for instance Macdonald, 1995), the literature is quite clear in identifying the terms of trade (which can be affected by the evolution of commodity prices), the productivity differentials and the direction (and size) of capital flows (net of interest and dividends payments) as the three main variables responsible for these changes.1 In this context, a relevant question for the desirability of a monetary union is whether these determinants converge in a particular region. Obviously, if the three North American economies tend to experiment similar changes in these three variables, they will also tend to require the same change in their equilibrium real
North American Integration and the Single Currency
183
exchange rate. With a common inflation target, they could easily float together and if they decided not to do so, their bilateral real exchange rates would nevertheless tend to stay constant. In the next section, this approach is applied to Canada and the United States, as some of the required data were not available for Mexico.
IDENTIFYING THE KEY ASYMMETRY BETWEEN CANADA AND THE UNITED STATES In order to examine the convergence in the determinants of the bilateral equilibrium real exchange rate, three ratios comparing Canada to the U.S. have been computed: the ratio of one plus net capital inflows as a percentage of GDP (excluding changes in official reserves and netting out interest and dividend payments), the ratio of the terms of trade and the ratio of the productivities in the business sector. These ratios are compared to the Canada-U.S. bilateral real exchange rate computed as in Eq. (5). The Canadian GDP has been converted into nominal U.S. dollars by using the 1995 average nominal bilateral exchange rate, multiplying the result by the U.S. GDP price deflator (in base 1995 = 1). The quarterly exchange rates, the required price indices, the GDP’s and the balance of payments series (complete starting in 1977) all come from the IMF. The quarterly productivity series (available for Canada starting in 1987) come from Statistics Canada and the Federal Reserve Bank of Saint-Louis. In Fig. 1, a four-quarters simple moving average of the relative capital inflows series is compared to the bilateral real exchange rate.2 The least that can be said is that, except for a brief episode, which occurred around the Asian crisis, the two series are closely related one to the other. The same can definitely not be said of the relative terms of trade and the bilateral real exchange rate presented together in Fig. 2. In particular, the relative terms of trade are unable to explain the persistent downward trend in the real exchange rate observed since the beginning of the 1990s. Although they fluctuate very much, the relative terms of trade have stayed roughly constant on average after an abrupt fall at the beginning of the 1980s. For the 1990s, Canada’s relative productivity in the business sector appears to be a better candidate (see Fig. 3) although it fails to match bilateral exchange rate movements as closely as do relative capital inflows. The absence of a strong correlation between the relative terms of trade and the bilateral real exchange rate is a bit surprising given the Bank of Canada’s insistence that fluctuations in the international price of commodities are the main culprit for the wide fluctuations in the Canadian exchange rate. To be true, there is a strong correlation between a subset of the terms of trade emphasized by the Bank of
184
MARTIN COITEUX
Fig. 1. Relative Capital Inflows and the Bilateral Real Exchange Rate.
Fig. 2. Relative Terms of Trade and the Bilateral Real Exchange Rate.
North American Integration and the Single Currency
185
Fig. 3. Relative Productivity and the Bilateral Real Exchange Rate.
Canada, the relative price of non-energy commodities, and the exchange rate. It remains unclear however why only non-energy commodities should matter and not other commodities, not to mention the other goods and services, which represent the bulk of Canadian exports. Of course, if the Canadian economy is perceived as an essentially non-energy commodities oriented economy; relative capital inflows will tend to turn in its disfavour when the price of these commodities falls. Figure 4 gives some credit to this hypothesis as there indeed appears to be a correlation between the relative price of non-energy commodities used by the Bank of Canada and Canada’s relative capital inflows. However, the correlation is far from being as strong as the one observed between the real exchange rate and relative capital inflows. Clearly, there is much more at work than the price of commodities. Moreover, given that the relative terms of trade have been fluctuating around an essentially constant mean since the mid-1980s, the widely held perception of Canada as being an essentially commodities oriented economy needs be revised. Let us then focus on what has been the essential asymmetry between Canada and the United States over the recent period. Capital inflows, especially for countries with such an easy access to world financial markets, cannot be considered exogenous. The relative supply of financial capital to the firms of a particular country obviously depends on the comparative expected risk-adjusted returns. Even if much of it turned out to be a bubble, the productivity and profit boom that characterized much of the 1990s in the United States has not been paralleled in Canada.
186
MARTIN COITEUX
Fig. 4. Relative Capital Inflows and the Bank of Canada’s Relative Price of Commodities Series.
This factor, plus the fact that the price of non-energy commodities has been simultaneously on a downward trend, have probably contributed to attracting a larger share of world savings to the stock and bonds markets of the United States at the expense of Canada’s. On the demand side and perhaps more importantly, Canada’s decision to tackle its public debt and deficit problems in the 1990s dramatically reduced the need for foreign financing, as the increase in the private sector financial needs did not fully match the increase in public savings. Even if the U.S. has eliminated its budget deficit over the same period (before recently going back into deficit), the required fiscal adjustment has been more modest south of the Canadian border. Other factors have certainly played an important role in creating the net capital inflows asymmetry that seems to be responsible for the 1990s decline in the value of the Canadian dollar. According to OECD figures, the stock of Canadian foreign direct investment in the U.S., evaluated in constant 1995 U.S. dollars, was representing only 54% of the stock of U.S. foreign direct investment in Canada in 1984. In 2000, this proportion had reached 83%.3 This supply factor, which in part reflects the successful integration of the two economies within a single goods and services market, has also had some demand side consequences. In a recent conference organized at HEC Montr´eal to discuss the extent of North American financial integration, the CFO of an important Canadian transportation services company explained to the participants how his company was hedging its acquisitions in the U.S. by levying the required funds in the U.S. market. As more than one Canadian firm has most probably adopted this quite natural and
North American Integration and the Single Currency
187
logical hedging strategy, the catching up of foreign direct investment has also probably been partly responsible for the increase in the demand for funds in the U.S. market at the expense of the Canadian market. One last ingredient to mention is the increase in the share of U.S. dollar denominated assets in the investment portfolio of the Canadian household sector observed during the 1990s. This development, which probably reflects an independent trend towards the international diversification of assets, has been facilitated by the partial liberalization of the restrictions applying to the pension funds and registered retirement savings plans (RRSP) accounts. The sum of these developments gives us a pretty plausible story for the observed decline in the relative capital inflows to Canada. But which of the individual developments weighted more? Without the required detailed empirical analysis, the question cannot be answered for sure. However, given the relatively large share of the government sector in Canada, the safer bet probably lies in the asymmetry in fiscal policies. In relation to GDP, the Canadian government sector accumulated more debt in the 1970s and the 1980s than the American government sector. This implied a much stronger fiscal adjustment in the subsequent period. In order to preserve a satisfactory level of economic activity and employment in this environment, Canadians had to accept a lower value for their currency. In other words, it would have been difficult for Canadians to live with the U.S. dollar in the 1990s. This, however, reflects the past. Towards the end of the 1990s and the beginning of the current decade, there had been a remarkable fiscal convergence between the two countries and the business productivity trends also looked more even. Factoring in the stability of the relative terms of trade over the medium-run, living with the U.S. dollar, in a country that depends more on the U.S. market for its economic well being than Belgium and Luxembourg depend on the European Union market, did not look like an heresy. But of course, the future must also be considered.
LOOKING TOWARDS THE FUTURE This section of the paper is definitely more prospective. If the analysis presented in this paper is correct, the Bank of Canada’s main line of defence in favour of a floating Canadian dollar is a bit overstated. The price of non-energy commodities may be falling at times but the capital inflows and outflows that seem to be responsible for the changes in the value of the Canadian dollar obey other powerful forces. In particular, at the beginning of the present decade, the asymmetry in fiscal policies that would have made living with the U.S. dollar difficult for Canada in
188
MARTIN COITEUX
the 1990s, looked behind us rather than beyond us. This promised to bring more homogeneity in the capital inflows to Canada and the United States even if Canada still need to change the not too well founded perception that it is an essentially commodities-oriented economy. However, the recent deterioration in the fiscal accounts of the United States may bring a new period of divergence. For now, this has not translated into larger relative capital inflows to the United States because the U.S. private economy has considerably reduced its investment driven demand for funds. Canada’s fiscal accounts have also deteriorated but generally remain in surplus, with a private economy that is for now faring better than its U.S. counterpart. Where does this discussion lead us? The European idea of convergence prior to monetary union was not such a bad idea after all. Convergence in fiscal policies, long term interest rates and inflation is certainly instrumental to more stability in relative capital inflows and consequently the exchange rate. Without the selfimposed discipline of a Maastricht-like pact, Canada and the United States have already achieved convergence in interest rates and inflation rates. Monetary conditions in Canada and the United States are basically the same, except for the frequent large changes in the value of the bilateral exchange rate. Down the road to a hypothetical monetary unification, there thus remains the need for some form of harmonisation in fiscal policies. Up to very recently, a tacit consensus on what constitutes sound economic policies seemed to be delivering for North America what Maastricht had delivered for Europe. This does no longer appear to be the case as the fiscal choices of the current U.S. administration differ markedly from the choices made by the Canadian federal and provincial governments. This situation is unlikely to change in the immediate future and given the overwhelming size of the U.S. economy, North America is unlikely to deliver a Maastricht-like pact of fiscal harmonisation. Does this mean that the idea of a common North-American currency, which up to very recently seemed to gain momentum in some sectors of the Canadian business and academic circles, will lose all its steam? This is unlikely, even if the debate will be postponed for some time. Given the extreme dependence on the U.S. economy reached over the last decade, business and public opinion in Canada will not easily tolerate new possibly large changes (upward or downward) in the exchange rate. For this reason and a few others, the pressures in favour of more harmonisation and exchange rate stability will not abate. Nevertheless, as the United States is currently revising in a fundamental way its role in the world, North America is also entering a period of great uncertainty. In this period of uncertainty, Canada will most likely adopt a wait and see attitude and it is very unlikely that NAFTA will evolve towards a European like economic and monetary union.
North American Integration and the Single Currency
189
CONCLUSION By looking at two trade related indices of regional dependence that correct for the size of the U.S. economy, this paper argued that the North American economy is much more regionally oriented than the European economy. In particular, Canada depends more on the (mainly U.S.) regional market than Belgium and Luxembourg depend on the European market. Mexico is another stand out and the United States are more regionally dependent than France and Germany. If monetary union has been thought to be good for Europe, why couldn’t it be good for North America? If the asymmetry in power is always present as a fundamental difference between Europe and North America, the paper nevertheless generally focused on the potential economic asymmetries. In a framework that emphasizes the exchange rate over output and prices fluctuations, the asymmetry in fiscal policies has been identified as the key factor that would have made living with the U.S. dollar difficult for Canada in the recent past. The asymmetry represented by the larger share of commodities in the Canadian vector of exports may be a factor but it does not appear to be the decisive factor as so often claimed by the Bank of Canada. Up to very recently, fiscal policies in Canada and the U.S. were converging to an extent such that living with the U.S. dollar in an economy so dependent on the U.S. market could make perfect sense. However, current fiscal and political developments in the U.S. inhibit and possibly revert convergence. In this context, the case for adopting the U.S. dollar will momentarily lose its momentum in Canada. Given the strong degree of regional dependence, pressures in favour of more harmonisation and exchange stability will remain but it is unlikely that NAFTA will evolve in the near future towards more than what it currently is, a free trade area.
NOTES 1. Some of this literature is surveyed in Coiteux and Geandet (2002). 2. Required to smooth out erratic quarter-to-quarter fluctuations. 3. Author’s calculations using OECD figures.
ACKNOWLEDGMENTS The author would like to thank Alan Rugman for organizing the conference that motivated the writing of this paper and Lorraine Eden for her most useful help in identifying some of the earliest references to the trade intensity index in the literature.
190
MARTIN COITEUX
REFERENCES Bayoumi, T., & Eichengreen, B. (1994). Monetary and exchange rate arrangements for NAFTA. Journal of Development Economics, 43, 125–165. Beine, M., & Coulombe, S. (2001). Should Canadian regions adopt the U.S. dollar? Mimeo, University of Ottawa. Blanchard, O. J., & Quah, D. (1989). The dynamic effects of aggregate demand and supply disturbances. The American Economic Review, 79(3), 655–673. Blank, S., & Coiteux, M. (2002). The state of North American integration. Paper presented at the First Meeting of the Study Group on North American Monetary and Financial Integration, HEC Montr´eal, October 31st and November 1st. Coiteux, M. (1998). A tale of two fortresses: Measuring the regional bias in trade relations. Working Paper. Centre for International Business Studies (CETAI), HEC Montr´eal. Coiteux, M., & Geandet, R. (2002). Does monetary union make sense for Brazil and Argentina? Paper presented at the 37th meeting of the Argentine Association of Political Economy (http://www.aaep.org.ar/espa/anales/02.htm). Tucum´an. November 13th–15th. DeSerres, A., & Lalonde, R. (1994). Sym´etrie des chocs touchant les r´egions canadiennes et choix d’un r´egime de change. Bank of Canada Working Paper, Ottawa, 94–99. Drysdale, P. (1988). International economic pluralism, economic policy in East Asia and the Pacific. New York: Columbia University Press. Dupasquier, C. R., Lalonde, R., & St-Amant, P. (1998). Optimum currency areas as applied to Canada and the United States. Money Affairs, 11, 1–35. Eden, L., & Appel Molot, M. (1992). Comparative and competitive advantage in the North American trade bloc. Canadian Business Economics, 1(1), 45–59. Frankel, J. A. (1997). Regional trading blocs in the world economic system. Washington, DC: Institute for International Economics. Harris, R. (2001). Symbol of weakness. As globalization increases the power of the U.S. greenback, the currencies of small countries such as Canada risk disappearing altogether. Financial Post, November 1st. Hufbauer, G. C., & Schott, J. J. (1993). NAFTA: An assessment. Washington, DC: Institute for International Economics. Kojima, K. (1964). The pattern of international trade among advanced countries. Hitotsubashi Journal of Economics, 5(June), 16–36. ´ Lalonde, R., & St-Amant, P. (1993). Zones mon´etaires optimales: Cas du Mexique et des Etats-Unis. Bank of Canada Working Paper, Ottawa, 93–112. Macdonald, R. (1995). Long run exchange rate modeling. A survey of the recent evidence. IMF Staff Papers, 42(3), 437–439. Rajan, R. S. (1996). Singapore’s bilateral merchandise trade linkages with Japan and the U.S. Asian Economic Journal, 10(2), 378–389. Rugman, A. (2001). The end of globalization. Why global strategy is a myth and how to profit from the realities of regional markets. New York: American Management Association. Summers, L. (1991). Regionalism and the world trading system. In: Federal Reserve Bank of Kansas City (Ed.), Policy Implications of Trade and Currency Zones (pp. 295–301). Wyoming: Jackson Hole. Tavlas, S. G. (1993). The new theory of optimum currency areas. The World Economy, 16(6), 663–685.
North American Integration and the Single Currency
191
Tellier, P. (2001). The dollar question. Sooner or later, there will be just a handful of currencies through which all trade will be conducted – the euro, the U.S. dollar and the Chinese yuan. Financial Post, December 19th. Tower, E., & Willett, T. D. (1976). The theory of optimum currency areas and exchange-rate flexibility. Princeton University Press. Yeats, A. J. (1998). Does Mercosur’s trade performance raise concerns about the effects of regional trade arrangements. The World Bank Economic Review, 12(1), 1–21.
CANADIAN INVESTMENT IN THE UNITED STATES: A REGIONAL PERSPECTIVE Prem Gandhi and Wayne Glass ABSTRACT It is widely accepted that international trade and investment are complementary to each other at the macro level. Investment flows to areas where the market opportunities are the greatest, resulting in greater integration of the markets. However, once the decision to invest abroad is made by companies, where they locate, (i.e. which states/regions in the host country the company will choose to locate) is not very well appreciated. In order to gain greater understanding of the factors affecting such decisions, this study seeks to identify the factors that are important to Canadian companies for investing in the United States, given their choice of locations in the U.S. The managers’ views of further U.S.-Canadian economic integration are also investigated and verified.
INTRODUCTION This study is a three-dimensional effort to bring together theories of international trade and investment with regional and location analysis. It is widely accepted that international trade and investment are complementary to each other at the macro level. Investment flows to areas where the market opportunities are the North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 193–209 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10011-9
193
194
PREM GANDHI AND WAYNE GLASS
greatest, thus resulting in greater integration of the markets. However, once the decision to invest abroad is made by companies, where they locate (i.e. which states/regions in the host country the company will choose to locate) is not very well appreciated by researchers. The choice of location is particularly important to economic development agencies that seek outside investment to increase income and employment in their regions. Given these issues, this study seeks to identify the factors that are important to Canadian companies in: (1) making a determination to invest in the United States; and (2) refining their choice to a given location in the U.S. To address these factors, this study focuses on three questions: (1) Why do Canadian companies invest in the United States? (2) Why do they invest in the particular regions/states in the U.S.? (3) What is their attitude towards further integration of the two economies, beyond NAFTA? This question is extremely important given the present level of integration of the two countries.
BACKGROUND To put the study in context, it is well established that U.S. and Canada are each other’s largest customers and that both are large investors in each other’s economy. U.S.-Canadian trade is nearly $1.3 billion a day, estimated to be $445 billion in 2001. In fact, between 1988, when the Canada-United States Free Trade Agreement was signed, and 1994 when Mexico joined NAFTA, trade between the two countries increased from $194 billion to $259 billion, a 34% increase. By 2002, trade between Canada and the United States had risen to $445 billion, a 58% increase since 1994 (Canada, 2002). All states in the U.S. contribute to this trade with 38 states considering Canada as their leading export market. One of the reasons to explain this pattern is the greater integration and complementarity of industries between the two countries. Almost 58% of U.S. exports and 53% of Canada’s exports include the same product categories: transportation equipment, telecommunication, metals, machinery, and equipment and tools. Moreover, most of the top trading states (Michigan, Ohio, Illinois, New York, Pennsylvania, Indiana, New Jersey and Wisconsin) are located near the U.S-Canadian border and are in the U.S. manufacturing belt. Economists have always concentrated on the international flow of goods and services to measure the economic welfare of the trading partners. The flow of international investment has been relatively neglected for study, either because in a tariff-ridden economy investment is an alternative to trade, or because investment flows are not a part of country’s GDP. It is only lately that trade and investment
Canadian Investment in the United States
195
have come to be realized as two sides of the same coin. International investment strengthens trade flows, as demonstrated by Canada and the United States. The United States is the largest investor in Canada with a total stock of direct investment estimated to be $139 billion in 2001. Canada has the 2nd largest concentration of U.S. investment behind the United Kingdom with 10.2% of U.S. total investment abroad, representing 67% of Canada’s total stock of international investment. At the same time, Canada is the sixth largest investor in the U.S., with an estimated value of $102 billion or 8.2% of foreign assets held in the U.S. (Canada, 2002). That is, 50% of total Canadian investment abroad is held in the U.S. While Canadians have always lamented the domination of U.S. investment on their economy, the reality is that Canada has historically invested twice as much in the U.S. on a per capita basis than the U.S. has in Canada (Myers, 1999). The result is that while U.S. investment was 8 times larger than Canadian investment in the U.S. in 1967, that ratio today stands at less than 1.4 times. While attraction of the U.S. as a location for Canadian investors is still prominent, the proportion of Canadian investment coming to the U.S. is declining. In 1997, 52% of total Canadian foreign direct investment was destined for the U.S. compared to 67% in 1987. Still, this record is not unimpressive when one realizes that Canada is 1/10th the size of the United States (Rugman, 1987). Further, most studies on the subject have come from Canada and have concentrated on the Canadian multinational corporation in the U.S. Their emphasis is at the macro level, which authors like Alan Rugman have captured in their research about the strategic decision making of such corporations going back as far as the late 1970s (Rugman, 1980, 1985, 1987, 1990). But what about the investment made by small and medium-size companies that have made substantial contributions to integrating the two economies? A few Canadian authors such as Litvak and Maule (1981), in their study of twenty-five Canadian small companies in the United States point to the relative size, growth potential, and increased service to American customers as the most important reasons for investing in the U.S. Among other important reasons, companies pointed to the presence of tariff and non-tariff barriers in the United States, along with the technological strength of Canadian companies and their experience with international business. All other considerations at this juncture would seemed less critical. Rugman (1987) includes reference to two studies made in 1985. One, by Forget and Denis, found the pull factors on the United States in terms of market size, potential to expand, and proximity to the customers were more important than the Canadian push factors, including taxation, unit labor cost, and more restrictive regulations. They did not find tariffs significant in the companies’ decision to locate in the United States. However, the issue of foreign tariffs influencing a
196
PREM GANDHI AND WAYNE GLASS
company’s decision to invest abroad remains unsettled. The second study by Neil Matheson, in analyzing the reasons for eighteen Canadian companies that invested abroad, found tariffs and the desire to service the foreign market to be the principal reasons in half of the cases. While such studies concentrate on the international flow of capital between Canada and the United States, they do not concern themselves with the choice of locating such investment in particular regions of the United States once the decision to invest in the United States has been made. In other words, these studies do not address the question of why, once it decides to invest in the United States, a Canadian company chooses a particular location for its operations, when it can operate from any point in the United States. Canadians, like others, have to choose a location consistent with their corporate objective(s). A large proportion of the Canadian investment is destined for the Eastern United States, especially the Northeast. Canadian companies have invested 60% of their total investment in the mid-east and the Great Lakes regions of the U.S. These two regions account for almost 36% of the U.S. gross domestic product (O’Hagan & Anderson, 2000). In the U.S., a few authors have studied Canadian investment in their region/states. Studies by McConnell and MacPherson (1990) and McConnell (1999) in Western New York, Drennen (1997), Kasoff, Benedict and Lauer in Ohio (1997), and Richardson in Washington State (1997) are some examples of the studies of the impact made by Canadian companies in these areas. Gandhi (1991) has kept track of Canadian inflows in New York and Vermont, starting with Clinton County in Northern New York in the early 1970s. However, each of these studies has been location-specific. No study has taken a comparative regional approach, which this study attempts to rectify. That is, the earlier studies do not explain the choice of one specific location over another, (e.g. western New York v. Ohio or West Virginia). Moreover, calculating the cost-benefit of choosing a given location to produce and to service a market is complex. J. R. Meyer (1968) believes that location theory “synthesize{s} elements from almost all other places of economic analysis.” This study expects to make a contribution in that direction. Furthermore, the results of the study could be potentially helpful to economic development agencies as they plan their strategies to attract such investments to their regions/states. Lastly, the 1990s saw many significant changes in North America. Starting with the U.S.-Canadian Free Trade Agreement in 1989 and NAFTA in 1994, North American markets have become more closely integrated. The question is, where to next? It is interesting to note in this context that questions such as these have often kept the discussion on Canadian-U.S. relations alive with policy studies, particularly in Canada. For example, the possibility of free trade between Canada
Canadian Investment in the United States
197
and the U.S. raised questions in Canada about the fear of both the losses of U.S. investment inflows and the loss of Canadian identity. In regions such as the St. Lawrence borderland of Northern New York/Vermont the question was being asked with some anxiety by local development agencies and policy makers. If tariffs between Canada and the United States had already benefited the St. Lawrence region from the inflow of Canadian investment due to its proximity to the rich Canadian market, would free trade alter this relationship? Similarly, the onset of NAFTA raised serious concerns in both Canada and the United States of losing businesses to Mexico, especially manufacturing, because of cheap labor. If that happened, would regions such as Western New York or Northern New York/Vermont lose if Canadians decided to pack up and either go home or move operations to the “warmer” climate of Mexico? Now there is a growing discussion among Canadians to push trade integration further, especially between the U.S. and Canada, with monetary integration or development of a single currency. The third objective of this study is to determine Canadian companies management’s impressions about the appropriateness of this action or of developing larger integration with Europe or South America. How preferred would greater integration schemes be for their operations?
RESEARCH METHODOLOGY A survey questionnaire was mailed to 243 Canadian companies in Northern New York/Vermont, Western New York, Ohio, and Washington State from lists developed in conjunction with the Canadian Consulates, state departments of Commerce, and economic development agencies in each state. Some 129 individual pieces of data were requested of firms. A completed questionnaire was received from 69 companies, resulting in a little over 28% response rate. The questionnaire provided quantitative data about the three main concerns of this study, in addition to other useful information about the make-up of the Canadian companies and their impact on the four regions/states under consideration. It should be noted that the term “region” is used to denote a multi-jurisdictional area that may encompass a portion of a state or more than one state. The rationale for this determination was due in part to the availability of segmented data and in part due to the geographic and economic uniqueness of the regions selected.
THEORETICAL FOUNDATIONS Why do countries/companies invest abroad? The earliest explanation was provided by economists in terms of the differences in the marginal productivity
198
PREM GANDHI AND WAYNE GLASS
of capital. That is, countries with large stock of capital find additional stock to be less productive at home relative to the countries with a low stock of capital. Thus, capital-rich (industrialized) countries will find it profitable to invest abroad in capital-poor (developing) countries in order to maximize their overall returns. Unfortunately, this explanation neither explained the continuous shortage of capital in the developing countries nor did it explain the fact that industrialized countries were increasingly investing in each other, especially since the end of WWII. In addition, industrialized countries were not only witnessing increasing investments, they were also benefiting from increasing trade with each other. In response to this new reality, several economists provided alternative explanations to this apparent contradiction. Stephen Hymer (1976) discussed the oligopolistic nature of companies while Vernon (1966) talked about the product life cycle theory as possible explanations for increasing trade and investment among the industrialized countries. Dunning (1988) brought these disparate explanations together through his emphasis on the “internalization” process by which companies make best use of their certain advantages. His Eclectic Paradigm suggests the reasons for a firm to engage in foreign direct investment due to certain advantages (e.g. ownership advantage, advantage to “internalize,” and locational advantage). Locational advantage, in turn, may depend upon an area’s economic, social and political factors (Eden, 1993). Such advantages may include the quantity and quality of the area’s factor endowments, accessibility to a wider market, the overall availability of a quality infrastructure, a region’s receptivity to foreign firms, the facility to conduct businesses between a subsidiary and the head office, as well as the access to any tax advantages or tax differentials, etc. A further consideration to FDI by any company will depend upon the nature of the company’s operations, e.g. whether the company is resource-oriented, market-oriented, or a foot-loose one. Accessibility to markets is becoming a major consideration for manufacturing firms, which use their firm-specific knowledge by internalizing that uniqueness. Empirical studies on foreign direct investment have thus looked at the attractiveness of the host countries by pulling such investment to their economies. The present study uses a variety of pull factors for the attractiveness of the host country (the U.S.) for foreign investment (Canada). With the increasing flow of Canadian investment in the U.S. in the 1990s, there has been growing number of studies of the phenomenon. For example, in 1999 alone, Canadian companies invested $5.6 billion in the U.S., an increase of 8% over the previous year. This investment went mostly to the manufacturing, finance and insurance industries (Bargas, 2000). Meyer and Green (1996), O’Hagan and Anderson (2000), Rao, Legault and Ahmad (1994), and Knubley, Krause and Sadeque (1991) have analyzed the phenomenon of Canadian investment abroad. Meyer and Green look at the “location and functional biases of Canadian FDI
Canadian Investment in the United States
199
activity, from the perspectives of both the ‘point of control’ and the ‘point of investment,’ and then offer some explanation for the spatial trends realized.” Through a stepwise regression procedure, they determine pull factors rather than push factors, concluding that “market, trade with (Canada), labor and crime conditions were critical place-specific criteria that affected the spatial decision of Canadian direct investors.” O’Hagan and Anderson (2000) examine the distribution of Canadian investment across states between 1974 and 1994 to extend the discussion of pull factors as attractions for Canadian firms to invest in the U.S. With the use of a multinominal logit model they determined that the Canadian investment is attracted to states with large markets, good accessibility to international inputs and markets, skilled labor, and expanding work forces. At the same time, distance to the Canadian source location, and corporate tax rates have negative impacts on investment location choice (Myers, 1999).
SCOPE AND FINDINGS OF THE SURVEY One of the difficulties encountered in the survey was to reconcile the various lists of Canadian companies received from the variety of sources. Duplication of names, addresses and facilities of Canadian companies and their U.S. affiliates created additional problems. McPherson (1999) described similar concerns in a previous study. For example, Canadian companies are located throughout New York State. Since the focus of the study was on Northern and Western New York, the questionnaire was mailed only to those companies that had addresses in those regions. Companies with more than one facility either in the same state, or in other states, were sent only one questionnaire. As a result, the study identified 116 companies located in New York State regions under study, with the majority located in Northern New York, concentrated around Plattsburgh in Clinton County. Similarly, 98 companies were identified in Ohio (compared to 118 noted by Ohio’s Department of Development), fourteen in Vermont and in Washington State, resulting in a grand total of 243 usable companies for this study. Of the 69 returned questionnaires, 39 came from Northern New York/Vermont (42%), twelve from Western New York (31%), fourteen from Ohio (14%), and four from Washington state (27%), giving an overall return rate of over 28%. See McPherson (1999) for comparison to previous survey response rates. The questions in the survey centered around four areas of investigation; profile of the Canadian company in the U.S., attractiveness of the U.S./regional location, satisfaction with the location decision, and the attitude towards further integration between Canada and the U.S. Table 1 provides a profile of the companies whose
200
PREM GANDHI AND WAYNE GLASS
Table 1. Properties of Canadian Companies Investing in the United States (Percent Responding). Characteristics
Total
NNY-VT
WNY
OH
WASH
Type of company Mfg/assembly Warehousing Sales/service Construction Other
54 14 11 5 9
69 11 11 3 11
36 18 9 – 45
42 25 17 8 8
– 25 – 25 50
Location of parent company Ontario Quebec Alberta British Columbia NA
42 38 3 1 16
29 60 3 – 8
82 9 – – 9
91 9 – – –
25 25 25 25 –
23 26 17 6 6 20
36 18 9 9 – 9
33 42 8 – – 18
– 25 25 – – 50
9 91
25 75
25 75
Value of company’s assets (dollars) <1 million 31 1–3 million 28 3–6 million 15 6–9 million 5 9–15 million 5 >15 million 16
Political or economic conditions in Canada led to locate in U.S. Yes 29 39 No 71 60 Source: Survey Questionnaire.
responses were used. All except for three companies have parent organizations in Canada. Given the centrality of Ontario to the economy of Canada, it is not surprising that the majority of the headquarters are located in that province. Some of the Canadian affiliates in the U.S. were located prior to the 1950s (one as early as 1868), though the majority of them established their U.S. location within the last 20 years. Sixty percent of the respondents had no U.S. presence prior to their establishment in the U.S. The majority of the companies either exported directly from Canada or used a sales representative in the U.S. Of the remaining firms, 31% had another location in the U.S. prior to establishing their location in the regions under study. As one would suspect, the majority of the respondents are in manufacturing/assembling (51%) and are located in New York State and Ohio, both of which have traditionally been in the manufacturing hub of the United States. It is also not surprising to find 14% of the respondents have a warehousing facility,
Canadian Investment in the United States
201
a phenomenon common among firms trying to test the American market prior to locating their full operation. Those who responded in the “other” category (19%) included a diverse mixture of engineering, engineering designers, business development, consulting, customs brokers, investment bankers, and other miscellaneous industries. Given the types of companies, it is not surprising that 31% had assets of less than a million dollar (20% had assets less than $500,000). For the firms in “other” category, their main assets were in human capital and less in equipment and machinery.
REASONS FOR LOCATING IN THE U.S. Since the main objective of the study was to find the reasons for the Canadian companies to invest in the United States, the questions in this area pertained to four major categories: regional/local considerations, which include those factors that give the region/state/local area an absolute advantage over other areas in the U.S. (that is, factors that are unique to a give area); market considerations, which have been given top billing among motives to invest abroad in the literature (that is factors that account for the survival-and-growth strategy of the companies); cost considerations, which give areas a comparative advantage; and economic considerations that would pull Canadian companies to the United States, but not necessarily to a particular area. In other words, the list of questions contained all factors that pull Canadian companies to the U.S. as well as those that attract them to a given location. Tables 2 and 3 summarize the results of this portion of the survey. At the macro level, Canadian companies ranked the four most important reasons for investing in the United States as: potential for growth in the U.S. market (80%); proximity to the parent organization in Canada (69%); proximity to customers (54%); access to highways (53%). That is, at least 50% of the respondents noted these factors as their main reasons to invest. At the micro level, the primary reason for choosing a particular state (or region) for location was: (1) proximity to the Canadian operation; and (2) access to highways. This is not surprising since the majority of respondents were from the New York/Vermont where excellent north-south highways provided the shortest driving distance to a parent organization located in Ontario and Quebec. It also makes sense when one considers the fact that most of the companies in the area were small and/or family-run businesses, and in such cases it is easier for the owner(s)/managers to commute back and forth between the parent organization and its subsidiary. Proximity may also be critical in a local venture’s initial stages, when it has to depend upon the parent company for clerical and managerial
202
PREM GANDHI AND WAYNE GLASS
Table 2. Motives for Canadian Companies to Invest in the U.S.: A Regional Perspective (Percent Responding). Factors
State/regional considerations Proximity to Canada Local incentives Public infrastructure Important materials Cultural amenities Community attitude toward business Educational resources
Total Sample (vi/mi/ni∗ )
NNY-VT (vi/mi/ni)
WNY (vi/mi/ni)
OH (vi/mi/ni)
WASH (vi/mi/ni)
69/22/9 35/39/26 30/30/40 26/26/48 2/31/67 45/45/10
71/23/6 37/34/29 31/37/32 34/23/43 3/34/63 46/40/14
63/9/27 18/27/54 9/–/91 18/9/73 –/9/91 9/45/45
50/25/25 25/42/33 33/17/50 8/42/50 8/3/58 42/42/17
50/25/25 25/25/50 25/25/50 –/25/75 50/25/25 75/25/–
14/53/34
14/51/35
27/9/64
8/58/34
–/100/–
Market considerations Potential for growth in the United States Proximity to customers Proximity to suppliers Export facilities
80/16/4
71/17/12
55/27/16
67/17/17
100/–/–
54/27/17 29/45/26 23/36/41
57/26/17 29/46/25 29/31/40
36/18/46 18/18/64 18/36/46
58/17/25 33/42/25 33/17/50
100/–/– –/75/25 –/75/25
Transportation Highways Railways Airways Seaways
54/43/3 5/30/65 19/30/51 6/15/81
51/40/9 9/29/62 6/32/62 3/14/83
18/63/18 9/–/91 9/27/63 –/–/100
67/17/17 –/42/58 33/33/33 –/33/67
50/50/– 25/25/50 100/–/– 25/–/75
34/49/17
18/45/37
50/25/25
75/25/–
43/43/14 40/29/31 46/29/25 31/46/23
9/27/63 18/18/64 –/27/73 –/27/73
50/25/25 42/42/17 25/50/25 8/58/33
75/25/– 25/25/50 –/50/50 25/50/25
24/54/22
23/57/20
9/55/36
25/42/33
75/–/25
17/30/54 22/50/28
17/23/60 20/51/29
9/27/64 9/27/64
8/17/75 17/50/33
–/75/25 25/50/25
Production and cost considerations Availability of skilled 39/48/13 labor Trainable labor 44/42/14 Existing industrial base 40/36/24 Low electric rates 33/40/28 State taxes 22/51/27 Economic considerations Banking & financial facilities Export facilities State incentives
Note: Symbols: vi = very important; mi = moderately important; ni = not important. Importance in terms of the percentage of respondents considering that factor, i.e. at least 50% of the respondents considering that factor as important or not important. Source: Survey Questionnaire.
Canadian Investment in the United States
203
Table 3. Most Important and Least Important Factors for Locating in the U.S.: A Regional Perspective (Percent Responding). Factors importanta
Total Sample
NNY-VT
WNY
OH
WASH
vi/mi/nib
Most Growth potential Proximity to HQ Access to highways Proximity to customers
80/10/4 69/22/9 54/43/3 54/27/19
vi/mi/ni 71/17/12 71/23/6 51/40/9 57/26/17
vi/mi/ni 55/27/11 63/9/27 18/63/18 36/18/46
vi/mi/ni 67/17/17 50/25/25 67/17/17 58/17/25
vi/mi/ni 100/–/– 50/25/25 50/50/– 100/–/–
Least importanta Seaways Cultural amenities Railways Export facilities Airways
ni/mi/vi 81/15/6 67/31/2 65/30/5 54/30/17 51/30/19
ni/mi/vi 83/14/3 63/34/3 62/29/9 60/23/17 62/32/6
ni/mi/vi 100/0/0 91/9/– 91/–/9 64/27/9 63/27/9
ni/mi/vi 67/33/– 58/33/8 58/42/– 75/17/8 33/33/33
ni/mi/vi 75/–/25 25/25/50 50/25/25 25/75/– –/–/100
Source: Survey Questionnaire. a Importance in terms of the percentage of respondents considering that factor, i.e. at least 50% of the respondents considering that factor as important or not important. b Symbols: vi = very important; mi = moderately important; ni = not important.
personnel, product design, and investment funds. The parent company may also like to have the branch nearby when flexibility in production runs and/or design change is critical, again in the initial stages of production. The availability and easy access to warehousing facilities in the four U.S. regions make the above point more understandable since the Canadian companies are able to use the location as a distribution center to service the American market. Regional economists have singled out easy access to a good transportation network in the U.S., especially highways, as an important factor in attracting business investment. Trucking is the most economical way of shipping goods over distances of five hundred miles or less. Highways not only help move resources and output in and out of the region in the most economical fashion, they also facilitate commuting between headquarters in Canada and the subsidiaries in the U.S. The most important reasons for Canadian companies to invest in the U.S. are market considerations (potential for growth, proximity to customers, and access to highways), and local considerations (proximity to the parent organization in Canada). Of all the other important local considerations, community attitude towards business, especially towards foreign business, received a relatively strong positive response. Several other points deserve attention. Major efforts are made in the United States by regional, state, and local agencies to attract businesses to their location through advertising or personal visitations. In many cases, they emphasize such
204
PREM GANDHI AND WAYNE GLASS
factors as quality of life, availability of public infrastructure, availability of labor, and trainable labor at reasonable cost in marketing the region. Development agencies provide potential businesses a variety of incentives such as rent-free space, property tax rebates, job training rebates, tax abatements, and other financial considerations to encourage them to invest in their respective regions/states. Respondents were asked about the effectiveness of these strategies and to rate various advertising sources in their location decisions. Very few respondents indicated that advertisements in general (including U.S. and Canadian media less than 4%) and trade journals, mailed material, and the Internet were not particularly important (about 8%) to their decisionmaking. Visitations or a personal contact by a state representative was more important (25%) but less than a similar visit by a local representative of a community (36%). However, the effectiveness of the incentives was questionable. In the costbenefit analysis, Canadian companies consider them as sweeteners, and no more. Asked specifically, state incentives were important to only 22% of the respondents, and 93% would still have moved to the same state without such incentives. Similarly, 35% of the respondents found local incentive as important, but without local incentives, nearly 87% would have still moved to the same area within the state. What was more important, as mentioned earlier, was the community attitude towards business (45%). But the most surprising result of the survey has to do with the relatively small importance attached to the availability of labor. Regional economists have mentioned this as a critical factor influencing the choice of location. The empirical evidence on the issue is mixed. Litvak and Maule (1981) have pointed to the importance of favorable labor costs (15–40% lower in the United States) and less labor unrest as inducements for Canadian companies to locate in the U.S. Forget and Denis (1985), on the other hand, found that American labor costs and other cost factors were not a major element in Canadian investment decisions. Respondents in the current survey reported labor supply (trainable as well as skilled) to be not an important factor for them in choosing their location. However, more than 70% of the firms found the local workforce met or exceeded their expectations. Only 31% found the workforce to be less available or qualified than expected. Such a finding may be comforting to those concerned with the economy of the studied regions in the sense that such companies would not be attracted to the cheaper labor in the offshore countries of Asia and Latin America. How does the study findings compare with earlier studies of the same nature? Table 4 summarizes the results of most important reasons for Canadian investment in the U.S. and compares these with the findings of previous studies. The most important reasons in each study are indicated in agreement by “yes,” those
Reasons
Present
Gandhi
McPherson and McConnell
Meyer and Green
O’Hagan and Anderson
Forget and Daniel
Matheson
Kasoff et al.
Potential for growth in U.S. Proximity to customers Proximity to Canada Accessibility highways Community attitude Availability of labor (cost) Trade barriers Year published
Yes Yes Yes Yes N/a N/a No 1991
Yes Yes Yes Yes N/a N/a No 1990
Yes Yes Yes No No No N/a 1996
Yes Yes Yes No No Yes Yes 1996
Yes Yes Yes Yes No Yes No 2000
Yes Yes No No No N/a No 1985
Yes Yes No No No No No 1985
Yes Yes Yes Yes No Yes N/a 1997
Canadian Investment in the United States
Table 4. Consistency of Results with Previous Studies.
205
206
PREM GANDHI AND WAYNE GLASS
that do not agree are indicated by “no,” and “n/a” indicates factors that are not comparable.
REACTIONS TO FURTHER INTEGRATION The last part of the survey had to do with the Canadian companies’ management’s responses to further integration between Canada and the United States and their reactions to the effect it would have on their operations. Table 5 summarizes the companies’ responses to these questions. Respondents were asked to rate on a 1–5 scale the extent the Canadian company favors further integration between United States and Canada as well as towards the creation of a Free Trade Zone of the Americas or a Common Currency Area. It is clear from the survey that Canadian companies care deeply about further liberalization of international trade and are not reluctant to see further integration between Canada and the United States on a line similar to the European Community. However, that support, though still high, declines significantly for either the political integration between the two countries or for creating a common currency zone like the Euro zone. In a way, their opposition is stronger for any political integration between Canada and the United States than for either creating a Free Trade of the Americas or establishing a Common Currency Area.
Table 5. Company Attitude Toward Further Integration (Percent Responding). Possibilities
Strongly Favors
Favors
Moderately Opposes
More free trade Canada/U.S./EEC
64
31
5
–
–
Complete free trade Canada and U.S.
65
30
5
–
–
Common market type Canada and U.S.
55
27
15
3
–
Political integration Canada and U.S.
25
14
25
20
15
Common currency Canada and U.S.
25
28
23
8
13
Free trade of the Americas
27
29
22
14
8
Source: Survey Questionnaire.
Opposes
Strongly Opposes
Canadian Investment in the United States
207
Table 6. Effects on Canadian Companies’ Operations Under Expanded Integration (Percent Responding). Effects No change Expand somewhat Expand significantly Open additional facilities in the U.S. Consolidate operations with Canadian parent Close U.S. facility Downsize Do not know (negatively stated) Close Canadian facility
Yes
No
30 30 17 30 13 3 2 8 –
70 70 83 70 87 97 98 92 100
Source: Survey Questionnaire.
Table 6 summarizes the effects that expanded market integration between Canada and the U.S. would have on their operations, as perceived by company management. If anything is certain, it seems that the Canadian companies know very well the effects such steps would have on their operations both in the United States as well as in Canada. In all certainty, they will not close their Canadian operations. Neither would they close their U.S. operations. For the most part, their responses seem to suggest a status quo. One conclusion that can be drawn from these responses is that economic development agencies in the four regions should not worry about Canadian operations in their region. They are likely to stay. If the literature on economic integration is any guide, the regions would continue to attract Canadian investment due to the dynamic effects of rationalization leading to specialization of product lines, establishment of product niche in different plants, product life cycle, and creating new product lines etc. To some extent this has been going on at present.
CONCLUSIONS While more research can be undertaken on both the data collected from this study and additional studies of regional location decision making, a number of interesting conclusions did result from this study. First, regional economic development organizations should recognize that some of the suppositions they have been using to design industrial recruitment might not be as applicable to Canadian firms. These same suppositions have an impact on the retention of existing Canadian firms in a region. Recruitment efforts would best be focused on firms with headquarters within a reasonable drive of
208
PREM GANDHI AND WAYNE GLASS
the U.S. community and would best be conducted by communities with efficient highway links to Canada. When considering greater economic, currency, or political integration between Canada and the U.S., policy makers would do well to include those Canadian firms with U.S. locations in the discussion. The firm’s views are based on recognition of the economic realities of doing business across borders and the impact (or lack thereof) that any movement to increased integration would have on their operations and the communities in which they are located.
REFERENCES Bargas, S. (2000). Survey of Current Business, 80(7), 58–69. Drennen, C. (1997). State concentration of Canadian direct investment in the United States. Presented at the 14th biennial meeting of the Association of Canadian Studies in the U.S. (ACSUS). Minneapolis. Dunning, J. H. (1988). Explaining international production. London: Unwin Hyman. Eden, L. (1993). Bringing the firm back in: Multinationals, international political economy. In: L. Eden & E. Potter (Eds), Multinationals in the Global Economy. Calgary: University of Calgary Press. Forget, C. E., & Davis, D. (1985). Canadian foreign direct investment in the United States: Reasons and consequences. In: A. M. Rugman (1987), Outward Bound: Canada Direct Investment in the United States (p. 37). Toronto: Canadian American Committee. Gandhi, P. P. (1991). St. Lawrence borderlands: The free trade agreement and Canadian investment in New York and Vermont. In: R. Lecker (Ed.), Borderlands: Essays in Canadian-American Relations (pp. 194–221). Toronto: ESW Press. Government of Canada (2002). United States-Canada: The world’s largest trading relationship. Ottawa: Department of Foreign Affairs and International Trade. Hymer, S. (1976). The international operations of national firms: A study of direct foreign investment. Cambridge, MA: MIT Press. Kasoff, M. J., Benedict, M. E., & Lauer, S. (1997). Canadian direct investment in the United States: An Ohio perspective. Canadian Journal of Administrative Sciences, 14(2), 178–187. Knubley, J., Krause, W., & Sadeque, Z. (1991). Canadian acquisitions abroad: Patterns and motivation. In: L. Waverman (Ed.), Corporate Globalization Trough Mergers and Acquisitions. Calgary: University of Calgary Press. Litvak, I. A., & Maule, C. J. (1981). The Canadian multinationals. Toronto: Butterworths. Matheson, N. (1985). Canadian investment abroad. In: A. M. Rugman (Ed.), Outward Bound: Canadian Direct Investment in the United States (p. 31). Toronto: Canada-American Committee. McConnell, J. E., & MacPherson, A. D. (1990). Canadian establishments in Western New York: Some preliminary findings. Canadian Journal of Regional Science, 12(Summer/Autumn), 189–204. McPherson, A. D. (1999). Canadian direct investment in Western New York: A tracking study of subsidiaries and parent companies. Canadian Journal of Regional Science, 20(3), 329–340. Meyer, J. R. (1968). Regional economics: A survey. In: L. Needleman (Ed.), Regional Analysis (pp. 19–60). Baltimore: Penguin Books.
Canadian Investment in the United States
209
Meyer, S. P., & Green, M. B. (1996). Foreign direct investment from Canada: An overview. Canadian Geographer, 40(3), 219–237. Myers, J. (1999). Money talks when money walks: Foreign direct investment in Canada is drying up. Canadian Business, 72(1), 88. O’Hagan, S., & Anderson, W. P. (2000, Summer). Canadian foreign direct investment in the U.S.: A discrete choice analysis approach. Canadian Journal of Regional Science, 23(2), 213–231. Rao, S., Legault, M., & Ahmed, A. (1994). Canadian-based multinationals: An analysis of activities and performance. In: S. Globerman (Ed.), Canadian Based Multinationals (p. 106). Calgary: University of Calgary. Richardson, K. (1997). Whatcom County, Washington and the Greater Vancouver Region – What is so enticing over the border? Presented at the 14th Biennial Meeting of ACSUS. Minneapolis. Rugman, A. (1980). Multinationals in Canada: Theory, performance and economic impact. Boston: Martinus Nijhoff. Rugman, A. (1985). Megafirms: Strategies for Canada’s multinationals. Toronto: Methuen/Nelson. Rugman, A. (1987). Outward bound: Canada direct investment in the United States. Toronto: Canadian American Committee. Rugman, A. (1990). Multinationals and Canada-United States free trade. Columbia: University of South Carolina. Vernon, R. (1966). International investment and international trade in the product life cycle. Quarterly Journal of Economics (May), 190–207.
PART III: NAFTA AT AGE TEN: SECTORAL STUDIES AND RELATED ISSUES
CANADA-U.S. AGRICULTURAL TRADE Andrew Schmitz and Hartley Furtan ABSTRACT The U.S. 2002 Farm Bill provides sizeable direct and indirect subsidies to U.S. farmers, which has created increased competition in markets where the United States and Canada compete. Target prices were reintroduced and the overall level of U.S. Government support was increased. Canadian farmers will find it more difficult to compete in grains, oilseeds, and pulses. Government support in Canada for these crops is significantly below U.S. support. Canada and the United States have a significant two-way trade in agricultural products, including beef and pork. The outbreak of Mad Cow Disease in Canada in 2003 clearly illustrates the need for cooperation between the two countries.
INTRODUCTION In 2002, the United States passed a new farm program that provides sizeable direct and indirect subsidies to U.S. farmers. As a result, the new U.S. farm program created increased competition in markets where the United States and Canada compete. U.S. farm subsidies are generally not decoupled from production; hence they are both production and trade distorting. Added U.S. production, due to subsidies, will result in a loss of market share for Canadian products. As of July 2003, Canada had not increased subsidies to Canadian farmers in response to the U.S. farm program. North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 213–235 © 2004 Published by Elsevier Ltd. ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10012-0
213
214
ANDREW SCHMITZ AND HARTLEY FURTAN
Canada and the United States have a significant two-way trade in agricultural products. For example, the United States is an importer of large quantities of Canadian beef and pork while Canada imports large amounts of U.S. fruits and vegetables. The United States is the largest market for Canadian agricultural exports; hence the magnitude of trade barriers between the two countries has important consequences for the economic well-being of agriculture. In this paper, we highlight the importance of agricultural trade between the United States and Canada and the key elements of both the U.S. and Canadian farm programs. Data on producer-support levels suggest that for several of the agricultural sectors, government farm support in the United States is much greater than in Canada. We also discuss many of the border disputes that have arisen between the two countries and conclude with comments on the likely impact of the 2002/2003 U.S. Farm Bill on Canada-U.S. trade.
OVERVIEW OF CANADA-U.S. AGRICULTURAL TRADE U.S. agricultural exports to Canada have increased over time, reaching U.S. $8 billion in 2001 (Fig. 1). Likewise, Canadian exports to the United States have increased significantly over time, exceeding U.S. $10 billion in 2001 (Fig. 1). Canadian agricultural exports to the United States exceed U.S. exports to Canada by over U.S. $2 billion. The largest increase in exports for both countries has been in bulk commodities with relatively little export expansion in produce and horticultural products, processed intermediates, and high-value processed products. U.S. dairy exports to Canada rose sharply between 1997 and 2000 (Fig. 2) while Canadian exports to the United States increased sharply between 1994
Fig. 1. Canada-U.S. Agricultural Exports, 1989–2001. Source: IBAT (2002).
Canada-U.S. Agricultural Trade
215
Fig. 2. Canada-U.S. Dairy Exports, 1989–2001. Source: IBAT (2002).
and 1999. Since 1999, however, average Canadian exports to the United States have fallen dramatically and U.S. dairy exports to Canada since 2000 have fallen sharply. Historically, Canada’s exports of grain to the United States has exceeded imports, although in 2002, this was no longer the case when U.S. corn exports to Canada far exceeded U.S. imports of Canadian wheat (Fig. 3). Even so, U.S. trade-dispute actions against Canadian wheat shipments are far greater in number than are Canadian trade actions against corn imports from the United States.
Fig. 3. Canada-U.S. Corn and Wheat Trade, 1990–2002. Source: Industry Canada (2002).
216
ANDREW SCHMITZ AND HARTLEY FURTAN
THE 2002/2003 U.S. FARM PROGRAM The following are key provisions of the U.S. Farm Security and Rural Investment Act of 2002/2003 (FSRI): Income support for wheat, feed grains, upland cotton, rice, and oilseeds is provided through three programs: direct payments, countercyclical payments, and marketing loans with loan deficiency payments (LDP). Target prices were reinstated along with the associated countercyclical payments – a major change from the 1995/1996 Farm Bill. Support for peanuts was changed from a price-support program with marketing quotas to marketing loans, countercyclical payments, direct payments, and a quota buyout. Sugar is to operate as a no-net program. The nonrecourse loan program was reauthorized at U.S. $0.18 per pound for raw cane sugar and U.S. $0.229 per pound for refined beet sugar. For sugar, nonrecourse loans are extended to in-process beets and cane syrups. Both marketing assessments on sugar and forfeiture penalties were terminated. The nonrecourse sugar-loan program was reauthorized. The producer payment in kind (PIK) program continues and the U.S. Secretary of Agriculture can exchange Commodity Credit Corporation (CCC)-owned sugar for reductions in sugar-crop acreage plantings. Federal milk marketing orders continue. The minimum support price for milk was fixed at U.S. $9.90 per hundredweight (cwt) for milk containing 3.6% butterfat. In addition, a national Dairy Market Loss Payment (DMLP) program was established. The Dairy Export Incentive Program (DEIP) was extended to 2007. There were no changes in the basic crop insurance program. The Agricultural Risk Protection Act of 2000 (ARPA) provided an additional U.S. $8.2 billion for insurance-premium subsidies for fiscal years 2001 through 2005.1 The ARPA provision, scheduled to go into effect in 2006, allows for selection of continuous insurance levels rather than coverage levels at fixed intervals. Under FSRI, however, the ARPA was eliminated. (Crop insurance costs increased from U.S. $748 million in 1998 to U.S. $4.4 billion in 2002. Thus future possible payouts may require additional emergency funding.) Country-of-origin labelling (COOL) requirements were introduced. Under FSRI, retailers are required to inform consumers of the country of origin for covered commodities at their final point of sale (muscle cuts of beef, lamb, and pork; ground beef, ground lamb, and ground pork; farm-raised fish and shellfish; wild
Canada-U.S. Agricultural Trade
217
fish and shellfish; peanuts; and perishable agricultural commodities, such as fresh fruits and vegetables). Funding for the Environmental Quality Incentives Program (EQIP) was increased. A new conservation security program was introduced. Land-retirement programs were expanded, particularly for wetlands. Funding was expanded for farmland protection, and a new Grassland Reserve was created. The maximum acreage enrolled in the Conservation Reserve Program (CRP) was increased from 36.4 million acres to 39.2 million acres. Under the Wetlands Reserve Program (WRP), the maximum acreage was increased from 1.08 million acres to 2.8 million acres. New trade programs were authorized, including the McGovern-Dole International Food for Education and Nutrition Program, the Biotechnology and Agricultural Trade Program, the Technical Assistance for Specialty Crops Program, and the Exporter Assistance Initiative Program. The Public Law 480 Program (PL480) was reauthorized. The Export Enhancement Program (EEP) extended funding through 2007 at the current funding level of U.S. $478 million per year. Under the FSRI, the definition of unfair trade practices was expanded to include: (1) practices of state-trading enterprises that “are not consistent with sound commercial practices conducted in the ordinary course of trade;” (2) subsidies that decrease market opportunities for U.S. exports or unfairly distort agricultural markets to the detriment of the United States; and (3) unjustified trade restrictions or commercial requirements, such as labelling, that affect new technologies, including biotechnology.
Fig. 4. Canada-U.S. Fresh and Frozen Pork Exports, 1989–2001. Source: IBAT (2002).
218
ANDREW SCHMITZ AND HARTLEY FURTAN
Fig. 5. Canada-U.S. Live Pork Exports, 1989–2001. Source: IBAT (2002).
A sizeable trade exists between Canada and the United States in pork and beef. Little wonder there is considerable interest on the effect that COOL will have on the meat trade. Canadian exports of pork (fresh, frozen, and live) to the United States far exceed Canadian imports of like products (Figs 4 and 5). In 2001, Canadian exports of fresh, frozen, and live pork exceeded imports by roughly U.S. $800 million. Likewise for beef (fresh, frozen, and live), Canadian exports far exceeded Canadian imports (Figs 6 and 7). In 2001, total Canadian beef exports exceeded imports by roughly U.S. $1.6 billion.
Fig. 6. Canada-U.S. Fresh and Frozen Beef Exports, 1989–2001. Source: IBAT (2002).
Canada-U.S. Agricultural Trade
219
Fig. 7. Canada-U.S. Live Cattle Exports, 1989–2001. Source: IBAT (2002).
Under the 2002/2003 U.S. Farm Bill, loan rates for grains increased, especially for oats, sorghum, and wheat (Table 1); however, target prices under the 1991 Farm Bill were generally higher than in the 2002/2003 FSRI (Table 2). Target prices were eliminated in the 1996 Farm Bill program. FSRI includes oilseed crops (sunflower, flaxseed, canola, and mustard seed) and pulse crops that were not in the 1995/1996 Farm Bill (e.g. small chickpeas, lentils, and dry peas). The 2002/2003 Farm Bill, unlike previous U.S. Farm Bills, established marketing, loans and LDPs for these pulse crops (Table 3).
Table 1. U.S. Agricultural Commodity Loan Rates (1991 and 2002). Crop
1991
2002
Percentage Change
Corn ($/bushel) Sorghum ($/bushel) Wheat ($/bushel) Upland cotton ($/pound) Rice ($/hundredweight) Barley ($/bushel) Oats ($/bushel) Soybeans ($/bushel) Minor oilseeds ($/pound)
1.89 1.69 2.58 0.519 6.50 1.71 1.14 5.26 0.10
1.98 1.98 2.80 0.52 6.50 1.88 1.35 5.00 0.096
4.8 17.2 8.5 0.2 0.0 9.9 18.4 −4.9 3.2
Source: Babcock (2002).
220
ANDREW SCHMITZ AND HARTLEY FURTAN
Table 2. U.S. Agricultural Commodity Target Prices. Crops
1991
2002–2003 (Estimated)
2004–2007 (Estimated)
Corn ($/bushel) Sorghum ($/bushel) Wheat ($/bushel) Upland cotton ($/pound) Rice ($/hundredweight) Barley ($/bushel) Oats ($/bushel) Soybeans ($/bushel) Minor oilseeds ($/pound) Peanuts ($/ton)
2.74 2.61 4.00 0.73 10.71 2.36 1.45
2.60 2.54 3.86 0.72 10.50 2.21 1.40 5.80 0.10 495.00
2.63 2.57 3.92 0.72 10.50 2.24 1.44 5.80 0.10 495.00
Source: USDA/ERS (2002).
Table 3. U.S. Oilseed and Pulse Crops: Loan Rates. Crops
2002 Farm Bill Rate
Oil-type sunflower ($/pound) Other-type sunflower ($/pound) Flaxseed ($/pound) Canola ($/pound) Rapeseed ($/pound) Safflower seed ($/pound) Mustard seed ($/pound) Soybeans ($/bushel) Small chickpeas ($/hundredweight) Lentils ($/hundredweight) Dry peas ($/hundredweight)
0.0915 0.1210 0.0698 0.0949 0.0947 0.1253 0.0988 5.00 7.56 11.94 6.33
Source: USDA/ERS (2002).
THE AGRICULTURAL ASSISTANCE ACT OF 2003 The U.S. Agricultural Assistance Act passed in February 2003 provided an additional U.S. $3.1 billion in government assistance to farmers above the provisions of the 2002/2003 Farm Bill: Crop Disaster Assistance: Provides disaster payments for crop losses for either 2001 or 2002, producers choose which year. It covers losses greater than 35% of the normal yield (i.e. 65% of the normal yield less the harvested yield). Quality losses are covered as well.
Canada-U.S. Agricultural Trade
221
Livestock Assistance: Expands eligibility for the 2002 Livestock Compensation Program (LCP), which was initiated by the U.S. Administration in the fall of 2002 to provide assistance for selected types of livestock in counties designated as disaster areas. The program re-establishes a more targeted Livestock Assistance Program (LAP) for producers suffering grazing losses in counties designated as disaster counties due to weather-related problems. Funding Assistance: U.S. funding assistance provided: (1) the government purchase of surplus fruits and vegetables and other commodities used by schools and food banks; (2) direct tobacco payments to producers; (3) losses to the cottonseed industry resulting from hurricanes and other conditions in 2002; and (4) disaster assistance to sugarcane producers for losses resulting from hurricanes in 2002 and for sugar-beet losses for either 2001 or 2002. Administration: Ensures that the reduction in CRP annual payments on CRP acres approved for haying and grazing in 2002 because of drought and other weather problems will be waived.
CANADIAN AGRICULTURAL PROGRAMS A wide range of programs has been used in Canada, many of which have been terminated. Some selected programs are described briefly below. Crop Insurance Act (CI) Passed in 1959, this was a voluntary production-guarantee program. Because the federal government had no constitutional authority to unilaterally establish a cropinsurance program, the CI was jointly managed by the federal and provincial governments. Under this arrangement, producers paid premiums that were matched by the federal government. Provincial governments paid the administrative costs. Producers and the federal government shared insurance costs until 1989 when federal-provincial cost-sharing changes were implemented, establishing a tripartite program paid for by producers, the provincial governments, and the federal government. These changes, which reduced federal government premium costs, were accompanied by changes to make the program more actuarially sound. Western Grain Stabilization Act (WGSA) Introduced in 1975, the WGSA provided a grain safety-net program that paid producers from the Prairie Region when net cash receipts from the seven major grains
222
ANDREW SCHMITZ AND HARTLEY FURTAN
fell below 90% of the preceding five-year average. WGSA allowed farmers to contribute a percentage of their gross sales to a stabilization fund. The Canadian government contributed to the fund by matching approximately 50% of the producer contribution. Payouts to the Prairie region equaled CDN $223 million, CDN $522 million, CDN $859 million, and CDN $1.4 billion for crop years 1983/1984 through 1986/1987.
Western Grain Transportation Act (WGTA) Enacted in 1983, the WGTA replaced the 1925 statutory “Crow’s Nest Pass Freight Rate” with a rail subsidy on the movement of grain and grain products from Prairie points to the ports of Vancouver, Thunder Bay, Prince Rupert, and Churchill. Government payments of about CDN $800 million per year were paid directly to the railways, which were obliged to charge subsidized freight rates for grain movement. When the WGTA was repealed on August 1, 1995, the freight subsidy was eliminated and producers were partially compensated with two lump-sum compensatory payments in 1996 called the Crow Benefit (Crow).
Special Canadian Grains Program (SCGP) Introduced in 1986, this program transferred CDN $1 billion from the federal government to Canadian grain and oilseed producers. The transfer was to offset the loss incurred by producers from the subsidy war between the United States and the European Union. An additional CDN $1.1 billion was paid out in 1987 as part of the same program.
Gross Revenue Insurance Program (GRIP) In 1991, the federal government entered into separate agreements with each provincial government to established GRIP as one component of the new safetynet program for farmers. It offered farmers revenue protection. Several billion dollars were paid out under GRIP to grain and oilseed producers. The premiums were shared by farmers, provinces, and the federal government under the tripartite system. The revenue guarantee for each crop was based on a percentage of the 15-year moving average of indexed past prices multiplied by crop yields. For calculating the revenue guarantee for a farmer average, yields were used
Canada-U.S. Agricultural Trade
223
or farmers could elect to use their own farm-coverage yields if those yields were above the area average. But GRIP had major problems and Saskatchewan withdrew from the program in 1995; Alberta and Manitoba withdrew in 1997.
Net Income Stabilization Account (NISA) In 1991, the federal government established NISA as a second component of the new safety-net program for farmers. NISA enabled farmers to set money aside, subject to an annual contribution cap, in a personal income stabilization account from which they could draw funding during difficult times. Farmers were permitted to set aside up to 2% of their gross sales in individual accounts, the sum of which would be matched equally by the federal and provincial governments. The program allowed a withdrawal from the farmer’s NISA account when the farmer’s gross margin fell below the five-year average or when his net income fell below CDN $10,000. Farmers cannot draw on their NISA account until sufficient funds have been accumulated, limiting the effectiveness of NISA as an income stabilization mechanism.
Agricultural Income Disaster Assistance (AIDA) In response to low farm incomes – particularly in the pork sector – the federal government introduced AIDA in 1998. When a producer’s net income (not including depreciation) falls below 70% of his three-year moving-average net income, he becomes eligible for an AIDA payout. (Net income levels below zero are not included in the averaging process.) Provinces that had disaster programs in place, like Quebec and Alberta, did not participate directly in AIDA. However, the federal government did transfer to those provinces their share of the total available support. The cost share on this program was 40% provincial and 60% federal. But AIDA had its drawbacks. The system of three-year moving averages resulted in payments to producers who experienced only one low-income year, but did not result in payments to producers who had low incomes in the three years immediately prior to the program’s inception in 1998. This program was sometimes costly in terms of the number of accountants and government employees needed to process an individual farmer’s AIDA application. In addition, AIDA may have discouraged diversification, as farmers who had beef cattle in combination with grain likely received little or no AIDA money, due to the high price of feeder calves throughout the 1998 to 2000 period.
224
ANDREW SCHMITZ AND HARTLEY FURTAN
Supply Management Supply management, which dates back to the late 1960s and early 1970s, covers such commodities as dairy, poultry, and eggs. For example, the Canadian Egg Marketing Agency was created in 1973, the Canadian Turkey Marketing Agency in 1974, and the Canadian Chicken Marketing Agency in 1978 (Schmitz, Furtan & Baylis, 2002). For supply-managed commodities, tight import quotas exist along with high external tariffs. In addition, producers control production.
NEW CANADIAN AGRICULTURE PROGRAMS In 2002, the Canadian government started a major review of all agricultural programs. It released a document called the Agricultural Policy Framework (APF), which listed five major areas of change: (1) risk management; (2) renewal; (3) food safety; (4) agricultural research; and (5) environmental issues. Each of these areas will require a new set of programs, which are yet to be developed. Any discussions of the supply-managed industries is what is missing from agricultural programs. In terms of direct subsidies to farmers, risk-management is extremely important. Programs include NISA, the current CI program, and the Canadian Farm Income Program (CFIP) that replaced AIDA. There is a new plan that will roll all of the above risk-management programs into one. Farmers will co-insure the margin of returns on their farms, made up of crop insurance and the existing NISA and CFIP
Fig. 8. Canada-U.S. Aggregate Producer Subsidy Equivalents (PSEs), 1986–2001. Source: OECD Database (2002).
Canada-U.S. Agricultural Trade
225
payments. For example, suppose a farm has CDN $100,000 of net margin. The farmer will be required to put up CDN $10,000 to fully insure the CDN $100,000 and to purchase crop insurance. The renewal plan is largely an adjustment program for farmers exiting agriculture, but at the same time there will be some attempt to assist new entrants. Food safety is a new emphasis for Canadian agriculture. In the past, Canada has produced a safe supply of food; however the recent concerns over bioterrorism have made the tractability of food an important issue. Environmental concerns – especially those of water quality – have become important to policymakers because of a few
Fig. 9. (a) U.S. Commodity Payments vs. Direct Support, 1986–2001. (b) Canadian Commodity Payments vs. Direct Support, 1986–2001. Source: OECD Database (2002).
226
ANDREW SCHMITZ AND HARTLEY FURTAN
Fig. 10. Canada-U.S. Individual Commodity Producer Subsidy Equivalents, 1986–2001. Source: OECD Database (2002).
serious water-contamination problems blamed on agricultural production. Finally, the focus on agricultural research is shifting from traditional commodities, such as wheat and canola, to more value-added and exotic crop and animal production.2
LEVELS OF PRODUCER SUPPORT From 1990 through 1995, government support for both Canadian and U.S. agriculture declined (Fig. 8). Support for U.S. agriculture increased beginning in 1995 Table 4. Average of Canadian Non-Tariff-Rate Quotas (TRQs) and Over-Quota TRQs, 2001. Product Wheat Corn Beef Pork Potatoes Soybeans Cheese Butter Eggs Chicken with spent fowl Chicken without spent fowl Source: AMAD database and AAFC estimates (2002).
Average Tariffs (%) 62.8 0.4 26.5 0.0 1.4 0.0 245.6 298.7 121.5 167.8 245.4
Canada-U.S. Agricultural Trade
227
Table 5. U.S. Agricultural Trade Issues and Concerns Under CUSTA, 1989–2001. Issue/Year
Commodity
Major U.S. Industry Concerns
U.S. International Trade Commission (USITC) 332 investigations 1990 Durum wheat Rising imports Assistance to Canadian producers under the WGTA, Western Grain Stabilization Act (WGSA), Canadian Wheat Board 1991 Dried alfalfa Competition in third-country export markets (Japan) Assistance to exports under the WGTA 1991 Fresh apples Marketing and grading practices Assistance under the Agricultural Stabilization Act (ASA), NTSP, and British Columbia Farm Income Insurance Program 1993 Dried peas and lentils Competition in third-country markets Producer assistance under the WGTA and GRIP 1993 Live cattle and beef Increasing live cattle imports Assistance under the NTSP 1997 Fresh and processed Canadian Federal, Provincial, and Municipal assistance potatoes programs effect on increasing U.S. imports 1992–1996 Restricted access for U.S. exports 1997 Live cattle and beef Impact of NAFTA and the Uruguay Round on U.S. imports and exports Countervailing duty/anti-dumping investigations 1989 Fresh chilled or frozen Alleged countervailable subsidies for swine pork
1999
Live cattle
Alleged countervailable subsidies Alleged selling below cost of production
Actions Taken
WGTA, WGSA discontinued and replaced with other assistance programs
WGTA replaced with other programs
NTSP, ASA replaced with other programs
WGTA and GRIP replaced with other programs NTSP replaced with other programs Potato issues included in 1998 ROU
Cattle issues included in 1998 ROU
USITC determined on second remand no material injury from subsidized imports DOC found no countervailable subsidies USITC found no material injury from dumped imports
228
ANDREW SCHMITZ AND HARTLEY FURTAN
Table 5. (Continued ) Issue/Year
Commodity
Major U.S. Industry Concerns
Actions Taken
1999
Sugar and syrups
1999
Live swine
Review of Anti-dumping (AD) order Review of AD order
2000
Red raspberries
Review of AD order
2001
Greenhouse tomatoes
Alleged selling at less than fair value
Order revoked, Oct. 1999 Order revoked, Nov. 1999 Subsidies found to be de minimus Order revoked, Jan. 2000 Preliminary affirmative determination
Sources: Taken in part from Jabara and Reeder (2002), which they compiled from reports of the U.S. International Trade Commission, USTR, GAO, and USDA.
and Canadian support increased beginning in 1997. It is interesting that for the 1986–1997 period, Canadian producer support was above that of the U.S. levels. After 1997, however, support for U.S. agriculture surpassed the support provided to Canadian farmers. When comparing the 2002/2003 U.S. Farm Bill with the new Canadian APF, there is likely to be a significant increase in the divergence between U.S. and Canadian support in future years in favor of U.S. producers. The data in Fig. 9(a) and (b) show the breakdown of government transfers, according to Market Price Support (MPS) and direct payments. There has been no significant decline in the MPS over time. There are significant differences between the government-support levels of the two countries (Fig. 10). The egg sector has much greater support in Canada than in the United States, which is also true for the beef and veal sectors. The support for milk is roughly the same in both countries. Wheat and oilseed crops are supported more heavily in the United States than they are in Canada, which is also true for the other grains (Table 3). Tariffs provide additional support for producers. In Canada, for example, the highest tariffs are levied on imports of cheese, butter, eggs, and chicken, which are supply-managed commodities (Table 4).
AGRICULTURAL TRADE DISPUTES Agricultural trade disputes between Canada and the United States have been numerous (Tables 5–7). Many of these cases were resolved in favor of the defendants. We highlight the tomato dispute between the United States and Canada, since it
Canada-U.S. Agricultural Trade
229
Table 6. U.S. Agricultural Trade Issues and Concerns Under CUSTA, 1989–2001. Issue/Year
Commodity
Major U.S. Industry Concerns
CUSTA/NAFTA binational dispute panels 1993 Wheat U.S. industry concerns that the CWB is selling durum wheat into the United States below cost High tariffs on subject 1996 Dairy, poultry, eggs barley, products possibly violate margarine Canada’s obligations under the CUSTA and NAFTA USDA/USTR actions 1995 Wheat
1997–1999
Sugar-containing products (SCPs)
2000
Wheat
NAFTA legislation requires end-use certificates on imports of commodities when Canada requires them
301 Petition alleging injury to U.S. wheat producers due to discriminatory pricing policy of CWB
Actions Taken
Panel upholds CWB acquisition price as the basis for any possible A/D action Panel upholds Canadian tariffs and TRQs
USDA implements end-use certificate requirements on imports of Canadian wheat U.S. allocates 59,250 mt (91.5%) of SCF TRQ to Canada Under 1998 ROU, the U.S. agreed to require export permits for an importer to claim in-quota tariff rate on Canadian origin SCPs USTR accepted petition Oct. 2000
Sources: Taken in part from Jabara and Reeder (2002), which they compiled from reports of the U.S. International Trade Commission, USTR, GAO and USDA.
illustrates a general conclusion that as a country’s exports grow rapidly, importers will generally challenge this growth in exports by imposing tariffs on those exports. A considerable legal effort has been expended over the import and export of fresh tomatoes between the two countries, due in part to the considerable exportation of tomatoes from Canada to the United States (Fig. 11). As recently as November 2001, the Canadian Tomato Trade Alliance (CTTA) filed a complaint initiating
230
ANDREW SCHMITZ AND HARTLEY FURTAN
Table 7. Canadian Agricultural Trade Issues and Concerns Under UCSTA. Issue/Year
Commodity
Major Canadian Industry Concerns
Countervailing duty/anti-dumping investigations 1989–2000 Red delicious apples Alleged selling below cost of production 1990, 1995, & 2000 1992–1997
Fresh potatoes into British Columbia Iceberg lettuce
Alleged selling below cost of production Alleged selling below cost of production
1995
Refined sugar
Alleged selling below cost of production
1998
Prepared baby food
2000
Corn into Western Canada
2001
Tomatoes
Alleged selling below cost of production Alleged countervailable subsidies Alleged selling below cost of production Fresh tomato dumping investigation
Actions Taken
AD duties imposed Duties revoked under 5-year review, August 2, 2000 Duties sustained under 5-year reviews, 1990, 1995, 2000 AD duties imposed Duties sustained under 5-year review, November 28, 1997 AD duties imposed Duties sustained under 5-year review, November 2000 AD duties imposed Preliminary affirmative determination announced October 2000 Commissioner of CCRA supported dumping claims June 24, 2002 CITT rejected dumping claims-no material injury to domestic industry June 26, 2002
Sources: Taken in part from Jabara and Reeder (2002), which they compiled from reports of the U.S. International Trade Commission, USTR, GAO and USDA.
an investigation to determine whether fresh tomatoes were being dumped by the United States. In June 2002, the Canadian International Trade Tribunal (CITT) determined that U.S. imports of fresh tomatoes had not caused material injury to the Canadian domestic tomato industry. In the past 10 years, there has been no significant growth in U.S. exports of fresh tomatoes to Canada; this has not been the case for Canadian exports of fresh tomatoes to the United States. Since 1994, Canadian exports have increased roughly eightfold. In 2000 and 2001, U.S. tomato exports to Canada were less than Canadian tomato exports to the United States, with Canadian exports totalling roughly U.S. $170 million in 2001. It is important to note that some of the disputes between Canada and the United States evolved independent of U.S. farm programs (Tables 5–7). For example,
Canada-U.S. Agricultural Trade
231
Fig. 11. Canada-U.S. Tomato Exports, 1989–2001. Source: IBAT (2002).
tomatoes, apples, and potatoes are not covered under U.S. farm programs, but trade disputes in these commodities occurred nonetheless.
COUNTRY-OF-ORIGIN LABELLING (COOL) COOL is one of the major items under the 2002/2003 U.S. Farm Bill, which will become mandatory on October 1, 2004. It is likely to have a major impact on Canada-U.S. trade and will be costly, especially to the beef and pork processors. This is an extremely complex regulatory framework, as can be seen in the following discussion. The 2002/2003 FSRI regulations will require producers to adhere to the following (Smith, 2003): Calves must be born, raised, and slaughtered in the United States in order to receive the “product of the United States” label. Thus, U.S. cow/calf producers can no longer background or finish their calves in Canada and still receive the U.S. label. Canadian calves cannot be purchased by U.S. cattle feeders, raised on U.S. grain using U.S. management and technology to finish the cattle, and receive the “product of the United States” label. U.S. cow-calf producers must maintain detailed records proving where every head they graze, background, finish, and market was born and raised. These records will provide an audit trail for an individual meat product back to the
232
ANDREW SCHMITZ AND HARTLEY FURTAN
feedlot, farm, or ranch where the animal was produced. However, this detailed recordkeeping will also add costs for the cow-calf producers, feeders, packers, and retailers. Every individual animal processed will have to be identified and followed from origin through to final processing. COOL does not apply to poultry; hence chicken integrators will not incur the added costs that beef and pork processors will.
CONCLUSIONS Many commodities traded between the United States and Canada are not covered by the 2002/2003 U.S. farm program. These include fruits and vegetables such as potatoes, tomatoes, and lettuce. Trade in many of the Canadian supply-managed commodities will not be greatly affected by the U.S. farm program, as it does not alter tariffs, import quotas, or the production controls underlying Canadian supply management. For these commodities, negotiations for trade liberalization fall under the jurisdictions of CUSTA, NAFTA, and the World Trade Organization (WTO). COOL will likely affect trade negatively for live pork and beef as well as for pork and beef products. Processors are likely to lose financially and the two-way trade in beef and pork could be greatly reduced due to the added transaction costs resulting from COOL. The discovery of Mad Cow Disease in Canada in May 2003 adds to the COOL debate. As a result of the discovery of Mad Cow Disease in Canada, the United States placed border restrictions on certain imports of beef and beef products from Canada, as did South Korea and Japan. The manner in which the border dispute is resolved will have far-reaching implications for agricultural trade and the well-being of the Canadian cattle industry. Under the new 2002/2003 Farm Program, target prices were reintroduced. In addition, the overall level of U.S. government support was increased significantly. This will add greatly to the U.S. production of certain crops such as wheat and oilseeds, which both Canada and the United States produce. The U.S. 2002/2003 Farm Bill will make competition for Canadian farmers more difficult, especially in grains and oilseeds, in view of the relatively low levels of farm support in Canada. Increased U.S. production coupled with lower Canadian farm supports will intensify the competition confronting Canadian agricultural exports to the United States and elsewhere. Trade is determined in part by the initial endowment of resources (such as arable land and climate), and partly by factors such as investment and specialization. Because the United States and Canada have many of the same resource
Canada-U.S. Agricultural Trade
233
endowments and produce similar products, some believe trade in agricultural goods must be related to government intervention. Trade disputes surrounding durum wheat is an excellent example of agricultural-trade problems between the two countries due to like products. The Canadian Wheat Board sells durum wheat produced in Saskatchewan to millers in the Midwestern United States. This grain moves through the state of North Dakota, which is the largest producer of durum wheat in the United States. Why would a country like the United States import durum wheat and simultaneously export durum wheat? If we view trade of agricultural products as trade in heterogeneous products, then it is understandable why the United States may want to import Canadian durum wheat. The Canadian grain-grading system has a more rigid specification than that in the United States, so U.S. millers have a better idea of what quality of durum they are purchasing. In addition, Canadian supplies of durum are larger than those in the United States; thus U.S. millers can be assured of a continuous supply of high-quality durum wheat from Canada. Most trade disputes between Canada and the United States occur when there is trade in what appears to be similar products, when in fact they are not the same products. Durum wheat is only one example; others include Hard Spring wheat, tomatoes, softwood, lumber, cattle and calves, and hogs. Trade agreements, like NAFTA, fail to adequately recognize the subtle quality differences in these agricultural products. Roughly 90% of the Canada-U.S. agricultural trade has at least one multinational corporation involved, and often trade is within the same firm. (As an example, ConAgra Canada will export wheat to ConAgra in the United States.) In some cases, foreign direct investment (FDI) by these multinationals increases product trade and, in some cases, substitutes for product trade. Also, multinationals lobby governments. Because multinational firms have a far greater impact on farm policy in the United States than in Canada, they are often more supportive of U.S. farm subsidies. In addition, multinationals generally favor expanded trade (whether due to expanded production from farm subsidies and/or from lowering trade barriers) and often lobby to keep borders open. This was certainly the case in the beef countervail and AD duty cases that the United States launched against Canada in the late 1990s. Cargill, Inc. and IBP, Inc. lobbied heavily to keep the Canada-U.S. border open, and they were successful since the disputes were resolved in Canada’s favor. With the discovery of Mad Cow Disease in Canada in May of 2003, the United States closed its border to live cattle imports from Canada (which remains closed as of July 2003). We hypothesize that, once again, the multinationals are continuing to lobby intensely the U.S. government to open up its border. Agricultural trade is driven partly by government policy. For example, there is little trade in dairy products between Canada and the United States – a direct
234
ANDREW SCHMITZ AND HARTLEY FURTAN
result of the high tariffs and import quotas that both countries have on imported dairy products. Both countries protect their dairy producers. If these barriers were lowered, trade would increase if for no other reason than to minimize transportation costs. Some policies encourage exports. EEP, which was a direct transfer that reduced the price of export products, was part of the 1991 U.S. Farm Bill. Canada does not have such direct export subsidies. Many of the agricultural trade disputes arise because of government subsidies on exports and on production. In the Uruguay Round of the General Agreement on Tariffs and Trade (GATT), export subsidies were reduced, but they were not eliminated. Direct production subsidies were increased while commodity price supports were reduced. Nevertheless, U.S. producer subsidies in aggregate are not decoupled from production. It is difficult to imagine how there cannot be trade disputes, given the complete lack of policy harmonization and the resulting difference in subsidies between the two countries. The FSRI in the United States and the APF in Canada continue to move agricultural policies in different directions. Is it not paramount that some effort be made to harmonize policy so as to facilitate increased agricultural trade while reducing the threat of trade disputes?
NOTES 1. Crop insurance has long been part of the U.S. farm policy. The U.S. government administers a crop insurance program through the Federal Crop Insurance Corporation (FCIC), created in 1938. Over time, this program has grown from covering just a few crops and areas to covering most crops and areas. In addition, the U.S. Congress has periodically appropriated funds for disaster assistance to farmers when farming areas have suffered widespread crop losses due to weather conditions such as drought or flooding. Between 1980 and 1998, the U.S. Department of Agriculture (USDA) expanded the availability of crop insurance from 30 crops to 67 crops and from about one-half of the nation’s counties to virtually all U.S. counties. Participation, measured in percent of eligible acres insured, increased by over 40% between 1980 and the early 1990s. Under the Federal Crop Insurance Reform Act and the U.S. Department of Agriculture Reorganization Act of 1994, Congress required farmers wishing to participate in other USDA farm programs to purchase a minimum amount of crop insurance, which helped increase participation in crop insurance to over 70% of eligible acres. In 1996, the U.S. government introduced a revenue insurance program that had three separate options for farmers: crop-revenue coverage, revenue assurance, and income protection. The USDA subsidizes the premiums and allows private firms to market the insurance. The three revenue insurance plans differ in the level of protection each provides to farmers and differ in the costs required to administer the plans. Crop Revenue Coverage (CRC) provides a minimum revenue at planting time, which is determined by taking the futures price at planting and multiplying it by the guaranteed
Canada-U.S. Agricultural Trade
235
yield (based on the farmer’s yield history for that crop). At harvest, the revenue guarantee is calculated by multiplying the harvest price by the guaranteed yield. The farmer’s actual revenue guarantee is the larger of the two calculations. In contrast, both the Income Protection and the Revenue Assurance plans guarantee a farmer’s revenue at the time of planting. The guarantee is determined by multiplying a price established at planting by the guaranteed yield. If the total-crop revenue at harvest time is below the guarantee, the farmer receives a payout. Most of the contracts taken by farmers are for high-coverage levels (Makki, 2000). 2. The role played by SSAPF in terms of its impact on agricultural trade is far from clear. There is little doubt that Canada will continue to depend on the United States as its major agricultural trading partner. Interestingly, Canada is still making no attempt to harmonize its agricultural policies with those of the United States. Thus, we expect the APF framework to provide programs that have little or no impact on Canada-U.S. agricultural trade disputes.
REFERENCES AAFC (Agriculture and Agri-Food Canada) (2002). Internet Website http://www.agr.gc.ca/index. AMAD (Agricultural Marketing Access Database) (2002). http://www.amad.org. Babcock, B. (2002). Local and global perspectives on the new U.S. farm policy. Iowa Agricultural Review, 8(3), 1–13. IBAT (International Bilateral Agricultural Trade) (2002). Internet Website: http://www.ers. usda.gov/data/fatus. Industry Canada (2002). Internet Website: http://strategis.gc.ca/sc mrkti/tdst/tdo/tdo.php. Jabara, C., & Reeder, J. (2002). Major agricultural trade issues under CUSTA: One U.S. perspective. In: W. Koo & W. Wilson (Eds), Agricultural Trade Under CUSTA (Chap. 2). New York: Nova Science Publisher. Makki, S. S. (2000). Crop Insurance: Inherent Problems and Innovative Solutions. Paper presented in honor of Luther Tweeten, sponsored by the Ohio State University and the Economic Research Service, U.S. Department of Agriculture, Washington, DC. OECD (Organization for Economic Cooperation and Development) (2002). Agricultural policies in OECD countries-PSE tables and support material. Paris: OECD. Schmitz, A., Furtan, H., & Baylis, K. (2002). Agricultural policy, agribusiness, and rent-seeking behaviour. Toronto, Ont.: University of Toronto Press. Smith, R. (2003). Sobering message on COOL to cattle industry. Feedstuffs, 75(6), 1, 23. USDA/ERS (2002). UN Comtrade data using ERS Ibat classifications. Communication with Dr. Vollrath.
THE CANADA-U.S. SOFTWOOD LUMBER DISPUTES Harry Nelson and Ilan Vertinsky ABSTRACT Disputes about Canadian exports of softwood lumber to the U.S. have persisted for more than a century. In this paper the roots of the disputes and the prospects for their resolution are examined. The focus is on the following key factors: (1) the nature of supply and demand; (2) the normative differences underlying the systems of timber management in the two countries and differences about what constitutes a “level playing field”; (3) rent seeking by stakeholders; and (4) weakness in bilateral and multilateral trade dispute resolution institutions. The paper concludes that there are good reasons to expect short term solution to the current dispute but persistence of the disputes in the long run.
INTRODUCTION Despite increasing integration of the North American economy disputes about softwood lumber exports from Canada to the U.S. persist. From a pure economic perspective free trade will benefit both economies as a whole. Some constraints on export flows including export taxes and quotas appear to increase social welfare in Canada and decrease welfare in the U.S. Yet, Americans resist free trade and Canadians object to restrictions on lumber flows even when the benefits accrue to Canada. In this paper we examine this puzzle and evaluate alternative North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 237–262 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10013-2
237
238
HARRY NELSON AND ILAN VERTINSKY
proposals to resolve the softwood lumber dispute both in the short and long run. We also examine the trade dispute resolution infrastructure in place to identify both its potential contribution and its flaws in providing sustainable solutions to the conflict.
BRIEF HISTORY Periodic disputes over softwood lumber between Canada and the U.S. have flared intermittently throughout the past two centuries (Reed, 2001 provides a detailed chronology). The most recent trade action is part of a dispute that dates back to the early 1980s. The dispute has been particularly persistent through the past two and a half decades involving U.S. trade actions, the use of formal dispute resolution procedures, as well as negotiated settlements. The most recent trade investigation culminated in countervailing and anti-dumping duties totaling 27.2% levied on imports of Canadian softwood lumber that are currently being contested. The investigations have been numbered, informal shorthand used by participants in the dispute reflecting their personal involvement and familiarity with the dispute process as well as the issues raised in the dispute. Lumber I commenced with a petition filed by the Coalition for Fair Canadian Lumber Imports in 1982, arguing that low provincial stumpage fees subsidized the Canadian softwood lumber industry and harmed American producers. This followed an initial complaint by the Northwest Forest Industries Association that Canadian imports had led to increased unemployment in the region after a short recession (Gorte, 2001). The U.S. International Trade Administration (ITC) reached a preliminary finding that the U.S. industry had been injured, but the U.S. International Trade Commission (ITC) ruled that stumpage programs did not constitute subsidies, ending the complaint.1 In 1985 the U.S. trade representative requested that the ITC update its 1982 study. It reported back that Canadian production costs were substantially lower than those for American producers. In October of 1985, the Court of International Trade issued a ruling over “carbon black” (a petroleum product from Mexico) that was interpreted by some as reversing the precedent set three years earlier so that stumpage would be found to be offered at preferential rates (Cashore, 1997). Lumber II started in 1986 when the Coalition (now renamed the Coalition for Fair Lumber Imports) re-filed its earlier petition. By the end of 1986, the ITA had reversed its earlier finding and issued a preliminary finding justifying a 15% CVD, pending a final determination to take place at the end of 1986. As a consequence, Canadian negotiators and provincial officials acquiesced to a Memorandum of Understanding (MOU) agreeing to a 15% export tax for lumber exports. The Maritime Provinces were initially included but subsequently excluded, apparently because of the large amount of private land
The Canada-US Softwood Lumber Disputes
239
found within New Brunswick and Nova Scotia, leaving British Columbia, Alberta, Ontario, and Quebec (with the majority of their timber coming from government lands) subject to the tax. The MOU set out a process by which provinces could increase their stumpage rates and reduce the export tax; British Columbia did raise its stumpage such that within a month exports from BC entered the U.S. duty-free, while Quebec progressively increased its stumpage rates so that by 1990 they were reduced to 6.2% CVD (Gorte, 2001). During 1986 Canadian and American government officials had started negotiating the Free Trade Agreement (FTA). The agreement was concluded with the MOU in place as several prominent American politicians indicated that they considered softwood lumber a “deal-breaker” and that Congress would not pass the trade agreement unless softwood lumber had been “dealt with” (Ritchie, 1997). Lumber III started in the Fall of 1991 when British Columbian government and industry officials asked the Canadian government to assert that the provinces had had met their obligations under the MOU and therefore that Canada would be terminating the agreement (as both parties were permitted to do). The U.S. government initiated an investigation in response. The ITA initially found that Canadian softwood lumber exports enjoyed a subsidy of just under 15%. This was reduced in the final determination to 6.51%. In the finding, the ITA noted that the subsidy it measured varied across the four different provinces but levied a uniform rate. Approximately half of the measured subsidy was attributed to log export restrictions in BC that the ITA claimed depressed log prices, creating an indirect subsidy. In the summer of 1992, Canadian officials initiated dispute resolution procedures through the Free Trade Agreement. The following spring, the bi-national panels formed as part of the procedure remanded back to the ITA its subsidy determinations, noting that it had neither proven specificity nor shown that stumpage fees had distorted markets. In response, the ITA reaffirmed its original findings and increased its determination to over 11%; the panel then asked for a second remand and a determination that there were no subsidies. The ITA subsequently confirmed that determination in early 1994 and the duties collected were subsequently refunded to Canadian firms (approximately $800 million in total) (Gorte, 2001). In September 1994, Congress passed the Uruguay Round Act Amendments (URAA) as part of the U.S. implementation of its WTO commitments. As part of the Act, Congressional leaders included in the preamble words to the effect that nothing in the act should be construed as restricting the ability of the U.S. government to take trade action against Canadian softwood lumber (Gagn´e, 2001). Canadian officials and Canadian forest industry representatives, well aware of the implications of such language, entered into negotiations that culminated in the Softwood Lumber Agreement (SLA). The Agreement, which voluntarily restrained Canadian lumber exports from four provinces (again BC, Alberta,
240
HARRY NELSON AND ILAN VERTINSKY
Ontario, and Quebec) consisted of a tariff rate quota under which firms could ship specified volumes of lumber to the U.S. duty free; above that limit, firms faced a two step schedule, where they could ship an incremental volume of lumber paying a $50 mbf fee and beyond that, they could ship as much additional lumber as they wanted at U.S.$100 (both of these fees were indexed to inflation). The Canadian government would collect those fees and return them to the provinces. The quotas were non-transferable between companies and provinces; therefore, firms with multiple plants in a covered province could allocate quota among its mills in the province but could not transfer it to other companies or facilities in other provinces. The quotas were largely based on historic shares of producers shipping lumber to the U.S. adjusted for special circumstances and more recent growth trends. The agreement also allowed adjustment of quotas on the basis of past use (i.e. companies not fully utilizing their quotas would lose the unused part). Lumber IV commenced when the agreement expired on March 31, 2001. On April 1, 2001, the Coalition for Fair Lumber Imports filed its trade complaint with the U.S. Department of Commerce. Part of the dispute involved for the first time the use of anti-dumping duties and the recently enacted Byrd Amendment-which permitted firms that felt they had been damaged by foreign dumping to directly claim the dumping duties collected by the U.S. government. The recurrent theme of all past disputes was the U.S. insistence that provincial governments, the major owners of Canadian timber resources, engage in “practices that distort trade.” These practices the U.S. claimed provide subsidies to Canadian producers. Canada kept insisting that the provincial system of administered pricing did not distort or provide subsidies to Canadian lumber producers. All international or binational panel reviews to date convened either under the FTA, NAFTA, or WTO have found such subsidies did not exist. Braudo and Trebilcock (2002) suggest that the persistence and intensity of the U.S.-Canada softwood lumber trade dispute “demonstrates that there are economic, political, and legal factors driving the dispute . . . [despite the emergence] of bilateral, multilateral, and global trade agreements . . . Lumber I–IV shows that the intensity of the dispute has become greater than ever.” The economic factors underlying the disputes stem from the volatile nature of demand, its inelasticity in the short run, and the lagged response of supply. They are also rooted in the development path and geography of wood supply in Canada.
NATURE OF DEMAND AND SUPPLY The Canadian industry is the world’s largest exporter of softwood lumber while the U.S. is the world’s largest consumer of softwood lumber. In recent years,
The Canada-US Softwood Lumber Disputes
241
Fig. 1. U.S. Lumber Demand and Domestic Supply, 1984–2001, in Million bf.
over 60% of all Canadian production has been exported to the U.S. Econometric studies have found that lumber demand is inelastic in the short-term (Adams & Haynes, 1996). In the long-term, lumber demand is influenced by income, population growth, and interest rates reflecting its primary use in new home construction and remodeling (USITC, 1999). Lumber prices tend to be highly volatile, changing as much as 30% from one month to the next (Zhang & Sun, 2001). Softwood lumber is largely a commodity product although there is some scope for lumber manufacturers with access to unique species or larger trees to produce more specialized products that yield higher prices. The softwood lumber industry is a competitive industry and not highly concentrated. In the U.S., most lumber manufacturers are small firms (USITC, 1999). Within Canada, average mills are larger but again the industry is not highly concentrated (although there has been a trend towards consolidation in both countries). Figure 1 shows U.S. demand and U.S. domestic supply (measured by total production and imports less U.S. exports). While U.S. demand has grown over time, U.S. supply has stayed relatively constant and the increase in U.S. demand has been principally met through increased Canadian imports of softwood lumber. Figure 2 shows that the Canadian share of the U.S. market has fluctuated between 25 and 35% over the past twenty years, with the actual level of Canadian lumber exports strongly influenced by exchange rates (Adams et al., 1986).
242
HARRY NELSON AND ILAN VERTINSKY
Fig. 2. Canadian Share of U.S. Lumber Consumption, 1984–2001.
Over time, lumber prices have stayed constant in real terms although displaying increasing volatility. The lumber disputes in the past have been highly correlated with recessions, when lumber demand decreases and Canadian exports compete with American production for market share. The collapse of the Japanese market in 1997, which has historically been a significant importer of forest products from the West Coast of North America, has further reduced demand in recent years. The long-term trend has been towards increasing use of substitute wood products, including engineered wood products and non-wood products. The location of lumber manufacturing is determined principally by access to an economic supply of standing timber (Prestemon & Buogiorno, 1997). The high cost of transportation and relatively low value of logs favours the establishment of sawmills close to their wood supply sources (Polzin, 1994; USITC, 1999). There is some international and intra-regional trade in logs but most of it tends to reflect either localized trade in neighboring wood supply regions that happen to span national boundaries or high value logs used to make specialty products. Within North America, the traditional sources of supply have been the western U.S., utilizing timber from the large expanse of private and public forestlands found in the Pacific Northwest, and the U.S. South, where lumber mills source their wood from predominantly private lands. Over time, the U.S. South has grown in importance, especially as harvests on national U.S. Forest Service lands have been
The Canada-US Softwood Lumber Disputes
243
Table 1. Annual Harvest by Ownership in the U.S. in 2001 (Million ft3 ).
National forest Other public Forest industry Non-industrial private Total
Volume
% of Total
819.4 929.2 4,677.9 9,506.8
5.1 5.8 29.4 59.7
15,933.3
100.0
Source: USDA (2002).
curtailed or sharply reduced in recent years. Table 1 shows the annual harvest by ownership in 2001. Figure 3 shows the regional timber supply for the U.S. Pacific Northwest, U.S. Inland West, and U.S. South. The share of lumber production in the U.S. PNW and Inland West fell from just under 60% in 1984 to 44% in 2001, reflecting the reduction in harvests on federal forestlands that are mainly found in the U.S. West, and historically have been an important source of timber for sawmills in the region. In recent years, expansion in timber processing facilities and maturing of commercial softwood plantations in other countries has led to a small (but steadily growing) increase in exports to the U.S. market, abetted in part by a
Fig. 3. Domestic Lumber Supply in the U.S. from Selected Regions, 1978–2001, in Million bf.
244
HARRY NELSON AND ILAN VERTINSKY
Table 2. Softwood Lumber Production in Canada for Selected Provinces in 1992 and 2002, in Billion Board Feet.
2002 1992 Change
BC
AB
ON
PQ
Maritimes
Other
Canada
14.1 14.1 –
2.8 1.8 +1.0
3.3 1.8 +1.5
7.6 4.2 +3.4
2.3 0.9 +1.4
0.6 0.3 +0.3
30.7 23.1 +7.6
Source: Natural Resources Canada (2003).
strong U.S. dollar. While Canada has historically provided the U.S. with most of its softwood lumber, in recent years offshore exporters have entered the North American market. While still small relative to the Canadian share, these imports have been growing quickly, and are expected to increase in the future as other countries take advantage of low cost timber supplies and newer, more efficient manufacturing facilities. Within Canada, the main source of supply has historically been British Columbia. Starting in 1980, there was a deliberate attempt to expand the lumber industry within the BC Interior. In recent years, the forest base has been expanding throughout Canada as other provinces seek to develop their forest resources, as shown in Table 2. The largest component of costs for sawmills are delivered wood costs. Newer mills have been developed to take advantage of economies of scale through machines that permit high speed processing with little labour requirements. There is some scope for firms to cut costs through rationalization (by shutting down high cost facilities) but the ability to take either of these cost-saving actions is predicated on having access to more timber or multiple plants situated close together. Mills in the U.S. South are older and smaller, and hence have lower productivity, so that increases in log costs are harder to offset (Wood Markets, 2002). There are significant differences in industry structure and costs across Canada that leads firms to adopt different positions on the impact of trade restrictions and proposed border measures. For example, lumber manufacturing costs in BC are higher than elsewhere in Canada: estimated costs on the BC Coast are $375 U.S.$/mbf and in the BC Interior $230 U.S.$/mbf; while costs in Eastern Canada are $204 U.S.$/mbf and on the Canadian Prairies $183 U.S.$/mbf (Wood Markets, 2002).2 At the same time, the presence of high valued species and larger trees in BC permits the development of firms that specialize in niche markets, such as Western Red Cedar and wide dimension lumber, and firms can typically sell these products for several times the price of standard construction lumber. Supply response tends to be lagged especially in a Canadian context where government policies attempt to maintain employment levels (and hence production). Firms will typically be
The Canada-US Softwood Lumber Disputes
245
willing to operate below break-even for extended periods of time where they face shut down costs and the possibility of losing a skilled work force. The slow supply response helps accentuate sudden price swings as it takes time for supply to adjust to changes in demand. Irland (1986) notes that the industry is chronically plagued by excess capacity and that in mature markets there is little opportunity to seek market growth. Spelter and McKeever (1999) estimate total North American softwood lumber capacity as 67 billion bf.3 Their estimates also provide regional detail as well as changes since 1995. Within Canada, BC accounted for 47% of the softwood lumber capacity in 1999, followed by Quebec with 28%, Ontario with 9%, Alberta with 8%, the Maritimes with 6%, and the Prairies (Saskatchewan and Manitoba) with 2%. Spelter and McKeever note that over the period 1995–1999, 55 mills were closed with a loss of capacity of 3.4 billion bf in North America (more than 50% of this loss took place in BC). However, this reduction in capacity in mill closures was more than offset by the construction of new mills and upgrades to existing mill capacity leading to a net gain in capacity of 4 billion bf. Of the net gain, approximately 45% was in Quebec and another 45% was in the U.S. South. When prices fall below certain thresholds the resistance of governments and labour unions to rationalization and reduction in capacity weakens. Thus, often during severe recessions the industry as a whole manages to reduce its costs and gain a competitive advantage. Overall, there has been a migration of production and capacity within Canada away from BC towards lower cost regions. Within the U.S., capacity was closed in the Western U.S. while it expanded in the U.S. South. Within the U.S. South and U.S. Inland West, wood costs and therefore total costs remain high. Spelter and McKeever (1999) estimate the net drain of softwood timber (removals less growth) in the U.S. and Canada and note that throughout many regions of the U.S. South, the net drain is positive with removals exceeding the growth and note that timber prices are higher in these regions, reflecting limits to timber supply. There are a few key factors that can be highlighted as drivers of trade disputes. Canada’s abundance of timber resources compared to the size of its economy and thus its domestic demand meant that its lumber industry has developed to serve export markets. Indeed, the opportunities in the American market are at the heart of the expansion of the sector and the increasing share of Canadian exports of the U.S. market. There are limits, however, to the supply of wood. As harvesting locations move further from population centres, costs increase and competitive positions change. The cost competitiveness positions within Canada (between provinces and regions) have shifted over time as a result of changes in wood supply, stumpage policies, regulatory costs, and the age of plants. Different
246
HARRY NELSON AND ILAN VERTINSKY
cost competitiveness positions means differences in economic interests and thus differences in negotiating positions with respect to trade with the U.S. Thus there are substantial disagreements among Canadian provinces and firms with respect to alternative means of resolving trade disputes as they arise. The inelasticity of demand means that the rewards of constraining supply are very tempting. Thus, the emergence of a coalition of U.S. producers attempting to restrict Canadian imports is not surprising. But the positive factors (i.e. the distribution of economic benefits and costs of trade) tell only part of the dispute story. Normative differences underlying the systems of timber management in the U.S. and Canada also lie at the heart of the dispute.
INSTITUTIONAL DIFFERENCES BETWEEN CANADA AND THE U.S. Political Systems Under the parliamentary system, Canadian provinces enjoy a great deal of autonomy (more like a federation when it comes to natural resource issues) and the federal government has been reluctant to exercise much of a role unless provinces are in agreement. In Canada the federal government has primary responsibility for international trade. Hoberg and Harrison (1994) note that the provinces have been quick to assert provincial jurisdiction in areas affecting natural resources, including environmental protection. When the provinces and federal government can reach a consensus, however, it is quite capable of implementing substantive policy changes associated with trade issues (Feigenbaum et al., 1993, pp. 73–74).
U.S. and Canadian Forestlands The U.S. and Canadian forest products industries are marked by many similarities. Lumber mills in both countries produce mainly commodity grade softwood lumber used principally in the home construction and remodeling markets. The timber is harvested from a mix of private and public lands. In the U.S., those public lands are state and federal forest lands. In Canada, the public lands are principally provincial or Crown lands as well as some federal land. In the U.S., private land is further subdivided into that held by companies for the express purpose of growing commercial timber crops (industrial timberland) and non-industrial timberland, forested land held by small landowners for amenity purposes and
The Canada-US Softwood Lumber Disputes
247
Table 3. Ownership of Timberlands in the U.S. and Canada. Ownership of U.S. Timberlands
U.S. Timberlands (000 Hectares) (Defined as Productive and Non-Reserved Forests)
Ownership of Canadian Forestlands
Canadian Forestlands (000 Hectares) (Defined as All Forestland)
Non-industrial Industrial Government
116,394 28,513 53,215
Private Provincial Federal
25,056 296,496 96,048
Total
198,123
Total
417,600
Source: U.S. Department of Agriculture (2002), Natural Resources Canada (2002).
where timber sales are incidental (although they are a significant source of timber – see Table 1). Table 3 shows the significant difference between U.S. and Canadian ownership of forestland. The principal difference is in the amount of harvest derived from these different land types and the different institutional arrangements governing harvests on those lands. Under the Canadian Constitution, provinces were granted the rights to forest resources in 1867.4 Therefore, forestlands are owned predominantly by the provinces, although private ownership is significant in certain regions of the country. Minor amounts of forestland are held collectively (if one considers the timberlands held by the federal government on behalf of First Nations), and federal ownership of forestland is concentrated in the Yukon and Northwest Territories. Within Canada, there are significant differences in the amount of private land found within each province (see Table 4). This reflects historical patterns of settlement as Canada developed east to west. Most private land in Canada is found in eastern Canada and around agricultural land since it had the most value at the time of settlement. By the time the forest industry started developing in the BC Coast in the early 1900s, the U.S. moved towards the retention of public ownership of forestlands. This influenced the adoption of a similar policy in Western Canada. U.S. and Canadian Forest Policies Canadian provinces share many common objectives they hope to achieve through their forest policy. Among one of the most important is to provide economic opportunities through development of the forest resource. This has been accomplished through a series of policies designed to develop a capital base and processing sector in order to extract higher value from the trees as well as provide employment opportunities. The main feature of these policies are tenure arrangements, or
248
HARRY NELSON AND ILAN VERTINSKY
long term leases under which companies assume management responsibilities and other commitments in exchange for access to Crown timber. As part of these tenure arrangements, provincial governments determine timber fees for stumpage. In general, most timber harvested in Canada is under large-scale, area-based renewable tenure arrangements held by companies (Ross, 1995). Much of the timber harvested under these arrangements is dedicated to a company’s own processing facilities, which is reinforced by appurtenancy conditions attached to the cutting rights. For the most part, the obligations increase with the duration and size of the tenure held by the firm; for example, area-based, large-scale renewable agreements typically carry the most responsibilities in terms of planning, management, and silviculture. Within Canada, different provinces utilize various formulas to determine the timber fees payable for timber cut on public land. While the formulas differ somewhat in their design, they all utilize product prices and some measurement of costs in determining the fees equivalent to the residual value of the wood. In the U.S., where most of the timber is derived from private land, the Federal government and state governments (all of which administer their own forestland) use timber auctions to determine the fees payable.5 U.S. trade complaints have primarily targeted Canadian timber fees. The U.S. position has been that government determination of harvesting fees acts as a subsidy where those fees are presumed to be set lower than those that would prevail in a competitive market. Other Canadian policies have also been attacked through trade actions. The U.S. has claimed in the past that Canadian restrictions on log exports (similar to those that exist on public lands in the U.S.) have indirectly subsidized the Canadian industry by depressing prices within Canada. The U.S. has also argued that the provincial governments’ grants of long-term cutting rights have acted as subsidies since firms made no upfront payment for these rights; instead, the grants were often conditional on the establishment of processing and manufacturing facilities. Finally, the U.S. has also targeted Canadian government funding of reforestation programs and other programs designed to promote the forest industry in rural areas, provide technical assistance to private landowners, and help firms modernize. Differences in ideology with respect to roles of governments and markets and the nature of forests as public goods have played an important role in the dispute over the years. “Canada has chosen to maintain public ownership and management of lands containing valuable harvestable timber. Canada also believes that it can manage the harvesting of timber for softwood lumber on these levels through government-administered stumpage programs in an efficient manner” (Braudo & Trebilcock, 2002). Provincial governments also saw the forest resources as a means for developing their relatively less developed regions and maintaining
The Canada-US Softwood Lumber Disputes
249
stable communities. In the U.S., in contrast, market mechanisms are perceived to be the best means of delivering almost all goods and services. Thus, Canadian administered pricing systems are viewed not as legitimate Canadian choices but as a vehicle for distorting the playing field in favour of Canadian producers. At the heart of the U.S. approach to trade are norms concerning what constitutes a level playing field in trade. The Idea of the Level Playing Field Abbott (1996) suggests that the shift away from tariff reductions in international agreements has meant a move towards negotiations over countries’ policies and practices and that in the U.S., this has led to a focus on practices by other countries that are considered unfair. He examines Section 301, the U.S. trade law that governs the use of trade actions and remedies, and argues that political pressure to seek relief from “unfair” trade practices, in part, stems from a belief in the importance of “fairness” and that it is important to understand the norms underlying this belief in the development of policy processes of the U.S. government (separate from traditional interest group analysis).6 These beliefs play a role in trade actions and trade legislation and can promote harmonization (at least as measured by the conformance of other countries domestic policies with the U.S.). The belief in market mechanisms is a strong part of the U.S. worldview. Government intervention that can be construed as interfering with market mechanism is met with a great deal of skepticism by Americans (Goldstein, 1988). Implicit in U.S. industry challenges of Canadian softwood lumber pricing policies has been the idea that government determined stumpage cannot possibly accurately determine the economic value of the wood being harvested. No definition of fairness exists although the term is repeatedly used by participants (as well as equity, level playing field, and similar metaphors), but Abbot claims several consistent norms have emerged over the past 25 years in debates over the “fairness” of foreign countries trade practices. Some of these norms are widely held, while others are more specific to the U.S. and often provide the grounds on which parties seek relief. Several of the arguments employed by U.S. parties in softwood lumber disputes have tapped into these beliefs: Norm of nullification or impairment (where the unfairness arises because a party evades a prior commitment); Norm of repricocity, where fairness means that concessions by one party are matched by equivalent concessions by the other. These matching concessions can be specific, or involve more general concepts such as equal access, national repricocity, or sectoral repricocity;
250
HARRY NELSON AND ILAN VERTINSKY
Fairness as expressed in terms of equal outcomes or measurable results (so regardless of whether or not there are matching concessions, the party looks for a demonstrable impact, such as an actual increase in sales, or a particular market share); Norm of the free market (there is a predisposition in the U.S. towards the view that government policies inherently cause market distortions and that the widespread use of government policies in other countries therefore favours foreign firms); and More general mercantilist claims (in which no claims are made as to what is fair but rather the argument is made that imports take away domestic employment opportunities and offer no benefits to the economy, an argument often advanced by trade unions in the U.S.). The application of these norms is selective and largely self-serving. For example, during the period of the SLA, U.S. firms have complained that Canadian firms have taken measures to evade their commitments (such as the shipment of rougher-headed lumber and pre-drilled studs that fell outside of the tariff classification used to implement the SLA) and have invoked these examples to press for more stringent restrictions or as evidence Canadian firms are not acting in good faith. U.S. firms also draw comparisons between selected U.S. and Canadian policies, arguing that there should be equal treatment: for example, the absence of log export restrictions on private lands in the U.S. should be matched by Canada dropping any restrictions on log exports from Crown lands, while national U.S. environmental policies governing practices on federal forestland (which are viewed as more stringent) should be applied to provincial forestland.7 Abbot examines the internal consistency and coherence of these norms with Section 301 rules and their applicability to international trade rules. He finds that there is no inconsistency with either set of rules and the underlying ideas of evasion expressed in terms of strict non-compliance or discrimination. However, where ideas of (failed) repricocity underpin trade complaints, the issue becomes more problematic, since negotiators are balancing off competing interests through different concessions. It becomes even more difficult when market norms are implicitly utilized, since the basis for arguments will be based on measuring the distortion attributable to government polices. Governments intervene in myriad ways and the effects are complex and far-reaching. It is difficult to apply neo-classical theory empirically and appropriately and different countries have different ideas about the appropriate mix of state and market. Abbot concludes that this limits the legitimacy of market-distortion arguments and would be unworkable as a principle of national trade policy.
The Canada-US Softwood Lumber Disputes
251
Irland (1986) suggests that that the American belief in free markets is not shared by Canadians, who tend to be more skeptical of market forces, and perceive government intervention as an appropriate counter-balance to problems of market power and potential over-exploitation of natural resources. The regional importance of the resource, differences in economic interest within and across borders, and differences across borders in norms and perceptions of what is legitimate all combine to generate a persistent trade dispute concerning softwood lumber. To understand, however, the manifestation and timing of disputes one must consider the political economy of the forest sector. Political economic analysis also provides the answer why both nations reject solutions which provide for increases in their social welfare.
THE POLITICAL ECONOMY OF THE NORTH AMERICAN FOREST SECTOR We start with a puzzle. Why did Canada and the U.S. reject resolutions to the trade dispute that would have benefited both parties? Economists that have analyzed each of the various agreements have all drawn the same general conclusions: American consumers have been harmed while American lumber producers have benefited. Depending upon the nature of the trade restriction, either provincial governments or Canadian forest product firms (and Canadian consumers where they have measured the effect on them) have benefited from a redistribution of income from American consumers created by higher lumber prices. Studies of the impact of various trade restrictions include an examination of the impact of the MOU (Myneni et al., 1994; Wear & Lee, 1993) and the SLA (Van Kooten, 2002; Zhang, 2001). They generally suggest that there are significant transfers: Wear and Lee (1993) found that American producers saw a gain of $2.6 billion while American consumers saw a loss of $3.8 billion (all in $1982) as a consequence of the MOU, and Zhang (2001) estimated that the SLA had increased lumber prices by just under $59/mbf, with producers benefiting by $7.7 billion and American consumers facing a loss of consumer surplus of $12.5 billion (all in $1997). Van Kooten (2002) estimates that the SLA benefited Canadian consumers by $109 million annually. Producers on both sides of the borders gained at the expense of U.S. consumers. Adams (2003) predicts that the current 27% tariffs, if maintained, would increase U.S. lumber prices by an average of 4.5% over 2002–2010 and U.S real stumpage prices by 2.4% in the U.S. South and 5.1% in the U.S. West over the same period. As with many problems, concerns about the distribution of benefits overwhelmed consideration about the potential size of benefits. Several factors make
252
HARRY NELSON AND ILAN VERTINSKY
agreements difficult to reach. In Canada, the interests of different provinces and producers vary significantly. For example, an intense debate within Canada took place over whether to reach an agreement similar to the SLA upon its expiration. Allocation of quota according to historical levels of exports to the U.S. excluded industries and regions which had exported in the past to other markets (e.g. the Coastal industry in BC exporting to Japan). These industries and regions were reluctant to agree to a quota system where they received relatively low quotas. Efficient producers with large capacity saw that rationalization of the industry could be brought about under a system of export taxes, which was preferable to a system where they were constrained from expanding because of quotas. Inefficient producers, however, preferred the quota system because of the rents accruing from supply restrictions. Finally, not all softwood is the same. Cedar producers in Coastal BC, for example, enjoyed market power in their niche market and exercised it. They saw no benefits from an agreement that limited their flexibility to manage their markets. In the U.S., differences in forest land-holdings and cost structures created differences in incentives and the intensity through which forest companies pursued trade actions, Firms with large land holdings saw trade restrictions as measures to increase the value of those holdings. Processors without forestlands were less concerned. Consumers and homebuilders associations objected to trade restraints, while forest labour unions and other forest-based community organizations lobbied for such constraints.
U.S. Interests and Perspectives Diffused benefits of free trade (enjoyed by consumers and builders) yield politically in the U.S. to concentrated pain (suffered by forest product firms and forest dependent communities). Led by large forest product firms with large land holdings coalitions emerged to constrain imports of Canadian softwood lumber. As noted earlier, the Coalition first filed an unsuccessful petition with the U.S. government in 1982. In 1986, the Coalition petitioned again this time successfully. It has persisted ever since, using U.S. trade law instruments to harass Canadian exporters and force Canada to enter into agreements, which constrain the flow of exports to the U.S. The Coalition has successfully played the political game. Aided by U.S. Senators from forest industry states (principally in the U.S. Pacific Northwest and U.S. South) the Coalition managed to pressure U.S. trade institutions and the office of the U.S. Trade Representative to pursue its cause, succeeding in winning changes in U.S. laws to ensure success in further rounds of the dispute. Part of its strategy has been a threat to “go outside the system” whenever it doesn’t produce satisfactory results (e.g. meaningful restrictions on Canadian imports).
The Canada-US Softwood Lumber Disputes
253
Examples include: a number of Congressional bills introduced at the behest of the Southern Lumber Manufacturers Association to limit Canadian lumber imports or change the definition of subsidy in 1985 (Braudo & Trebilcock, 2002); the Coalition filing a constitutional challenge to the FTA review process following the ruling of Extraordinary Challenge Committee in 1994 (Gorte, 2001); and recently introduced bills in Congress that would levy duties of up 46% today to replace current duties (Hamilton, 2003). A more recent strategy that has been introduced is a portrayal of Canadian forest management practices as “unsustainable” and environmental standards as “too low,” thereby implicitly representing another subsidy by which Canadian producers benefit. Although environmental standards have not been used directly in any trade actions to date, they have been used indirectly to appeal to other domestic constituencies within the U.S. (environmentalists) to win broader Congressional support for trade action (Cashore, 2001). In recent years, institutional ownership of forestland in the U.S. has become important (Binkley et al., 1996). From their perspective, increases in timber prices lead to increases in land values. The emergence of forest landholders in the U.S. has added a new party to the dispute. Saunders (2003a, b) notes that two of the largest forest landowners sit on the executive committee of the Coalition, and some observes see the dispute having more to do now with maintaining timberland values (which are higher with higher lumber prices) rather than U.S. employment in the forest industry or domestic lumber supply. Zhang (2002) who analyzed the political process involved in the dispute has documented several instances where letters from senators brought immediate action from the administration in support of the Coalition. He concluded that senators clearly were responsive to important timber related interest groups. He noted that “a small but concentrated softwood lumber industry can successfully lobby their elected officials such as Senators and demand protection from foreign competition, despite the fact that such protectionism harms the economic welfare of the nation as a whole.”
Canadian Interests and Perspectives The analysis of the political economy in Canada is somewhat more complex. Canadian provincial governments concerns for employment and stability of forest dependent communities, an industry attempting to secure immobile investments and reduce uncertainty objected to fundamental reforms involving a move toward a market system pricing of timber, a move which at least would challenge the basic legitimacy of U.S. trade actions. Reaching agreement on short-term dispute resolution alternatives was more difficult as we have noted since provincial economic and social interests vary. Even within provinces reaching a consensus and
254
HARRY NELSON AND ILAN VERTINSKY
a compromise solution with the U.S. is difficult. The interests of firms vary. The Federal government largely plays coordinating roles, rather than a leadership role, given the constitutional power of the provinces over natural resources. Thus, for example, while BC has responded to the trade complaints by raising its stumpage or seeking other negotiated settlements, reluctance by Eastern Canadian Provinces to adopt similar measures makes it difficult to develop a unified position. Provinces with abundant but undeveloped resources are not interested in restrictions that stop the development of their resources, while provinces that have reached the limits of their wood supply may welcome such restrictions. The political economy has become even more complex as environmental groups within Canada have found the dispute as a means to reduce production of timber in Canada’s forests, while First Nations communities see it as an alternative means to press their political concerns. Provincial goals and objectives also differ. The different provinces are also at different stages of utilizing their timber resources; some are expanding while some are reducing annual harvest levels. Industry structure and cost differences can also lead to differential impacts of alternative trade measures in different provinces and thus to different negotiating positions.
INSTITUTIONS FOR DISPUTE RESOLUTION: CAN THEY SOLVE THE PROBLEM? Trade between the U.S. and Canada is covered under both a regional agreement and a multilateral agreement that both contain dispute resolution procedures. The same dispute cannot be heard simultaneously under the WTO and NAFTA; therefore, the arguments crafted are particular to the venue chosen. In the current Canada-U.S. softwood lumber dispute, Canada is proceeding on several grounds, arguing under NAFTA that the application of countervailing measures and anti-dumping duties is inconsistent with U.S. trade law and that the U.S. has failed to meet its obligations requiring consultation and review of appeals for exclusion from the duties by a number of Canadian companies. Under the WTO, Canada is arguing that changes to U.S. trade law (which contributed to the findings of subsidy and dumping) were inconsistent with their GATT obligations and that the cross-border comparisons used in determining the CVD were invalid and disallowed under WTO rules. The NAFTA Process The NAFTA Agreement contains a dispute settlement process found in four chapters similar to those under the WTO, involving a quasi-judicial legalistic process that has a consultative phase followed by hearings before an arbitral panel. The panel is appointed from the member countries. Each chooses one or two panelists
The Canada-US Softwood Lumber Disputes
255
who then choose the final panelist (panels may consist of three or five members depending upon the chapter the dispute is heard under). The enforcement mechanism consists of the right of the prevailing party to receive financial compensation or retaliatory suspension of benefits. Two chapters have been utilized in the context of the U.S.-Canada Softwood lumber dispute. Chapter 11 covers investments and is available to North American firms that can sue host governments (where they are not the same nationality as the firm) for government policies or actions that they believe either discriminate against them or constitute an expropriation of their property.8 Chapter 19 covers anti-dumping and countervailing duties. It permits countries to use their domestic trade laws and remedies, which also retain the right to modify their laws so long as they are not inconsistent with their GATT obligations (Article 1902.2.i). Braudo and Trebilcock (2002) argue that there is an institutional failure in the NAFTA dispute resolution process and that it is incapable of solving a problem as complex (and high profile) as the softwood lumber dispute (and that Canada should look to the WTO dispute resolution process).9 The bi-national panel was criticized quite heavily by American commentators, who argued that the results reflected the fact that the panel rulings were divided along national lines (where Canadians had been the majority on the panels).
The WTO Process The agreement reached in 1994 that provided the basis of the WTO was noticeable for the establishment of an institutional structure to address disputes. The agreement establishes procedural rules over the creation of a panel to hear the dispute (through identifying who can be eligible to serve on a panel), the evidence to be considered, and the type of judgments that can be reached (as well as the ability to appeal that judgment). The dispute resolution is a quasi-legal process that sometimes uses previous rulings in reaching decisions to assess whether government policies or action are in compliance with WTO rules. It differs substantially in that the procedure has now become more automatic; previously, where a report or ruling could be blocked by one country (since a consensus was required), the rule of negative consensus is now required to prevent the adoption of a dispute panel ruling or report. GATT rules incorporate an added element that is not found in U.S. trade laws-that there must be a causal link between subsidized imports and material injury, and that the subsidies are causing the injury. This is to correct for other exogenous factors, such as changes in exchange rates or productivity shifts that may result in increased imports. The WTO agreement also provides for monitoring and enforcement of judgments. For a member country that is found to be at fault, it has three options. It can choose to:
256
HARRY NELSON AND ILAN VERTINSKY
(1) Bring the offending practices into compliance; (2) Offer compensation; or (3) Allow the prevailing party to establish equivalent measures. It is important to note that the country may maintain its restrictive trade practices under two of these options; in addition, the prevailing party may choose to target industries other than those in which the dispute originally occurred if the offending party does not comply with the ruling. The WTO has already issued some preliminary rulings relevant outside of the dispute resolution process reflecting cases brought prior to the initiation of Lumber IV. The first is that log export restrictions (considered an indirect subsidy under Lumber III by the ITA) cannot be construed as acting as a subsidy, while the second is a finding that the Byrd Amendment, an act passed by Congress that remitted anti-dumping duties to U.S. firms that claimed they had been damaged by the dumping and had been utilized by U.S. lumber firms, violates GATT rules. Both the NAFTA and WTO processes of dispute resolution do not provide a final timely solution where compliance is guaranteed. A decision may legitimize one position compared to another. Here the WTO with its professional panels appears to have more legitimacy. However, the process is rather slow and the remedies weak. Canada may retaliate against the U.S. in case of non-compliance by imposing trade restrictions (these may hurt Canada more than the U.S. given the share of trade with the U.S. in Canada’s GDP). The stakes of non-compliance in NAFTA are higher but the perceived moral authority of its dispute panel is in doubt given the partisanship displayed by panelists. A negative decision can be overcome in further disputes by a change in domestic law that corrects the flaw so that the action can be brought again on similar grounds. Both processes are slow and expensive. Persistence by protectionist elements may lead to high uncertainty and thus severe economic damage to exporters. Even without winning, the objectives of protectionists may be achieved by constantly initiating new actions. So are these institutions valueless? The answer is no. While they cannot bring about a permanent solution to the dispute, they may lead to closure in the perceptual gaps concerning normative differences about fairness and provide a regularized channel for dialogue between the disputants and a forum to protect free trade.
The Current Positions and Proposals There have been intermittent discussions over how to resolve Lumber IV since the initiation of the trade action. The U.S. Commerce Department has developed a policy framework in which it spells out what it believes would provide the
The Canada-US Softwood Lumber Disputes
257
basis for a long-term settlement and brings an end to such disputes (U.S. Department of Commerce, 2003). In the proposal, the Department argues that resolution of the conflict requires Canadian provinces to move to a market-based system in determining timber prices, with the use of auctions as the preferred means of accomplishing that objective. In addition, the Department also looks to provinces to remove the non-market distortions it associates with other provincial forest policies, principally those around tenure arrangements that govern harvest levels. There has been a general reluctance to adopt such wide-ranging policies within Canada (Hamilton & O’Neil, 2003). British Columbia has tabled a draft proposal under which the two countries would reach an interim agreement in which the existing duties would be replaced by an export tax while they try to develop a durable solution (British Columbia, 2002). The original proposal called for a fixed ad valorem export tax but has been amended to be on a sliding scale; with the upper level between 17 and 21% when lumber prices are low, falling to 0% when prices are high (Hamilton & O’Neil, 2003). The political logic for such taxes is the fact that U.S. producers can argue more forcefully against Canadian imports when prices are low, many U.S. plants shutting down or working with unutilized capacity and unemployment in the sector is high. When demand is strong and prices are high the arguments for measures, which increase prices, may not have a strong political echo. Indeed, worries of consumers of lumber products about affordability and worries about inflation may counter-balance rent seeking by U.S. producers. Also, from a long-term perspective sustaining extremely high prices increases efforts to substitute lumber and decrease its use through technology change and improved utilization. Thus, U.S. producers may have less concern about Canadian supplies when prices are high. The short-term economic logic of variable taxes relies on the fact that supply functions are very elastic near shutdown prices. At very high prices supply becomes more inelastic since wood supplies may be a limiting factor and adding a third shift is rather expensive. When supply is very elastic more of the tax burden is shifted to the consumer, everything else being equal. So from the Canadian exporter point of view having higher taxes in the more elastic portion of the supply curve and lower taxes at the high-end helps reduce the tax burden on the producer. This is especially true for efficient producers as shutdown of a significant number of less efficient producers provide an effective floor to the price. From a longer-term point of view variable tax schemes may provide a signal to inefficient producers to get out of the sector permanently. The threat of many expensive shutdowns may encourage rationalization of the industry, which will benefit survivors. From a Canadian public policy point of view there are two concerns: (1) employment; and (2) long term appropriate use of the resource. Myopic concerns for employment may stop rationalization. In the long run rationalization will
258
HARRY NELSON AND ILAN VERTINSKY
mean more stable employment. Reducing the use of the resource during period of weak markets may lead to higher long-term returns from the resource. The imposition of the countervailing and anti-dumping duties has not stopped the flow of Canadian softwood lumber to the U.S.; indeed, some Canadian firms responded by cost-cutting and raising their productivity and increased their exports over this period. U.S. producers are now seeking some combination of a system that includes both volume restraints and tariffs.
THE FUTURE There are good reasons to expect short term solutions but persistence of the dispute in the long run. While free trade is an ideal commonly endorsed, perceptions of fairness will continue to diverge. The underlying economic interests will not change substantially. Canadian firms have an incentive to increase production to take advantage of economies of scale. Taxes may temporarily reduce or slow shipments but over time Canadian firms will become more efficient. Proposed changes to Canadian forest policy are unlikely to achieve a long-term solution. Expanding the use of auctions will not necessarily be a panacea. The use of auctions as a means of allocating timber has been heavily criticized in the U.S. as to whether they provide a sufficient return for timber harvesting (Saunders, 2003a, b) and criticized as to whether they create the right incentives for proper management of public forestlands (Gorte, 1995). Auctions do offer potentially greater legitimacy as a means of pricing timber and may better reflect the economic value of timber within Canada. Since the U.S. market ultimately determines the economic value of timber in Canada, Canadian timber prices will rise and fall more closely with changes in the U.S. lumber market. This may create a problem as Canadian exports might increase in poor markets as timber prices become more market responsive.10 Canadian firms will still be susceptible to charges of whether or not timber prices reflect market values; concerns over issues of market power and collusion raised over U.S. timber sales can also be raised as to whether or not Canadian timber prices reflect “fair market value” and offer other grounds on which the Coalition can advance its political arguments. In the long run, border restrictions do not address the main problem facing U.S. companies, namely access to timber and the relative cost of timber. In the long run, the most important factors, however, that affect the North American lumber market are substitution and imports from non-Canadian sources. Long-term U.S. capacity expansion is limited (e.g. because of environmental constraints) and changes in forest policies that have reduced harvesting activity on federal forestlands. Imports and substitution encouraged
The Canada-US Softwood Lumber Disputes
259
by high prices resulting from border measures may have, however, irreversible effects. As we noted earlier, offshore imports have shown dramatic growth in the past decade. The higher price levels resulting from border measures on Canadian softwood lumber encourage substitution to other products and technological change that reduces the use of wood in current applications. Developments in the market since the Softwood Lumber Agreement demonstrate the effects of high prices. Once new substitute inputs are used or new technologies developed and adopted, a shift back to lumber and traditional methods of its use are more difficult. The growth in the construction market in the U.S. in the past five years in combination with constraints in supplies from Canada resulted in relative high prices for softwood lumber and a decline in lumber usage (measured in per capita use). The decline reflects increased efficiencies of use and substitution. Substitutes include wood and non-wood based products. Engineered lumber (wood I-joist and LVL), steel studs for framing, brick and block for exterior use and composites and plastic resin for various applications are examples. The growth in the use of non-wood substitutes was especially significant in residential construction. The traditional North American method of lumber framing has declined from 86% in 1995 to 78% in 2000 and is expected to decline to 71% by 2005. Building methods that use steel studs, concrete exterior walls as well as new building methods such as factory-built panelized and modular homes are being rapidly adopted (Taylor, 2001). The effects of substitution are experienced more in the higher cost segments of softwood lumber products (wide-dimension lumber). The capacity of production of engineered wood is expanding, which suggests lower prices and higher substitution rates in the coming years for wide dimension lumber, with a reduction in the historical price premium. This trend may be constrained, however, by lower prices of lumber that may reduce the price differential and the incentive of homebuilders to shift to use engineered wood, plus a historic preference for lumber.
NOTES 1. Under U.S. trade law, the ITC is charged with determining whether or not there is injury, and the ITA with determining whether or not a subsidy exists (and if so, measuring it). To qualify as a subsidy, a government policy has to meet several tests. The first of these is that the policy or program is “specific,” which means that the policy or program is only available to a defined group of firms or industries (either explicitly or de facto) and whether the policy or program provides the good or service on a “preferential” basis. The determination of preferentiality involves price test comparisons between the good or service in question and external prices. Several different possible benchmarks exist, ranging from the government’s cost in providing the good to market prices for the good. To
260
HARRY NELSON AND ILAN VERTINSKY
be countervailable, it is necessary to find that a subsidy exists and that it is either causing or threatening to cause material injury to domestic producers. In this dispute, the ITC found that timber in Canada was made generally available to all forest product manufacturers, not just to softwood lumber manufacturers, and was therefore not specific, and also found that it was not offered on a preferential basis. 2. In comparison, they estimated costs of lumber production of $280 in the U.S. South and $293 in the U.S. PNW. 3. In 2001, U.S. lumber consumption and apparent Canadian demand (Canadian production less exports) was approximately 62 billion bf (Natural Resources Canada, 2003). 4. The exception were the Prairie provinces – Alberta, Saskatchewan, and Manitoba – that received jurisdiction over their provincial forests when the federal government returned control of natural resources to them in the 1930s (Myre, 1998). 5. Economic theory would suggest that in a perfect world (perfect foresight, competitive markets, and no transaction costs) the two methods-residual value and auction-would yield equivalent values. 6. Section 301 is called “Relief from Unfair Trade Practices.” The act itself uses the terms unfair, inequitable, and unreasonable, but does not define them (other than to say that unreasonable is inequitable and unfair) (Abbott, 1996, p. 419). 7. Log exports from Crown lands are subject to governmental approval, as is the case for log exports from public lands in the Western U.S., while there are no restrictions on log exports from private lands in Canada with the exception of BC. 8. A U.S. lumber producer with mills in Canada (Pope & Talbot) sued Canada over the quotas it received under the SLA, arguing that its application was discriminatory and arbitrary (it was unsuccessful in its main complaint), while several Canadian companies are currently suing the U.S. government over U.S. countervailing measures in the most recent round of the trade dispute. 9. They make the claim on two major points. First, any NAFTA panel is limited to remanding the matter to the domestic agency that made the decision and asking it to reconsider its original findings; therefore a successful finding against the imposition of duties cannot remove the duties, instead, the agency in question will re-determine the matter. Furthermore, the reliance on interpretation of each countries own trade laws, and the fact that panel members are drawn from the member countries to the dispute, gives rise to the possibility that decisions will be made along national lines and may therefore be viewed as lacking legitimacy by some, as happened in Lumber II. 10. Grafton, Lynch and Nelson (1998) estimate the economic rent collected by stumpage in BC and suggest that stumpage has not been very responsive to market forces; however, they argue that based on economic theory the concern is not over whether too little stumpage is collected (since it will have no bearing on production because Crown harvest levels are fixed) but rather that too much may be collected in poor lumber markets, leading to curtailed production.
ACKNOWLEDGMENTS We are grateful to participants of the U.S.-Canada Trade and Investment Conference at the University of Indiana for useful feedback and to the SFM Network for financial support.
The Canada-US Softwood Lumber Disputes
261
REFERENCES Abbott, K. (1996). Defensive unfairness: The normative structure of Section 301. In: Bhagwati & Hudec (Eds), Fair Trade and Harmonization (pp. 95–174). Cambridge, MA: MIT Press. Adams, D. M. (2003). Market and resource impacts of a Canadian lumber tariff. Journal of Forestry, 101(2), 48–52. Adams, D., & Haynes, R. (1996). The 1993 softwood timber market assessment model: Structure, projections, and policy simulations. USDA For. Serv. Gen. Tech. Report PNW-GTR-368. Adams, D., McCarl, B., & Homayounfarrokh, L. (1986). The role of exchange rates in Canadian-United States lumber trade. Forest Science, 32(4), 973–988. Binkley, C., Raper, C., & Washburn, C. (1996). Institutional ownership of U.S. timberland. Journal of Forestry, 94, 21–28. Braudo, R., & Trebilcock, M. (2002, May). The softwood lumber strategy: Implications for Canada’s future trade strategy. Faculty of Law, University of Toronto. Mimeograph. British Columbia (2002, December). Canada-US softwood lumber trade dispute: Suggested elements of a resolution agreement. Accessed at www.gov.bc.ca/for. Cashore, B. (1997). Flights of the Phoenix: Explaining the durability of the US-Canada softwood lumber dispute. Orono, ME: Canadian-American Centre, University of Maine. Cashore, B. (2001, April 20). What should Canada do when the softwood lumber agreement expires? Retrieved February 8, 2003, from http://www.policy.ca/PDF/20010205.pdf. Feigenbaum, H., Samuels, R., & Weaver, R. K. (1993). Innovation, coordination, and implementation in energy policy. In: R. K. Weaver & B. Rockman (Eds), Do Institutions Matter? Government Capabilities in the United States and Abroad (pp. 49–109). Washington, DC: Brookings Institution. Gagn´e, G. (2001). Canada-US “free” trade and the softwood lumber dispute. University of Ottawa. Working Paper accessed at www.policy.ca. Goldstein, J. (1988). Ideas, institutions, and American trade policy. International Organization, 42(1), 179–217. Gorte, R. (1995, October 30). Forest service timber sale practices and procedures: Analysis of alternative systems. CRS Report for Congress. CRS Report 95–1077. Gorte, R. (2001, February 2). Softwood lumber imports from Canada: History and analysis of the dispute. CRS Report prepared for Congress. Grafton, Q., Lynch, R., & Nelson, H. (1998). British Columbia’s stumpage system: Economic and trade policy implications. Canadian Public Policy, 24(2), S41–S50. Hamilton, G. (2003, January 31). U.S. move threatens talks. Vancouver Sun, p. D1. Hamilton, G., & O’Neil, P. (2003, February 3). Lumber talks take on a sense of urgency. Vancouver Sun, p. C1. Hoberg, G., & Harrison, K. (1994). It’s not easy being green: The politics of Canada’s green plan. Canadian Public Policy, 10(2), 119–137. Irland, L. (1986). Resource economies in emerging free trade. In: V. Konrad, L. Morin & R. Erb (Eds), Proceedings of a Maine/Canadian Trade Conference (pp. 311–326). Held January 9–10. University of Maine, Canadian-American Centre. Myneni, G., Dorfman, J., & Ames, G. (1994). Welfare impacts of the Canada-US softwood lumber trade: Beggar thy consumer trade policy. Canadian Journal of Agricultural Economics, 42, 261–271. Myre, P. (1998). Changing forest values, forest legislation, and management in Canada in forest management into the next century: What will make it work? In: Proceedings of a Conference
262
HARRY NELSON AND ILAN VERTINSKY
Held in Spokane, Washington (pp. 39–45). Held November 19–21, 1997. Forest Products Society, Madison, WI. Natural Resources Canada (2002). The state of Canada’s Forests 2001–2002. Ottawa. Natural Resources Canada (2003). Special data compilation. Polzin, P. (1994). Spatial distribution of wood products industries. Journal of Forestry (May), 38–42. Prestemon, J., & Buogiorno, J. (1997). Comparative advantage in U.S. interstate forest products trade. Journal of Forest Economics, 3(3), 207–228. Reed, L. (2001, May). Two centuries of the softwood lumber war between Canada and the United States: A chronicle of trade barriers viewed in the context of sawtimber depletion. Report prepared for the Free Trade Lumber Council. Montreal. Ritchie, G. (1997). Wrestling with the elephant: The inside story of the Canada-US trade wars. Toronto: MacFarlane & Ross. Ross, M. (1995). Forest management in Canada. Canadian Institute of Resources Law, University of Calgary. Saunders, J. (2003a, March 4). The root of the lumber spat: How U.S. timberland is valued. Globe and Mail, p. B1. Saunders, J. (2003b, March 27). U.S. lumber auction has critics. Globe and Mail, p. B6. Spelter, H., & McKeever, T. (1999). Profile 1999: Softwood sawmills in the United States and Canada. Res. Paper FPL-RP579. Madison, WI: Forest Products Laboratory. Taylor, R. (2001). The threat from substitutes. Wood Markets Monthly August. Vancouver, BC. U.S. Department of Agriculture (2002). U.S. forest facts and historical trends. Accessed at http:// fia.fs.fed.us/library/ForestFacts.pdf. U.S. Department of Commerce (2003, January 6). Proposed analytical framework softwood lumber from Canada. Accessed at http://www.for.gov.bc.ca/het/softwood/softwood lumber framework.pdf. U.S. International Trade Commission (1999, October). Conditions of competition in U.S. Forest Products Trade. Report on Investigation No. 332–400 under Section 332(g) of the Tariff Act of 1930. Publication 3246. Washington, DC. Van Kooten, G. C. (2002). Economic analysis of the United- States-Canada softwood lumber dispute: Playing the quota game. Forest Science, 48(4), 712–721. Wear, D., & Lee, K. (1993). U.S. policy and Canadian lumber: Effects of the 1986 memorandum of understanding. Forest Science, 39(4), 799–815. Zhang, D. (2001). Welfare impacts of the 1996 U.S.-Canada softwood lumber trade agreement. Canadian Journal of Forest Research, 31(11), 1958–1967. Zhang, D. (2002). From senators to the president: Solve the lumber problem or else. A paper presented at the 2002 Southern Economics Association Meetings, November 24–26. New Orleans. Zhang, D., & Sun, C. (2001). US-Canada softwood lumber trade disputes and lumber price volatility. Forest Products Journal, 51(4), 21–27.
TRUCKING SERVICES UNDER NAFTA Christine Drennen ABSTRACT NAFTA established guidelines for creating a seamless trilateral trucking market that has proven problematic to implement. Costs of border delays, traffic congestion, added pollution, and documentation requirements stymie North American policy makers. Use of just-in-time supply chain techniques by shippers exacerbates the inability of border infrastructure to meet time-sensitive traffic demands. This paper reviews NAFTA’s transport provisions and explores Mexican, U.S., and Canadian disputes in the context of continental integration. Progress toward harmonizing labor and equipment standards in the course of instituting NAFTA is investigated. Policies for additional border facilities investments and streamlining procedures are compared to heightened national security concerns.
INTRODUCTION The North American Free Trade Agreement is credited with substantial growth of merchandise trade across the Canada-U.S.-Mexico borders. The backbone of merchandise transport is the motor carrier industry, which moves approximately two-thirds of all North American international cargo. The NAFTA timetable for a seamless trilateral trucking market by the year 2000 has not yet been realized. Foreign ownership restrictions and geographic limits on foreign carriers persist North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 263–282 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10014-4
263
264
CHRISTINE DRENNEN
despite NAFTA articles 1202 (national treatment for cross-border services) and 1203 (most-favored-nation treatment for cross-border services) prohibiting these anti-competitive practices. Governments concerned about national sovereignty, security, illegal immigration, and smuggling are challenged to reduce congestion, pollution, and the economic costs of border delays. The goal of swiftly flowing merchandise is seemingly at odds with imperatives of national security and reducing smuggling and illegal immigration. Private business, labor unions, industry associations, environmental groups, and special interest groups have competing perspectives on free trade in trucking services. Labor and environmental groups, at times joined by the White House, oppose NAFTA implementation based on arguments about differences between U.S. and Mexican standards and enforcement of safety and pollution controls. Industry groups counter that waiting in line at the border reduces productivity through higher transportation costs and raises consumer costs. This paper reviews North American trucking services under NAFTA by exploring harmonization of labor and equipment standards and how provision of border infrastructure is proceeding. The changing dynamics of the trucking dispute and border regulatory regimes are assessed in the context of national security. The economic and societal implications of on-going congestion at the border conclude this paper.
BEFORE NAFTA The U.S. and Canadian trucking services markets harmonized as a result of deregulation that began during the late 1970s.1 Transborder trucking did not adjust without controversy, but by the 1988 Canada-U.S. Free Trade Agreement Canadian and U.S. systems had become quite integrated. The U.S. and Mexican trucking markets were not harmonized, and as will be shown, remain separate. In 1982, the U.S. stopped processing operating authority permits for Mexican motor carriers. Five Mexican-based motor carriers have permanent authority to operate in the entire U.S. so long as they meet U.S. regulations (U.S. General Accounting Office, 1996). Otherwise, Mexican carriers with U.S. cargo must remain within a narrow commercial zone along the southern U.S. border. The Mexican and Canadian trucking markets have taken steps towards harmonization. Carriers from Mexico and Canada, hauling Canadian or Mexican bound merchandise, can traverse the U.S. without loading or unloading so long as they meet U.S. regulations. Canada and Mexico signed a memorandum of understanding in 1994 that allowed Canadian carriers to cross into Mexico instead
Trucking Services Under NAFTA
265
of dropping their cargo on the U.S. side for Mexican carriers to haul it across the border. Canadian-domiciled Challenger Motor Freight was the first Canadian carrier to enter Mexico in October, 1994 (Industry Canada, 1998). U.S. carriers gained this privilege under NAFTA in 1995. Mexican foreign investment law restricted foreign ownership of freight carriers to 49%. U.S. and Canadian firms operating in Mexico were obliged to form joint ventures with Mexican firms. In all three countries, for reasons of immigration rules, foreign carriers are restricted from loading and delivering strictly domestic merchandise from one location to another within a host country, because they would then be seen as engaging in the local labor market. The trailer portion of combination tractortrailer trucks is considered merchandise and this distinction creates logistical challenges for the movement of empty trailers. Canada and the U.S. enjoy limited cabotage privileges with one another but not with Mexico.2 Canadian-based drivers entering the U.S. with southbound cargo can avoid an empty return trip by pre-arranging a backhaul load of international cargo. In addition, drivers can haul one load between the international drop off and pickup points. A Canadian driver can, for example, drop off at location A, pickup an empty trailer at point B (nearby), travel to location C (pre-arranged) to drop off the trailer and pickup a load bound for Canada. This “incidental” domestic haulage is permitted with certain restrictions. Ongoing domestic service and soliciting return loads once in the U.S. are both prohibited. The reverse situation applies to U.S. drivers in Canada. Since free trade in trucking does not exist with Mexico, the result is a costly, inefficient “drayage” system. Long-haul trucks from Mexico’s interior arriving at the border offload their contents onto short-haul trucks that shuttle back and forth across the border and remain within a narrow commercial zone. Northbound merchandise is then loaded onto a U.S. or Canadian carrier to complete the journey. Southbound cargo is subject to this same system. The result is that approximately half the drayage trucks return across the border unloaded. In contrast, completely empty backhauls on the northern border are estimated to account for 17% of all crossings (Transport Canada, 1998).
NAFTA PROVISIONS The intent of NAFTA regarding trucking services is to lift restrictions on geographic market and ownership. The integration hoped for under NAFTA led some to predict that over the next twenty-five years “The border between Canada and the United States will be no more intrusive than the German-Dutch border today” (Hufbauer & Schott, 1998). The following timetable for phasing
266
CHRISTINE DRENNEN
in the principle of national treatment for market operating authority and then for ownership was established: 1994: NAFTA goes into effect; 1995: U.S. southern commercial zone extended to encompass statewide access to Texas, New Mexico, Arizona, and California; 1995: 100% Mexican ownership of U.S. based companies and 49% (minority) ownership of Mexican firms by U.S. companies; 2000: Total tri-national access to U.S., Mexican, and Canadian trucking companies across the NAFTA region for both pickup and delivery; 2001: 51% (majority) ownership of Mexican firms by U.S. companies; and 2004: 100% (unrestricted) cross-border ownership.
THE NAFTA TRUCKING DISPUTE The NAFTA trucking dispute is a story of frustration and repetition. Opposition and delay have followed negotiation and agreement. Implementation deadlines have come and gone with very little change in how goods are transported across the southern U.S. border. The steps in the dispute are outlined below with an exploration of the merit of the arguments for delaying NAFTA trucking rules. The first delay occurred when U.S. actors prevented Mexican-based trucking companies from accessing the U.S. market. Deficiencies in the Mexican regulatory regime were noted as the initial reason, i.e. the inability of U.S. inspectors to verify Mexican carrier safety compliance. The Clinton administration responded in 1995 by not processing any Mexican-based applications for operating permits for points outside the border commercial zone. In that same year, Mexican ownership of U.S. companies was prohibited amid concerns over U.S. highway safety. As of 2002, about 130 Mexican applications for U.S. national operating authority awaited processing. Over 850 Mexican carriers have applied to operate in the commercial zone, of which approximately half have been granted permission. Safety is measured by the rate at which vehicles or drivers are taken outof-service upon inspection. Inspection criteria include driver qualifications (i.e. commercial licensing, drug testing, hours of service) and equipment standards (i.e. vehicle weight and dimensions, parts and accessories needed for safe operations). Fatalities associated with large trucks have fallen over the last thirty years. In 1997, U.S. inspections found the out-of-service rates for trucks was 44% for Mexico, 25% for the U.S., and 17% for Canada (U.S. Department of Transportation, 1998). Mexican trucks inspected were the short-haul drayage variety. These trucks are not as well maintained as long-haul vehicles and lead
Trucking Services Under NAFTA
267
to a higher out-of-service rate. Mexican trucks that would be found on U.S. highways under NAFTA would be the long-haul variety and are speculated to be more safety compliant (U.S. International Trade Commission, 2002). By 2000 the Mexican out-of-service rate fell to 37% while the U.S. rate barely fell to 24%. Three out of four of motor carriers operating in the U.S. have not been subjected to a compliance review (U.S. Department of Transportation, 2001). The Clinton White House policy was inspired by the International Brotherhood of Teamsters Union and strongly supported by other anti-NAFTA groups. Special interest groups such as Parents Against Tired Truckers (PATT) also welcomed this development. The Canadian Trucking Association (CTA) and the American Trucking Association (ATA) vociferously opposed U.S. policy. The ATA was unable to gain much support from the Democrat-controlled administration that was courting labor as part of its political election campaign strategy. The Mexican government, with Canadian support, argued before a NAFTA dispute resolution panel that the U.S. had erected a non-tariff trade barrier under the guise of safety concerns. While the outcome was pending, the 2000 federal elections ushered in new leadership in Mexico and the United States. The Bush administration attempted to fulfill a campaign promise and rescind the ban on Mexican carrier operations and ownership. Congress stopped this move by withholding funding needed to process Mexico permit applications. In early 2001, the NAFTA arbitration panel ruled in favor of Mexico concluding that the U.S. was in violation of the NAFTA’s national treatment provision. It found that motor carriers operating in any NAFTA country were subject to all applicable domestic safety standards. The U.S. could only prevent Mexico carriers from operating within the U.S. on a case-by-case basis and not in blanket fashion, as had been the practice (NAFTA Secretariat, 2001). A pro-Mexico U.S. administration, an assertive new Mexican president, and a NAFTA panel ruling were key ingredients needed to open the border. Before NAFTA provisions for cross-border trucking could be implemented, the U.S. Congress established twenty-two new safety conditions that took effect in May, 2002. Finally, in November, 2002, the U.S. Department of Transportation (U.S. DOT) was ordered, by Presidential determination, to begin processing Mexico operating authority applications. Export dependent Mexico instituted regulations modeled on U.S. standards coupled with more highway supervision and increased fines. Safety measures such as vehicle weight and dimensions and limits on emissions have been developed and implemented (U.S. General Accounting Office, 2001). Circumstances seemed favorable for ending the delay now entering its eighth year. In January, 2003 the 9th Circuit Court of Appeals issued a ruling that again delayed the opening of the southern U.S. border to Mexico carriers. The case,
268
CHRISTINE DRENNEN
based on environmental concerns, was brought before the court by a coalition of labor and other special interest groups including the Teamsters. The court has indicated that the U.S. DOT needs to conduct an environmental impact statement according to 1970 National Environmental Policy Act (NEPA) and 1990 Clean Air Act standards. The failure rate for emissions testing at the California border in 2000 was 8% for U.S. trucks and 12% for Mexican (drayage) trucks (U.S. General Accounting Office, 2001). The U.S. and Mexican trucking services markets have yet to operate differently from the years before NAFTA. An exploration of the international motor carrier market dynamics and industry structure precedes an analysis of the economic and societal implications of continued delay.
THE TRUCKING SERVICES MARKET Trucking services is a derived demand market that is vulnerable to short-term business cycles. Its performance is often used as a leading economic indicator. An important market in its own right, the share of gross domestic product accounted for by trucking (including warehousing and storage) is roughly 1.5% of Canadian, 1.2% of U.S. economy, and nearly 4% of Mexican GDP. Even without the benefit of NAFTA, U.S.-domiciled carriers claimed 10% of the $6.4 billion Mexico trucking services market (U.S. Department of Commerce, 1998). The scale and growth of the overall industry are depicted in Fig. 1. Trucks and highways are the backbone upon which the largest share of NAFTA trade flows. Shipping freight by train lacks some scheduling and pricing flexibility
Fig. 1. Total U.S.-Canada-Mexico Trade by Truck (U.S.$ Billions). Data source: U.S. Department of Transportation, Bureau of Transportation Statistics, Transborder Surface Freight Data.
Trucking Services Under NAFTA
269
Table 1. Domestic, International, & NAFTA Activity (Average Annual Growth Rate 1995–2000). Importsa (%)
USA Canada Mexicoc
Exportsa (%)
GDPb (%)
World
NAFTA
World
NAFTA
2.7 2.5 3.1
7.9 6.5 10.2
9.7 6.2 11.5
6.2 7.4 14.5
10.7 9.1 15.1
a OECD
(2001). International Trade by Commodities Statistics. (July, 2001). National Accounts of OECD Countries, Main Aggregates, Vol. 1. c 1994–1999. b OECD
so critical to just-in-time production methods limiting the demand for rail transport. Air cargo, though fast growing, is often reserved for higher value-added products. Marine and pipeline claim the smallest share of NAFTA trade. Trilateral trade has evolved under NAFTA in three main ways. First, NAFTA trade is growing two to three times faster than domestic gross product. Second, the countries of NAFTA are becoming more economically dependent on one another. Finally, swelling border activity necessitates adequate facilities and staffing to manage the increased traffic volume. North American two-way trade growth outpaces domestic and other international trade as detailed in Tables 1 and 2. Consequently, international trucking is growing faster than domestic trucking. In the case of Canada, the international market accounted for 30% of total trucking in 1989 (by volume) and increased to 48% by 1999 (OECD, 2002). Growth is expected to continue over the next decade as trade and integration deepen. In addition to growing demand, the intensity of demand for trucking services is increasing. Growing cross-border reliance among production facilities brought about through intra-firm trade, intra-industry trade, and tighter supply chain Table 2. Trade Dependence of NAFTA Partners (NAFTA Trade as a Percentage of World Trade). Imports (%)
USA Canada Mexico
Exports (%)
1995
1999
1995
1999
27.4 70.9 76.4
30.1 72.0 76.3
28.9 79.6 84.9
36.4 87.2 89.8
Data source: OECD (2001). International Trade by Commodities Statistics.
270
CHRISTINE DRENNEN
Table 3. NAFTA Truck Traffic, 2000. Merchandise Value (Millions Current U.S.$)a
Number of Crossingsb
U.S.-Canada Detroit-Windsor
$257,642 $85,914 (33%)
7,048,128 1,769,389 (25%)
U.S.-Mexico Laredo-Nuevo Laredo
$171,058 $60,046 (35%)
4,525,579 1,493,073 (33%)
a U.S.
Department of Transportation, Bureau of Transportation Statistics, Transborder Surface Freight Data. b Special tabulation, August 2001.
relationships all raise the intensity of trucking demand relative to the value of final products. NAFTA cargo traffic is concentrated on a relatively small number of crossings, with the busiest northern and southern gateway (as shown in Table 3) each accounting for one-third of the traffic on their respective borders. Seventy percent of Canada-U.S. trade funnels through just six crossings. Comparing northern and southern U.S. land ports suggests that northern ports are considerably more productive. Four lanes connect the Detroit-Windsor gateway, whereas eighteen lanes serve the Laredo-Nuevo Laredo gateway. Bottlenecks in border zones have forced the industry to adapt. To meet growing demand in the face of unpredictable developments such as September 11th and lingering U.S.-Mexican delays, the trucking industry continues to evolve.
INDUSTRY STRUCTURE International trucking is conducted by three, somewhat overlapping, segments of the motor carrier industry: the express-package, own-account, and for-hire segments. Some companies operate in more than one segment and owner-operators can contract for work in more than one segment. Despite these features there are broad but important differences between these segments. They are largely not in competition with one another as they vary in terms of type of customer, cost structures, driver-employer involvement, equipment, and insurance matters. Each segment has large well-known firms and many small firms. Drivers employed within the first two segments operate equipment owned and maintained by their employers. Employers, rather than the individual drivers, typically manage insurance and licensing procedures. The NAFTA trucking dispute is focused on
Trucking Services Under NAFTA
271
the highly competitive for-hire segment. This segment accounts for more than half of all cargo transport. (1) The Express-Package segment consists of firms such as United Parcel Service (UPS) and FedEx Corporation. These companies are in the courier, package, and parcel business to transport letters, envelopes, pouches, and boxes weighing less than 100 pounds under expedited conditions (although they both have subsidiaries that compete in other segments). They compete on same day and next day delivery schedules. Companies in this segment are often involved in air as well as road transport, and have myriad cargo sorting hubs and networks. Step vans are most common with tractor-trailer combination trucks making up a small percentage of the overall fleet. UPS alone has more trucks in its fleet than the entire number of trucks in Mexico. Its revenue is about seven times the largest for-hire carrier. (2) The Own-Account segment consists mostly of dedicated drivers of large manufacturing firms such as the big three automakers. There are numerous large and small own-account fleets in the U.S. These in-house fleets are employees of a manufacturer whose primary business is not cargo transport. The big three automakers and large retailers (Wal-Mart, Canadian Tire, etc.) have fleets dedicated to their own accounts. Their identifiable trucks are another form of visibility for these companies. NAFTA documentation, customs, immigration issues, etc. are generally handled by the parent company. When used exclusively for their own companies these fleets often have empty and expensive back haul miles. Some are now getting into the for-hire business and booking shipments from third parties for backhaul. (3) The For-Hire is a crowded segment consisting of self-employed owneroperators and trucking companies (Roadway Express, Canadian Freightways, etc.) with fleets of drivers who contract with manufacturers and other shippers. This segment has two sub-sectors, truckload (TL) and less-than-load (LTL), each with slightly different operating characteristics.
FOR-HIRE COMPETITIVE ATTRIBUTES A Canadian survey revealed that there are about 8,000 for-hire companies (Industry Canada, 1998). The Mexico for-hire segment is assumed to be much smaller and the U.S. undoubtedly larger. As its name implies, the TL sector serves a single shipper at a time with truckload quantities (over 10,000 pounds) that frequently go straight from origin to destination. This sector possesses many aspects of a
272
CHRISTINE DRENNEN
perfectly competitive market. TL carriers provide an undifferentiated product and compete on price. Few barriers to market entry in this sector – essentially a license, insurance, and a truck – mean that there are more carriers than in the LTL market. Driver shortages combined with increasing driver regulations have not translated into higher wages because new entrants have similar skills to incumbents. Existing carriers have little incentive to react to a new entrant since any single firm or driver has little impact on the overall multi-firm industry. Barriers to exit are present in the form of sunk costs for licensing and equipment, but may also be minimal under the assumption that equipment can be sold at an amount close to its remaining value. These characteristics indicate that TL trucking approximates a perfectly competitive market. The LTL sector handles smaller than truckload shipments. Small shipments from many customers are consolidated onto the same truck. Owing to its smaller individual shipments this sector is characterized by a system of terminals and platforms for consolidating shipments bound for similar destinations. The terminal system adds roughly 25% to operating costs over the TL sector and so favors larger carriers that are able to bear this added cost. This barrier to entry enables incumbents to enjoy a less competitive market than the TL sector. Variation on the location of terminals creates some product differentiation among the LTL sector, and enables firms to compete on both price and timing. Speed is a differentiator and companies market themselves based on number of days from pickup to delivery. Competition to serve time-critical shipper needs results in tightly scheduled delivery windows that are expressed in terms of hours, not days. The extra cargo handling and paperwork necessitated by the terminal system make quality and reliability important features. Competitive pressures in both the TL and LTL sectors have kept real prices lower than might be expected even amid growing demand. In Canada, 2000 overall trucking prices were below 1996 levels. Transborder trucking prices increased slightly during the same period, reflecting the relatively greater demand growth in the international sector and a premium for expertise needed to deliver across the border. Operating ratios (operating expenses as a percent of operating revenues) have held steady during the 1990s at 94.5 to 97.6%. This industry has experienced productivity gains but also copes with rising insurance and fuel costs (Transport Canada, 2001). Since trucking firms are price takers and real prices have been declining domestically, carriers have found ways to preserve their margins especially by limiting empty backhauls. Third party logistics firms (so-called infomediaries) make information about loads in need of carriers and carriers in need of loads available on an electronic internet exchange. This increased degree of cooperation between and among shippers and carriers has so far skirted antitrust attention and is a means of adaptation to trade growth in an uncertain regulatory environment.
Trucking Services Under NAFTA
273
WAGES AND LABOR The same industry traits that hold down prices also put downward pressure on wages. The two largest variable costs in trucking are wages and fuel. Current exchange rates give Canadian and Mexican drivers a built-in price advantage. The perception of lower wages in Mexico is the primary motivation for the Teamsters’ role in delaying free trade in trucking. The size of the wage differential is not as great as is commonly quoted in the press. Wage statistics on the driver occupational class across all three countries are not easily compared. Dissimilar levels of industrialization obscure information about wages, hourly compensation costs, value added per person, and other measures. Different working conditions including hours worked per day or week, social insurance expenditures, and labor taxes vary greatly between Canada, Mexico, and the United States. Cross-country employee turnover and training expenses are not captured in wage reporting. Were these items accurately accounted for the differences in pay-scales would be less but, given the size of the wage gap, would still favor Mexico. In one estimate, Mexican drivers earn 13 cents per mile, while U.S. non-union drivers earn 31–40 cents per mile and U.S. union drivers earn 50 cents (plus 25 cents more in benefits) per mile (Schultz, 2002). Employee turnover is a large cost in some U.S. trucking firms. The average firm replaces the equivalent of its entire labor force each year. The cost to replace a driver ranges from 3,000 to $12,000 and averages about $5,000. Thus, a 2,000-driver firm with 100% turnover incurs expenses of $10 million per year. Industry-wide annual driver turnover expenses are estimated to be at least $3 billion (Keller & Ozment, 1999). Organized labor continues to influence the industry even though the Teamsters now represent about 25% of the U.S. drivers, down from 60% in the early 1970s. The Teamsters bargain with large unionized carriers in the Motor Freight Carriers Association to set wages and benefits in the National Master Freight Agreement. Smaller unionized carriers then follow suit. Union members typically work fewer hours, are paid for overtime and time spent waiting during loading, and enjoy several other benefits that most of their non-union counterparts do not. Mexico has two labor unions. CONATRAM (the Mexico Confederation of Carriers) represents smaller carriers and has gained attention for work stoppage activities through road blockades. This union, like the Teamsters, is opposed to NAFTA rules for trucking – although for entirely different reasons. Whereas the Teamsters fear the influx of a large pool of experienced Mexican drivers accustomed to lower wages, CONATRAM fears the entrance of large, well-capitalized American carriers able to buy small Mexican carriers and put them out of business. CANACAR (the Mexican National Cargo Chamber) is the union of large Mexican carriers and
274
CHRISTINE DRENNEN
supports NAFTA as do the industry groups the ATA and CTA (American Trucking Association and Canadian Trucking Association).
LOGISTICS, TECHNOLOGY AND EQUIPMENT Shipper needs for faster transit times and trends toward more frequent shipments raise the importance of trucking in the supply chain. Carriers that use logistics software, sophisticated communications, and have better equipment gain an advantage. U.S. and Canadian firms are adopting electronic commerce faster than their Mexican counterparts. Internet order placement, track-and-trace, and billing allow carriers to add value to the supply chain. Transponders and onboard communications systems are common in the U.S.-Canada setting. Carriers are addressing the problem of losing track-and-trace capacity once loads have crossed into Mexico (Hansen Harps, 2002). Infomediaries use the internet to establish electronic exchanges that match shipments to carriers. Changes in the transborder environment are putting pressure on government agencies and all links in the supply chain to communicate electronically under a uniform system. Such a system is still under development, suffering from the complexities of numerous reporting needs of varied agencies on both sides of the border coupled with the added costs such a system may impose shippers, carriers, and governments. Newer equipment has fewer maintenance requirements and gets better mileage creating a competitive advantage in Canada and the U.S. over Mexico. Such vehicles are also likelier to be able to pass EPA emission inspections and FMSCA safety inspections. U.S. EPA standards are scheduled to tighten further in 2004 and 2007. Pollution control is especially relevant to trucking. Trucks pollute disproportionately more than cars. According to one analysis, large trucks and buses generate 25% of smog forming pollutants and 58% of particulate matter despite that they represent less than 6% of total vehicle miles driven each year. The cleanest diesel truck equates to 31 passenger cars in terms of generating smog forming pollutants and 395 passenger cars in terms of particulate matter (soot) emissions under EPA standards for 2004 (Mark & Morey, 2000). The recent U.S. 9th Circuit Court of Appeal’s decision delaying NAFTA trucking demonstrates the growing importance of passing emissions inspections.
INFRASTRUCTURE OF TRADE Merchandise inspection facilities, information management technologies, along with roads, bridges, and toll plazas, and trained personnel are all part of the
Trucking Services Under NAFTA
275
infrastructure of trade by truck. The Immigration and Naturalization Service, Border Patrol, and Customs Service – now combined into the Department of Homeland Security – the departments of Agriculture, Justice, and Transportation are some of the (U.S.) agencies involved in border policy and infrastructure. The mismatch between infrastructure supply and demand results from trade growth unaccompanied by sufficient investment in border facilities. Immigration aspects also exacerbate burdens at the border. Inadequate financial resources have reduced the effectiveness of overall border policy. Although programs put forth in the current era of heightened national security show promise, questions and criticisms remain. Trilateral production capacity has been built through the preferential trade and investment treatment that each NAFTA signatory gives the other. To limit the application of NAFTA to the three countries and guarantee NAFTA preferential tariffs, product origin certification is needed. Without such rules, manufacturers from other countries could use Canada, the U.S., or Mexico as a platform for investment or transshipping product. A foreign manufacturer could establish such facilities to circumvent the objectives of NAFTA. Similarly, a non-North American firm could determine which of the three NAFTA countries offers the lowest tariff and use that country to transship merchandise through to the other two. Preferential tariffs and rules of origin for product content require not only documentation, but also require a means of enforcement – that being physical inspections and review of documentation. The feasibility of moving such inspection procedures away from clogged gateways is under consideration. NAFTA also defines entry procedures for professionals, intra-company transferees, and business visitors. Millions of people in passenger vehicles cross the same U.S.-Canada land borders used by commercial trucking. Merchandise and services provisions of NAFTA facilitate tri-lateral economic integration while overburdening physical and electronic border infrastructure. Regular delays, so-called “wait times,” add millions to cross-border business costs. Environmental degradation generated by traffic congestion, particularly by idling diesel trucks, represents a great cost to society, especially border communities, in terms of air quality and health problems. Policies that reduce the number of inspections at the border do not imply trimmed border delays. The most prepared and secure carrier in line behind an unprepared carrier still waits just as long. Policies that reward prepared travelers and carriers with dedicated lanes and expedited processing add value and provide an incentive to carriers and shippers to make the investments in communications upgrades needed to participate. Even before September 11th there was wide recognition that border crossing facilities were congested and procedures could not keep up with traffic demand. North American governments struggle to develop policies to facilitate current
276
CHRISTINE DRENNEN
trade and predicted growth. The 1995 U.S.-Canada Shared Border Accord, the 1997 Border Vision initiative on immigration control, and the 1999 Canada-U.S. Partnership (CUSP) are representative of collaborative attempts by private industry and selected levels of government to draw attention to the lack of resources, complications, and extraordinary costs of doing business across the northern U.S. border. The NAFTA Super-Highway concept calls for North-South corridors to aid truck traffic flows through and within the NAFTA region. These and many other creative initiatives have improved the movement of people and goods, yet the financial and human resources to implement these various proposals have not been sufficient. While the economies of late 1990s North America were in the midst of an economic boom, governments were not overly focused on infrastructure investment. The post-September 11th era awoke public support to enhance national security through border infrastructure investments and policy changes. Immigration rules came under scrutiny first. Canada and the U.S. have tightened visa screening, passport security, and admissibility standards, yet each maintains its own immigration policy. The recognition that security and immigration policy could severely hamper economic performance led to several initiatives, reminiscent of the proposals put forth in the late 1990s. The Customs-Trade Partnership Against Terrorism (C-T PAT) is a collaborative initiative between business and government. The traditional private sector supply chain of shipper-manufacturers, brokers, and carriers is enlarged to encompass government Customs agencies. Customs audits the links of a particular supply chain based on anti-terrorism security measures. Companies certified by the C-T PAT program are eligible for reduced number of inspections at the border. The Canadian-U.S. Smart Border Declaration of 2001 proposes new levels of inter-governmental cooperation and business-government collaboration. The initiative recommends a thirty-point action plan covering security, flow of people and goods, infrastructure and information sharing, all with an emphasis on enforcement. The plan commits financial resources for personnel and data systems to reduce duplication of in-bound and out-bound country information queries and infrastructure to separate prepared from unprepared carriers. Trade-related points in the Smart Border Declaration are: Harmonized processing through complementary national systems that reduce the process to a single step; Clearance procedures are moved away from the border in an attempt to reduce congestion through pre-clearance procedures at the point of origin; Joint facilities and dedicated lanes; Data sharing of information critical to concerns about customs and immigration;
Trucking Services Under NAFTA
277
Transponder and electronic tracking; and Increased border personnel. Initially, sixty companies piloted a project involving advance e-notification of Customs officials so they can do vehicle, driver, and cargo risk-assessment before the truck reaches the border checkpoint. Manufacturers not closely connected with their suppliers or that have numerous small suppliers will find such a program more difficult to implement than counterparts in other industries such as the highly integrated automotive sector. High-level government officials from the U.S. and Canada made several joint announcements on policy to facilitate trade while strengthening security during 2002. Public resources are promised for more personnel, better computing ability, and additional facilities to enhance functionality at the border. The private sector is working to improve security through tamper resistant and tamper evident packing and sealing techniques and investments in supply chain communications. Mexico and the U.S. recently demonstrated effective international cooperation when they moved to meet the safety requirements established by the U.S. Congress in less than twelve months. Resources were committed to harmonize computer systems, conduct carrier safety audits, check driver credentials, and acquire land to build inspection facilities at the border.
IMPLICATIONS Growth in international trucking will continue to exceed growth in domestic trucking. According to one analysis, U.S. for-hire trucking is expected to increase by 39% (in volume) from 1996 to 2006, led by trade growth with Mexico and Canada (Logistics Management, 1998). Developed before September 11th, this projection now seems optimistic. Even under a more moderate growth scenario, trucking will continue to be the riverbed on which growing trade will flow. How much it will grow hinges on border checkpoint and trade facilitation policy adopted by NAFTA governments and adaptation by the trucking industry. The three North American countries have a strong interest in facilitating trade and raising quality of life on the continent while maintaining national sovereignty and security. Any disruptions to the road freight system, such as the temporary border shutdown immediately after September 11th, will thwart the best border management and bring selected sectors to a near halt. Several implications for producers and consumers, labor and management, and the trilateral trucking industry follow once NAFTA is implemented on the southern U.S. border. They are summarized and explained below.
278
CHRISTINE DRENNEN
(1) Trade patterns among the NAFTA countries will shift as Mexico becomes relatively more important as a consumer market. (2) The composition of the trucking industry will adapt to changing trade patterns by ingraining itself more deeply in the producer value chain. (3) Working conditions themselves will change in response to new regulations. Downward pressure on wages of drivers from the U.S. and Canada will create incentives to maintain income through longer hours. (4) Logistics and technology will play a larger role in the trucking business as firms respond to border management policy and increased information needs of shippers. (5) Equipment costs will rise as firms move to stay ahead of increasingly stringent air quality regulations and remain in compliance with safety standards. (6) Provision of physical infrastructure, particularly NAFTA superhighway corridors traversing the U.S., and additional lanes and inspection space will be increasingly difficult to accomplish and will remain a primary constraint to international trucking. As Mexican industrialization and consumption increase the U.S. and Canada are well positioned to participate in this growing market. Trade amongst the three NAFTA countries will gradually shift as the demographically younger Mexican population moves into higher consumption stages of life and the consumption of tradable goods by aging populations in the U.S. and Canada declines. The Mexican population is three times that of Canada giving strength, in the long-run, to the economic force of Mexico. Market integration has raised trade and it will continue to outpace domestic growth. The U.S. will become more reliant on its NAFTA partners due to four reasons. Increasing tri-lateral integration, the U.S.’s relatively lower level of trade dependence, industrialization in the Mexican interior, and growth in Mexico’s consumer market are crucial growth factors for the U.S. economy. Mexico is projected to overtake Canada as the largest U.S. trade partner, and an open border will hasten that inevitability. A border where all the provisions of NAFTA are invoked will change price and demand characteristics beyond demographic factors. Lower transportation costs associated with more efficient border procedures (no more drayage) and the enhanced ability of for-hire firms to integrate their service with Mexican shippers will result in cheaper Mexican imports to the U.S. and Canada. Consumers in the U.S. and Canada and producers in Mexico will benefit most. U.S. and Canadian goods will likewise be cheaper in Mexico through reduced border processing costs and better shipper-carrier integration. Use of lower wage drivers from Mexico will have implications for both wages and driver regulations. Organized U.S. and Canadian labor has long
Trucking Services Under NAFTA
279
feared the threat of the low-wage Mexican driver pool. If lower wages are fully translatable into lower shipping rates (debatable), Mexican drivers are likely to enjoy an enormous advantage in the short-term over U.S. and Canadian union and non-union drivers. This wage advantage has a tendency to be overstated as explained above. Strong demand for lower wage Mexican drivers will cause Mexican wages to rise while putting downward pressure on U.S. and Canadian wages. Incentives to cheat on maximum number of hours and logbooks will be greater as more drivers, especially independent owner-operators, attempt to make up for lost wages through driving more hours and faster. The result will likely be stricter rules and stiffer penalties. The U.S.-Canadian experience in the decade following liberalization of trucking services is that regional markets have become prominent. Two-thirds of all cross-border truck trips had origins or destinations in the Great Lakes and Northeast states and central Canada (Transport Canada, 1998). Regional markets will be even more important in the U.S.-Mexican realm given the concentration of the maquiladora industries along the border. This suggests that drivers based in California, Arizona, New Mexico, and especially Texas are likely to most experience effects of the entry of Mexican drivers. The Mexican wage advantage is further blunted by the increasing importance and expense of technology and equipment. Even for large carriers, investment outlays for electronic notification of container tampering and in-cab communication capability with border personnel will challenge the industry. To make these private sector investments pay off governments also need to invest in infrastructure and technology. The older Mexican fleet will have to be upgraded to be competitive with the communications, safety, and pollution control features of U.S. and Canadian vehicles. The influence of technology means trucking companies, logistics providers, freight forwarders, and brokers will become more integrated through merger and acquisition activities. Electronic communication with government agencies will facilitate this integration. While a certain amount of empty backhauls will remain in the system, the reduction of empty miles will boost productivity. Ironically, the environmental groups that joined with the Teamsters to create the latest delay in opening the U.S.-Mexican border are helping to sustain a system that results in more pollution. As long as drayage remains a reality, the proportion of empty trucks on the roads and bridges connecting the U.S. and Mexico will remain the same, creating more congestion and more pollution. The gradual tightening of air quality standards and deteriorating performance of pollution-control features over the operating life of a truck imply a growing role for the EPA. Roadside emissions testing will likely increase as the EPA moves to enforce pollution reduction targets.
280
CHRISTINE DRENNEN
With respect to safety, it was previously noted that one in four U.S. trucks are out of compliance and most (76%) go uninspected. This troubling record leaves very little room to criticize Mexican truck safety. Companies in the industry recognize that a good safety record is a valuable marketing tool and have developed training and reward programs for drivers. Additional inspections at the border are likely, however funding for more safety inspections in the interior of the U.S. seems less probable. The states are short on inspection resources and facilities and federal funding is currently preoccupied with border enforcement. Road links, more than rail links, are the current focus of public sector infrastructure improvements. No single corridor, such as I-35 or I-69, has been singularly defined as a NAFTA superhighway. Instead, several corridors have been so named by local highway advocates and municipal and state governments. Responsibilities for road building are dispersed among many agencies and jurisdictions. Strong coalition building and regional land-use policies might assist in completing some of these projects, but various coalitions are already fighting between one another for federal funds. Citizen and environmental groups in the anti-road community have delayed many projects by years. The 2003 Transportation Equity Act for the 21st Century, a $216 billion, six-year bill, will dictate how much investment will occur in road-building and border infrastructure. Integrated North-South transportation systems will require construction of new terrain highways and additional lanes on existing corridors – mostly within the United States. U.S. funding will support these new and expensive road projects. Benefits of these projects will accrue to Canada and Mexico through faster transiting times and less fuel consumption yet these federal governments will not have to pay for the projects. Provinces and states on the border are left to cope with unfunded security mandates and increased infrastructure costs. An unknown force that will assuredly influence NAFTA trucking is the newly created Department of Homeland Security (DHS). The DHS merges 22 agencies and 170,000 federal employees. The Department of Transportation and the Customs Service will give some of their authority to the DHS. To succeed in its mandate, DHS proposals for new security measures will need to be reviewed and commented on by industry to insure that they are feasible in the cross-border just-in-time shipping environment. The “common perimeter” concept in North America has sparked controversy and debate by implying that all three countries adopt the same standards for entry. Common standards for identity cards, refugee and immigration policy, shared command of customs and immigration personnel, etc., require that the three countries share common values regarding these issues. Supporters claim this would speed the delivery of merchandise and enhance security, pointing to Europe as a model. Detractors argue that the sovereignty forfeited is too great given
Trucking Services Under NAFTA
281
overwhelming U.S. power. Rather than the perimeter concept a goal to create security confidence through harmonization of reporting methods and processes, sharing of some facilities, and moving physical inspections away from congested gateways with limited space for infrastructure expansion addresses many of the same problems with less political controversy. A fully implemented NAFTA and an integrated trucking services industry will not revolutionize shipping. Borders will remain intrusive checkpoints for smuggling, immigration, and customs. Policies must universally reduce wait times at the border and also speed merchandise through dedicated lanes other new infrastructure, and technology driven risk assessment. Otherwise carriers will be right back where they were – waiting in line.
NOTES 1. See Daniel Madar’s Heavy Traffic: Deregulation, Trade, and Traffic in North American Trucking (UBC Press, 2000) for a thorough treatment of the evolution of trucking in Canada and the U.S. since the 1930s. 2. Cabotage is the transport of cargo between two domestic points by a foreign carrier.
REFERENCES Hansen Harps, L. (2002). Crossborder transport 2002: It’s a whole new ballgame. Inbound Logistics. 17 July 2001. http://www.inboundlogistics.com. Hufbauer, G. C., & Schott, J. J. (1998). North American economic integration: 25 years backward and forward. In: Canada in the 21st Century, No. 3. Industry Canada Research Publications. ix. Industry Canada, Canadian Trucking Research Institute (1998). Trucking in Canada – A profile. 17 July 2001. http://www.strategis.ic.gc.ca. Keller, S. B., & Ozment, J. (1999). Managing driver retention: Effects of the dispatcher. Journal of Business Logistics, 20(2), 97–119. Logistics Management and Distribution Report (1998). A Decade of Growth Seen for Carriers, 37(3), 33+. Mark, J., & Morey, C. (2000). Rolling smokestacks: Cleaning up America’s trucks and busses. Union of Concerned Scientists. 17 July 2001. http://www.ucsusa.org. NAFTA Secretariat (2001). Cross-border trucking services. File No. USA-MEXICO-98–2008–01. 18 July 2001. http://www.sice.oas.org. OECD (2002). Reviews of regulatory reform: Canada. Paris. Schultz, J. D. (2002). A race to the bottom? Traffic World, 266(9), 22–24. Transport Canada (1998). Transportation and North American trade. Transport Canada (2001). Transportation in Canada – Annual report. 3 March 2003. http://www.tc.gc.ca. U.S. Department of Commerce (1998). Mexican market for trucking services. Retrieved 17 January 2001. http://www.strategis.ic.gc.ca.
282
CHRISTINE DRENNEN
U.S. Department of Transportation, Office of Inspector General (2001). Motor carrier safety at the U.S.-Mexico border. Report No. MH-2001-096. U.S. Department of Transportation, Office of Inspector General (1998). Motor carrier safety program for commercial trucks at U.S. borders. Report No. TR-1999–034. U.S. General Accounting Office (1996). Commercial trucking under NAFTA. U.S. General Accounting Office (2001). North American free trade agreement: Coordinated operational plan needed to ensure Mexico trucks’ compliance with U.S. standards. U.S. International Trade Commission (2002). The economic effects of significant U.S. import restraints. Publication 3591.
COSTS OF THE U.S.-CANADA BORDER John C. Taylor, Douglas R. Robideaux and George C. Jackson ABSTRACT This paper reports on the results of a research project aimed at estimating the costs of border crossing transit time and uncertainty for the U.S. and Canadian economies. The cost estimates are based on a review of prior reports, some 20 site visits to seven key crossings, and 173 interviews of knowledgeable organizations/persons. The key finding is that border transit time and uncertainty are costing some U.S.$4.01 billion, or 1.05% of total 2001 merchandise trade, and 1.58% of truck-based trade levels. The primary implication of the research is that it provides a baseline estimate of costs that can be used in cost-benefit analysis of alternative border management strategies.
INTRODUCTION While the U.S. and Canada have the largest bilateral trading relationship in the world today, little has been done to liberalize or modernize border crossings and processes. In fact, the customs processes that are used to manage the flows in each direction are rooted in a system that was originally developed to collect customs duty and control transportation of goods and peoples. These processes were
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 283–297 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10015-6
283
284
JOHN C. TAYLOR ET AL.
thought to have had a significant impact on the economies of the two countries pre 9/11, and post 9/11 border controls are thought to have increased these impacts. According to most reports, post 9/11 tightening of the border has led to longer transit times and uncertainty about the time that border crossings will take. At the same time, some have questioned the degree of security that the current border management system can provide given the level of interaction across the border on a daily basis. This paper reports on the results of a research project aimed at estimating the actual cost impact of border transit time and uncertainty. An understanding of what these costs actually are will be important to any future effort to quantify the cost benefits of various border management strategies.
U.S.-CANADA TRADE AND TRANSPORTATION AND CONCERNS OVER FUTURE CANADIAN-U.S. TRADE LEVELS Merchandise trade between the two countries totaled U.S.$382 billion in 2001, with truck-borne trade totaling U.S.$235 billion. This trade, and personal travel for both business and pleasure, results in a great deal of border crossing activity. In 2001, 68.3 million personal vehicles crossed the U.S.-Canada border along with 13.4 million trucks. Table 1 provides a summary of this travel by type of vehicle and year, and by key crossing region. Personal vehicle travel was down a cumulative 11.89% from a peak of 77.5 million units in 1995, however, commercial traffic grew 29.7% between 1995 and 2001, and 122.5% over the 17 years since 1984. To put this traffic volume into perspective, consider that personal vehicle traffic in 2001 averaged 11,687 units per hour over a 16 hour day. Truck traffic averaged 3,350 units per hour over a five day per week, 16 hour day. Providing for proper anti-terrorism security checks on all these vehicles while facilitating the free flow of goods and people was a daunting task pre 9/11 and has become all the more difficult since 9/11. Table 1 also points out the extreme concentration in traffic at several key crossings, and the extensive growth in traffic at these crossings. For instance, out of 130 border crossings, the four Ontario-Michigan crossings accounted for 35.9% of the northern border’s total bidirectional truck traffic in 2001. Truck traffic at Windsor-Detroit has grown 133.2% since 1984, and by 446.5% at the Sarnia-Port Huron crossing. It is also important to note that the bulk of this truck traffic crosses the border at locations that turn out to be key points for auto traffic as well, further congesting these crossings and straining the functional capacity of the existing bridges and tunnels. This concentrated volume of truck traffic at equally congested auto crossings points out some of the problems in trying to increase
Crossing
St. Step./Milltown Lacolle 5 Crossings Fort Erie Rainbow Bridgea Queenston Whirlpool Bridgeb
Truck Traffic
Auto Traffic
1984 TRK
1995 TRK
2001 TRK
1995–2001 Cum Grwt (%)
1984–2001 Cum Grwt (%)
1984 Car
1995 Car
2001 Car
1995–2001 Cum Grwt (%)
1984–2001 Cum Grwt (%)
102 458 674 9 490 60
168 570 1146 1 782 6
225 785 NA NA NA NA
33.9 37.7 NA NA NA NA
120.6 71.4 NA NA NA NA
2004 1931 5609 2986 1955 1170
3107 2526 6388 3231 3409 1390
2411 2073 NA NA NA NA
−22.4 −17.9 NA NA NA NA
20.3 7.4 NA NA NA NA
Niagara subtotal
1233
1935
2301
18.9
86.6
11720
14418
14176
−1.7
21.0
Ont.-New York sub
1691
2505
3086
23.2
82.5
13651
16944
16249
−4.1
19.0
Ambassador Bridge Det-Win Tunnel
616 773
2233 275
NA NA
NA NA
NA NA
4360 5413
7492 8163
NA NA
NA NA
NA NA
Windsor subtotal Sarnia
1389 286
2508 1168
3239 1563
29.1 33.8
133.2 446.5
9773 2911
15655 3922
15336 4264
−2.0 8.7
56.9 46.5
Ont.-Mich sub
1675
3676
4802
30.6
186.7
12684
19577
19600
0.1
54.5
155 344 0
263 766 0
403 867 0
53.2 13.2 0.0
160.0 152.0 0.0
270 2656 3614
408 3215 4948
415 NA NA
1.7 NA NA
53.7 NA NA
344
766
867
13.2
152.0
6270
8163
5633
−31.0
−10.2
c
Emerson Pac Highway Peace Archa Brit Col.-Wash sub All Other
2039
2930
3983
35.9
95.3
19831
29251
24011
−17.9
21.1
Total U.S.-Canada
6006
10308
13366
29.7
122.5
54710
77450
68319
−11.8
24.9
285
Source: Statistics Canada and U.S. Customs. a Rainbow Bridge and Peace Arch no longer accept trucks. b Whirlpool was closed through most of May, 2001. c Emerson is estimated for 1984.
Costs of the U.S.-Canada Border
Table 1. Two Way Traffic into Canada and USA by Vehicle Type 000’s of Vehicles.
286
JOHN C. TAYLOR ET AL.
Fig. 1. Total U.S.-Canada-U.S. Economic Activity, Canadian Land Exports to U.S. and Trucks in October 2000 to June 2001 vs. October 2001 to June 2002 (Percentage Monthly and Cumulative Period Change). Sources: Statistics Canada, U.S. Customs, Ward’s Automotive and U.S. Labor Department.
border security without leading to major delays and uncertainty that could have the effect of reducing trade and transportation. And indeed, while trade and truck traffic grew substantially over the 90s, there has been a slowdown in Canadian exports to the U.S. since 9/11 and a slight reduction in truck traffic. Figure 1 shows U.S. economic activity, imports from Canada by land, and inward truck moves for the entire U.S.-Canada border for each of nine months October to June pre-9/11 compared to the same nine months post 9/11. On a cumulative level, while the U.S. industrial production index was down an average of some 3.66% over the nine month comparison period, and auto production was actually up 4.24% in the U.S., imports of goods by land from Canada fell 10.8%. This fall-off in Canadian exports to the U.S. by land will be of considerable concern in Canada where a number of trade associations, such as the Canadian Association of Manufacturers and Exporters, have expressed fears that post-9/11 perceptions of border delays and uncertainty might have this effect (Dobson, 2002; Lawson, 2002; MacFarlane, 2001; Windsor Star, 2002). Given that economic activity in the U.S. over the nine month period was flat to up, and that the Canadian dollar value was relatively flat over this period, one would have expected
Costs of the U.S.-Canada Border
287
imports from Canada to have held steady. The fact that they fell 10.8% may in part be due to U.S. industrial buyer’s concerns about the nature of the border now and in the future. While actual border transit times are not dramatically longer than they were pre-9/11 there has been considerable publicity about border problems immediately after 9/11, and there is greater uncertainty today over the time it will take to cross the border. In addition, there has been considerable press about the need to “secure” the northern border, and this may have led to buyers having some qualms about using Canadian sources. Border crossing processes and procedures, and their costs, were coming under scrutiny pre-9/11 (Kenna, 2001; National Post, 2001; Trickey, 2001) and have received much more attention in recent months. However, what are the cost impacts of border transit times and uncertainty on trade and personal travel? The objective of this paper is to identify border transit times and uncertainty costs for manufacturers, carriers, and personal travelers. Once these costs are identified it will be possible to estimate the benefits that might accrue from different border management strategies such as the “external perimeter” one being discussed in Canada.
METHODOLOGY In order to estimate border transit time and uncertainty costs, funding was obtained from the U.S. Department of Transportation, the Michigan and New York Departments of Transportation, and the Canadian Embassy. In order to estimate the costs of border transit time and uncertainty to the economy a combination of secondary source reviews and site visit/personal interviews were used. The information gathered in these reviews was then used, along with a variety of assumptions, to identify cost categories and to make cost estimates for each category of potential costs. Examples of specific cost categories include primary inspection booth transit time, secondary inspection yard processing time, reduced cycles for carriers, lost productivity from reduced trade, higher inventory carrying costs, etc. The secondary source review included identification, analysis and categorization of cost impacts from a review of some 750 newspaper articles on border issues. These articles appeared in 25 Canadian and 18 U.S. newspapers that were reviewed. Other secondary sources included some 45 border management reports on the overall border crossing environment or on specific border crossings, and several previous studies on the extent of and costs of border transit times. Key reports that were used are summarized in the references section. Several key secondary sources of data on trade, traffic, tourism, and transit time were also
288
JOHN C. TAYLOR ET AL.
used. These key sources included Bureau of Transportation Statistics reports on U.S.-Canada trade by year and month, U.S. Customs Service reports on monthly and annual traffic flows into the U.S., Statistics Canada reports on vehicle traffic entering Canada by year and month, and Canada Customs archives on border transit times for commercial and passenger vehicles entering Canada and the U.S. This transit time archive was critical to determining border crossing times and associated costs, a key portion of total border related costs. In order to gain a better understanding of the nature of various border crossings and to better understand the extent of transit time and uncertainty, a series of site visits to key border crossings were conducted during the summer of 2002 in order to make observations and conduct interviews. The seven key border crossing frontiers at Champlain, NY-Lacolle, Ont. (and Vermont crossings); Niagara Falls, Ont.-Niagara Falls, NY (three crossings); Buffalo, NY-Fort Erie, Ont.; Windsor, Ont.-Detroit, MI (two crossings); Port Huron, MI-Sarnia, Ont.; Emerson, MTPembina, ND; and Douglas, BC-Blaine, WA (four crossings) were visited. Based on these site visits and other sources, a total of 173 personal and/or telephone interviews were conducted in order to assess the impact of border transit time and other border related costs. These interviews were conducted with manufacturer, carrier, broker, trade development, and trade association organizations.
LITERATURE REVIEW Several trade articles and/or reports have referred to the “costs of the border” in general but have not been specific about the types of costs being considered. These reports have typically been from the Canadian side of the border and have focused on costs to Canada alone on exports to the U.S. One such report is a pre-9/11 quote from the Canadian Manufacturers and Exporters Association (Trickey, 2001) which suggests that the costs of the border result in an average 6% increase in the cost of Canadian manufactured goods, with some industrial sectors facing additional costs of 13% for border crossing delays and regulations. Former Prime Minister Brian Mulroney also suggested pre-9/11 that the “cost of crossing the border is at least C$30 billion per year to businesses in both countries” (MacDonald, 2001). Another estimate of border crossing costs is included in a May, 2002 report to the Canadian Parliament’s Standing Committee on Foreign Affairs and International Trade. This report by Dr. Alfie Morgan for the Windsor Chamber of Commerce estimates that removing remaining tariffs, reducing the needs for inspection at the border, and reducing NAFTA paperwork would reduce costs by some 2–3% of NAFTA trade (Morgan, 2002). However, none of these sources have discussed specific types of costs or provided documentation on costs.
Costs of the U.S.-Canada Border
289
FREIGHT RATE IMPACT AND DETAILED COST ESTIMATION Prior to examining the actual cost impacts, the researchers sought macro indicators of these costs that might be apparent in the level of cross-border freight charges to users as compared to purely domestic U.S. rates. Following that analysis, actual cost categories were identified and costs were estimated. Cost estimates were made using minimum, most likely, and maximum ranges of assumptions about the nature of the costs. The following subsections first review the cross-border freight rate information, and then the detailed cost categories and actual estimates of costs that result from this research.
CROSS-BORDER FREIGHT COSTS AND SURCHARGES The researchers believed that any cross-border freight costs that were in addition to typical domestic costs might show up in cross-border freight rates or possible border surcharges, although the ability to pass on such costs could be affected by market forces in the cross-border transportation market. These cross-border rates could be compared to domestic ones to calculate a border crossing cost impact as reflected by freight prices to shippers using the system. The total dollar value of these cross-border freight charge penalties could then be compared to any analysis of actual identified cross-border costs. Given the possibility of higher cross-border rates, the researchers conducted several interviews with knowledgeable sources. Interviews did indeed result in information indicating that cross-border trucking freight rates are considerably higher than would be the case for similar domestic U.S. moves (Trade Association and Carrier Interviews, 2002). While there are several reasons for these higher rates, including historical practice and the overall supply demand relationship, interviewees suggested that one of the key reasons relates to border crossing transit times, uncertainty about border crossing times and costs, and the costs of border related administration and information systems support. How much higher are cross-border rates? The Freight Carriers Association of Canada suggested cross-border rates are 10–15% higher than comparable domestic rates. However several carriers indicated their cross-border rates are 20–35% higher than domestic U.S. rates, with the lowest suggested rate premium for cross-border freight being 10%. It is also important to note that several carriers charge a border crossing premium routinely, and/or for “wait time” at the border. For instance, LTL and Roadway subsidiary (Carrier and 3 PL Interviews, 2002) charges a $20 per consignment fee for all cross-border freight (Carrier and 3PL Interviews,
290
JOHN C. TAYLOR ET AL.
Table 2. Cross-Border Freight Rate/Cost Penalty (Billions of U.S.$). Assumption 2001 U.S.$ trade by truck Domestic freight cost as % of sales Freight cost at domestic rates % by which cross-border exceeds domestic rates Cross-border incremental freight cost
Minimum
Most Likely
Maximum
235 4.00% 9.40 10% 0.940
235 4.50% 10.58 15% 1.59
235 5.00% 11.75 20% 2.35
2002). Con-Way Transportation has an $8 surcharge (Schulz, 2002). These two carriers alone estimate border crossing administration costs of U.S.$25 million. Table 2 summarizes the calculation methodology, and the minimum, most likely, and maximum estimates of total cross-border freight cost penalty charges to Canadian and U.S. shippers and receivers. The calculations result in cost estimates ranging from U.S.$.94 billion to U.S.$2.35 billion with a most likely estimate of U.S.$1.59 billion. These estimates start with the value of cross-border trade moving by truck, and assume typical domestic freight rates as a percentage of these trade values equal to 4–5% depending on the scenario. These freight costs percentages of revenue are based on published data from Herbert W. Davis and Company (2002). Penalty costs of 10, 15, and 20% on top of these typical domestic freight estimates are then assumed for each of the respective scenarios. While the above information suggests there may be cross-border costs, the freight rate information does not suggest what the specific costs are, or the actual level of these costs faced by carriers. Nor does it address other border related costs that may be borne by manufacturers or others. The following section summarizes the researcher’s efforts to identify and quantify costs related to transit time and uncertainty at the border.
TRANSIT TIME AND UNCERTAINTY RELATED COSTS Table 3 summarizes the cost impact categories and actual cost estimates that were identified and developed during the literature review, site visits, interviews and latter analysis. The cost categories that were identified relate specifically to impacts resulting from transit times and uncertainty about transit times, and affect carriers, manufacturers and personal travelers. For carriers, the primary source of extended transit time relates to backups at primary inspection stations, and time spent at secondary inspection yards at the border. Costs for each of these categories are calculated
Costs of the U.S.-Canada Border
291
Table 3. Transit Time and Uncertainty Related Impact Summary (Millions of U.S.$). Type of Cost Transit time/uncertainty costs Carrier related Primary delays Secondary delays Excess plan time Reduced cycles/other Driver documentation/fax time Carrier subtotal Manufacturer related Manufacturer sourcing benefits Extra inventory carrying cost Manufacturer subtotal Personal traveler Transit/uncertainty cost total
Minimum
Most Likely
Maximum
275.3 602.5 113.7 65.8 133.5
324.2 755.4 416.4 120.7 250.7
351.8 908.3 515.7 197.4 400.9
1190.8
1867.4
2374.1
1007.0 229.0
1530.0 458.0
2000.0 686.0
1236.0
1988.0
2686.0
96.7
159.0
209.6
2523.5
4014.4
5269.7
and summarized in Table 3. These transit times, and uncertainty about the extent of transit time results in carriers building excess border crossing time into their route planning, and this excess time is often lost time that cannot be productively used. As such the excess time over and above the actual transit time becomes a cost impact that can be estimated. At the same time, when not enough time is assumed for the border crossing, deliveries are late, and exchanges at terminals may be missed. Truckers also experience a number of costs related to the reduced number of cycles they can make in a given day, including the need for additional equipment and drivers to accomplish a set number of deliveries. The costs for these reduced cycles and late exchanges, etc. are summarized in a category called “reduced cycles and other costs.” Carrier drivers also spend considerable time preparing border crossing documentation and faxing documents ahead to brokers and these costs are estimated. Finally it should also be noted that personal travelers experience extended transit time as well, and these costs are also estimated later in this section. Manufacturers themselves also incur costs related to the border, over and above those specifically affecting truck transportation. The actual level of transit time and uncertainty about border crossing times, and more importantly, perceptions of such times by buyers, can have a significant impact on the level of cross-border sourcing, and the level of inventories that supply chain managers feel are necessary
292
JOHN C. TAYLOR ET AL.
to support their operations. Reduced cross-border sourcing that is due to border concerns, leads to foregone productivity benefits that would have otherwise been obtained and imposes a cost on the economy that are estimated here. The cost impact of additional inventory investment is also estimated. Table 3 summarizes estimates for each of the above identified cost categories and estimates that total transit time and uncertainty related costs ranged from U.S.$2.52 to U.S.$5.27 billion with a mid-range estimate of U.S.$4.01 billion. These costs are for the combined economies of Canada and the U.S. and are for a calendar year. For carriers the total mid-range cost impact is estimated at U.S.$1.87 billion using the hourly truck cost of U.S.$150. This hourly figure is for the costs of unplanned transit time, and is based on a report by two Federal Highway Administration (FHWA) officials that estimated the costs of planned and unplanned transit time to carriers at U.S.$144–371 per hour (ICF Consulting, 2002; Maring & Lambert, 2002). Using an hourly cost figure at the low end of this range, at U.S.$150 per hour, the primary inspection booth transit time backup costs were estimated to total U.S.$324.2 million in both directions of travel at the most likely scenario. Table 4 summarizes both primary inspection and secondary processing border costs for trucks. Secondary processing times will be addressed following a discussion of the methodology for estimating primary transit time. Primary transit time backup costs were estimated by using a comprehensive sample of May 1 to August 30, 2002 daily transit times maintained by Canada Customs (Canada Customs Border Transit Time Archive, 2002). This sample Table 4. Truck Primary and Secondary Delay Cost Impacts (Millions of U.S.$). Category/Assumption Entering USA Primary Secondary Total Entering Canada Primary Secondary Total Total U.S.-Canada Primary Secondary Total
Minimum
Most Likely
Maximum
213.4 392.5
256.1 483.4
277.5 574.4
605.9
739.5
851.9
61.9 210.0
68.1 272.0
74.3 333.9
271.9
340.1
408.2
275.3 602.5
324.2 755.4
351.8 908.3
877.8
1079.6
1260.1
Costs of the U.S.-Canada Border
293
has not previously been available to researchers. The transit time estimates are based on Canada Customs inspector reports at some 20 key crossings for each direction of travel for both cars and trucks. Reports to the central administrator are made daily every three hours. The average transit times for each of the key crossings in the study were then used as a representative sample of average transit time and multiplied by the post 9/11 annualized traffic at that crossing to arrive at total transit hours for that crossing for a one year time period. This was repeated for cars and trucks in each direction of travel for each key crossing and for the total of all other crossings in order to develop a border-wide estimate of total transit hours related to primary inspection. The border-wide primary transit time for trucks averaged just 9–10 minutes for entry to the U.S., but more importantly ranged from 10 to 180+ minutes on numerous occasions in the summer of 2002. At the Ambassador Bridge, the busiest commercial traffic entry point in the U.S., commercial primary transit times for entering the U.S. averaged 28 minutes in the summer of 2002, with a great deal of variability. Times on many days reached 1–2 hours for entry to the U.S. at Detroit. These primary inspection booth transit times cause major backups on city streets in Windsor and have caused an outcry in Windsor, with federal and provincial political leaders making almost weekly pronouncements on efforts to reduce the impact, even though the cause of backups is primarily related to the lack of U.S. Customs staffing for inspection booths. Time in Customs secondary inspection yards for completion of paperwork and occasional inspections represents a second major category of carrier border cost. While 10–40% of all trucks must enter secondary to visit brokers and/or to clear paperwork with Customs staff, just some 1% of vehicles are actually physically inspected with some of their contents removed. The costs of these secondary processing times on carriers is estimated at U.S.$755.4 million at the most likely assumption scenario. These estimates are based on dozens of interviews with Customs agencies and carrier management and drivers, and observation at key crossings (Carrier Interviews, 2002). In order to calculate this cost the percent of trucks that enter secondary annually at each crossing was estimated and multiplied times the annual truck volume at that crossing to determine the number of trucks that enter secondary in a year. For each crossing a minimum, most likely and maximum scenario of the number of minutes spent in secondary was then estimated based on the referenced interviews. These estimates ranged from 45 to 105 minutes per truck. While these times represent the average period in secondary, it is important to note that LTL’s with multiple consignments almost all go to secondary, and that 10–20% of the time they may be in secondary for anywhere from 2 to 10 hours. This variability leads to a great deal of uncertainty.
294
JOHN C. TAYLOR ET AL.
A third carrier cost relates to route planning times that are typically assumed at 2 hours (Carrier Interviews, 2002), even though the actual combined primary/secondary crossing time may be less. Excess planned time is in effect “lost” in many cases, thereby imposing an additional cost on carriers that end up needing more routes, drivers and equipment than would have been the case with certainty in the system. Because a portion of this two hour assumed crossing time cost has already been accounted for in the actual primary and secondary transit time costs discussed above, only the net “excess” time cost is reported here. These unused plan time costs were estimated at U.S.$416.4 million at the most likely level. Costs were calculated by taking the planned time hourly cost assuming the two hours per crossing and hourly cost, and subtracting the actual calculated primary and secondary transit time cost estimate for each of the three scenarios. For reduced cycles and related costs, the calculation assumed that between 7 and 12% of truck freight billings at domestic like rates incurred these kinds of costs. The most likely estimate for these costs was U.S.$120.7 million. Reduced cycle costs are due to the time necessary to cross the border and typically are due to the number of cycles per day being reduced from three to two. Other costs relate to driver-out-of-service regulation safety costs, missing exchanges, warehousing costs for missed deliveries, etc. The calculation assumed that a freight cost penalty of 10–14% would be incurred for these types of costs. Driver time involved with completing and sometimes faxing border paperwork is the last cost category for carriers. This cost was estimated at U.S.$250.7 million. The documentation relates to preparation of manifests and other customs paperwork, and the need to fax paperwork ahead in order to avoid the need to enter secondary yards. A far greater percentage of shippers and carriers are trying to use customs systems such as PARS and PAPS that allow for clearance at primary given the uncertainty over time it will take to get into and out of secondary customs and broker offices. Depending again on the scenario, we assumed between 20 and 30% of truck movements incurred these kinds of expenses. We further assumed a time commitment of between 20 and 40 minutes depending on the scenario. Manufacturers incur several additional costs over and above trucking carrier costs as a result of uncertainty in the border crossing process. Manufacturer cost impacts that result from uncertainty in the border crossing time were estimated to range from U.S.$1.24 to U.S.$2.69 billion with a most likely estimate of U.S.$1.99 billion. Two cost impacts were estimated – for lost productivity and for higher inventory carrying costs. The first of these impacts is for manufacturers who suffer reductions in productivity because of reduced sourcing from Canada. There is no impact in the other direction because Canadian imports from the U.S. held constant. Lost productivity benefits are estimated at U.S.$1.53 billion at the most
Costs of the U.S.-Canada Border
295
likely level. The calculation is based first on the annualized declines in Canadian sourcing observed in the nine months following 9/11 compared to the same period pre-9/11, less the reduction that would be expected as a result of lower U.S industrial production. These figures, as shown in Fig. 1, totaled to an annualized 10.8% reduction in imports to the U.S. from Canada, and a 3.66% reduction that one might have expected as a result of reduced industrial production in the U.S. across the time periods. This net reduction in imports not explained by industrial production changes totaled 7.14%, or U.S.$15.33 billion. The U.S.$15.33 billion reduction in imports was then multiplied by a lost productivity factor of between 7 and 13%, and a most likely 10%, depending on the scenario, to arrive at the overall loss in productivity benefits resulting from less Canadian sourcing. Manufacturers also incur higher inventory levels, and therefore higher inventory carrying cost expenses on their P&L, as a result of border uncertainty. This second category of expense was estimated at U.S.$458.0 million. This calculation assumes an extra 3, 4 or 6 days worth of sales in extra inventory, depending on the scenario. An 18% inventory carrying cost (ICC) percentage is then assumed to obtain the dollar expense impact, as opposed to balance sheet inventory impact. The extra days supply is based on many interviewees (Carrier and 3PL Interviews, 2002; Manufacturer Interviews, 2002) suggesting that small increments of additional inventory have been added to the supply chain. In addition, several articles that were reviewed suggested that additional inventory would be a prudent step for manufacturers to take to protect their cross-border supply chains from disruption due to crossing time variability (Cooke, 2002; Strong, 2001, 2002).
CONCLUSIONS The key contribution of this research is the development of cost estimates for the impact of border transit times and uncertainty on the two economies. These cost estimates are crucial for any effective cost-benefits estimate of the current border management system, and are critical to any future assessment of the benefits that would accrue from changes in the current border management system. This research suggests that the transit time and uncertainty costs of the border range from a minimum of U.S.$2.52 billion to a maximum of U.S.$5.27 billion, with a most likely estimate of U.S.$4.01 billion. The most likely cost estimate of U.S.$4.01 billion represents some 1.05% of total U.S.-Canada trade in 2001, and after adjusting out certain non-truck related costs, some 1.58% of all truck-borne trade. For carriers alone, not counting costs incurred by manufacturers, the costs total U.S.$1.87 billion for the most likely set of assumptions. This cost estimate compares to the estimated U.S.$1.59 billion in additional carrier freight charges
296
JOHN C. TAYLOR ET AL.
to users that was estimated earlier. The two numbers are fairly similar and we believe this offers a degree of validation concerning the cost estimates. Overall, the total cost estimate of U.S.$4.01 billion for carrier and manufacturer costs seems modest compared to some of the statements reviewed earlier. Various quotes in the trade press had estimated costs in the range of C$30 billion, to 13% of the cost of goods moved across the border, to a surtax on Canadian consumers equal to 5–7% of the value of goods. The cost estimate calculated here does seem to be more similar to the estimate put forth by Dr. Elfie Morgan for the Windsor Chamber of Commerce. As summarized in the literature review section, Dr. Morgan had estimated a cost impact of 2–3% of the value of NAFTA trade, including savings from elimination of duties not considered in this analysis. While the above cost estimates relate to transit time and uncertainty it should be noted that additional bilateral trade policy costs also exist. These costs relate to customs administration, duties and brokerage fees of the type included in Dr. Morgan’s analysis. Additional research needs to address these costs. Further research should also be aimed at attempting to validate the transit time and uncertainty-related costs analyzed here, and estimation of the input-output impacts on the economy that result from these direct costs of the border. Any such analysis will however have to address whether or not these costs result in any lost output, or simply a transfer of the output location, in the form of less cross-border sourcing, as was assumed here. Policy makers need to assess the costs of the border in making decisions about the future course of bilateral border management. On the one hand, attempts can be made to tighten the border, with likely additional costs in terms of both investment and border staffing, and additional costs for business. Whether or not the border can be tightened is an open question, given the level of trade and travel across the border on a daily basis (Handelman, 2001; Hart & Dymond, 2002; Trickey, 2001). The other option is to consider a move to a more open “external perimeter” oriented border. Such an approach would gradually reduce the role of the U.S.-Canada border in favor of maintaining border controls at the first point of entry to the U.S. and Canada. The external perimeter strategy has received considerable attention in Canadian policy circles (Conference Board of Canada, 2001; Dobson, 2002; Hart & Dymond, 2001; Windsor Star, 2002). One benefit of such a system would be the likely elimination of the U.S.$4.01 billion in border transit time and uncertainty costs estimated here. Some have even suggested that such a strategy could actually increase the level of security by reducing the focus on routine 30–75 second stops at the current border, and replacing this with more focused inspections at the first point of entry (Handelman, 2001; Hart & Dymond, 2001; Trickey, 2001). Whichever approach is taken, policy makers will need solid estimates of the cost to industry of the present system.
Costs of the U.S.-Canada Border
297
REFERENCES Canada Customs (2001–2002, November–August). Border delay archive database for commercial and personal travel north and southbound. Carrier and 3PL Management and Driver Interviews (2002, Summer). Conference Board of Canada (2001, October). Border choices: Balancing the need for security and trade. Cooke, J. (2002, January 1). Brave new world. Logistics Management. Crossing Operator Interviews (2002, Summer). Davis, H. W., & Company (2002, June). Davis database, 27(1). Dobson (2002, April). Shaping the future of the North American economic space (pp. 1–30). C. D. Howe Institute. Handelman, S. (2001, July 23). 49th parallel. Time Canada. Hart, M., & Dymond, W. (2001, November 20). Common borders, shared destinies: Canada, the U.S. and deepening integration. ICF Consulting (2002, January 16). Economic effects of transportation: The freight story. Kenna, K. (2001, September 4). Think tank urges borderless NAFTA. Toronto Star. Lawson, B. (2002, March 25). Extra security should speed U.S. border crossing. The Hamilton Spectator, A5. MacFarlane, S. (2001, October 8). Border delays won’t jeopardize JIT but tighter security essential to reassure the U.S. Canadian Business and Current Affairs. MacDonald, D. (2001, June 19). Consider customs union, Mulroney urges. . .. The Vancouver Sun, D3. Maring, G., & Lambert, B. (2002, October 3). The freight analysis framework: Status and future directions. Federal Highway Administration, Office of Freight Management and Operations, 35. Morgan, A. (2002, May 9). Presentation on the Canada-U.S. economic relationship, to the Canadian House of Commons, standing committee on foreign affairs and international trade, on behalf of the Windsor chamber of commerce. National Post (2001, September 3). Open borders editorial. Schulz, J. D. (2002, November 18). Con-way takes $8 security surcharge. Traffic World, 45. Strong, M. (2001, November 12). Just-in-time could be just-IN-case. Crain’s Detroit Business. Strong, M. (2002, February 11). Keeping the chain unbroken. Crain’s Detroit Business. Trickey, M. (2001, August 4). The undefended border: Economic concerns spur campaign to erase line between us and the U.S. The Montreal Gazette, A12. Windsor Star (2002, December 11). Perimeter border for North America editorial.
A CONTINENTAL ELECTRICITY INDUSTRY Paul S. Ciccantell ABSTRACT Creating a continental energy market, including an interconnected electricity industry, was a central motivation for the U.S. government in the negotiation of the CUSFTA and NAFTA. Free trade agreements and regulatory changes in North America have fundamentally altered the characteristics of the electricity industry and the strategies of its constituent firms over the past decade. Markets are replacing extensive regulation in many states, many new firms have entered the industry, long term stability and predictability of returns to firms and of electricity prices have been replaced with the uncertainties of competition, and blackouts in California have become global headline news. In this period of rapid transition in the electricity industry, firms, states and consumers confront both new opportunities and new problems that were unimaginable a decade ago. The essential role of electricity in all economic activity makes this industry a critical component of the North American economy, but the future of the industry is far from clear. This paper discusses the material characteristics of the electricity industry and outlines the provisions of the CUSFTA and NAFTA and regulatory changes that affected the electricity industry over the past decade. The paper then examines the evolution of the continental electricity industry, with particular emphasis on the efforts to create competitive markets. The paper then analyzes the strategies of particular firms to respond to and
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 299–315 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10016-8
299
300
PAUL S. CICCANTELL
take advantage of these processes. The conclusion analyzes the policy implications of these processes and firm strategies.
INTRODUCTION: THE CHANGING NATURE OF THE ELECTRICITY INDUSTRY Creating a continental energy market, including an interconnected electricity industry, was a central motivation for the U.S. government in the negotiation of the CUSFTA and NAFTA. Free trade agreements and regulatory changes in North America have fundamentally altered the characteristics of the electricity industry and the strategies of its constituent firms over the past decade. Markets are replacing extensive regulation in many states, many new firms have entered the industry, long term stability and predictability of returns to firms and of electricity prices have been replaced with the uncertainties of competition, and blackouts in California have become global headline news. In this period of rapid transition in the electricity industry, firms, states and consumers confront both new opportunities and new problems that were unimaginable a decade ago. The essential role of electricity in all economic activity makes this industry a critical component of the North American economy, but the future of the industry is far from clear. The massive economic and social costs of California’s electricity crisis and growing public opposition to the privatization of electric utilities in countries such as Brazil and Peru calls into question the future of what has become one of the leading sectors of the world economy of the past decade, the electricity industry. In the U.S., Western Europe, Latin America and many other areas of the world, this once staid, heavily regulated, and often government-owned public utility was transformed by deregulation, privatization, and the emergence of an increasingly global set of major firms into one of the most dynamic sectors of the world economy. Market capitalization of the twenty largest U.S. investor-owned electric utilities rose from U.S.$123.3 billion in 1991 to U.S.$218.7 billion in 2001, sales rose from U.S.$95.8 billion in 1991 to U.S.$378 billion in 2001, and profits rose from U.S.$9.8 to U.S.$16.6 billion (Business Week U.S. & Global 1,000 1992, 2002). Average market capitalization had almost doubled and average sales increased by a factor of four, but rates of profit measured as return on equity fell slightly (from 7.9 to 7.5%) and as a percentage of sales fell dramatically (from 10.2 to 4.4%). These increases in size derived in large part from restructuring, continentalization and globalization in the industry, and the decline in returns from the increased competition that these processes created. Investors, brokers, and stock market analysts hailed this transformation as creating tremendous new investment opportunities. Economists and government
A Continental Electricity Industry
301
officials argued that privatization and deregulation would attract badly needed new investment in generating and transmission infrastructure around the world, eliminate the inevitable inefficiencies of resource allocation and operation that resulted from heavy state involvement in the industry, and eventually lower costs and increase the availability of electricity to hundreds of millions of new consumers around the world (see, e.g. Brennan, Palmer & Martinez, 2002). In the wake of the collapse of Enron, one of the leaders of the transformation and globalization of the electric industry, the bankruptcy of Pacific Gas and Electric (one of the largest utilities in the U.S.), increasing worries about shortages of electricity in the U.S., and Congressional and state public inquiries into market manipulation and corporate accounting fraud, many analysts now argue that poorly designed markets and unethical corporate behavior are to blame for the industry’s problems. More effective regulation of electric markets and corporate behavior are intended to resolve these problems. In this paper, I argue that the restructuring, globalization, and continentalization of the electric industry and the current problems are actually the results of the normal functioning of the globalized world economy. In the globally competitive struggles for revenues, profits and investment capital, firms are always searching for profits above the average rate of profit by developing new products, patenting technological innovations, and opening new markets for their products and services. Efforts by the U.S. and state governments to guarantee long term security of electricity supply via deregulation and privatization combined with efforts by some firms in the electric industry to increase their profits, investment opportunities, and stock market value set in motion a gradual and uneven process of transformation in the U.S. The growing global pre-eminence of the neoliberal model of industrial and state restructuring dovetailed nicely with efforts in the U.S. and in many other nations to reduce the role of the state via deregulation and privatization of the electric industry, and the CUSFTA and NAFTA included provisions designed to create an integrated continental electric industry. The restructuring of the U.S. electricity industry and the closely linked effort to create an integrated continental energy industry led this broader global process in the electricity industry. The experiences of the U.S., Canada and Mexico represent the pioneering and most advanced efforts to transform electricity from a heavily regulated and often state-owned public utility into a highly competitive global industry. Analysis of the successes and problems in the evolving North American electricity industry provides insights both into this central dimension of the North American economy and into the future of many other nations around the world that are following this model of restructuring the electricity industry. This paper first discusses the material characteristics of the electricity industry. It then outline the provisions of the CUSFTA and NAFTA that affected the
302
PAUL S. CICCANTELL
electricity industry, as well as the key regulatory changes in the U.S., Canada and Mexico over the past decade. The paper then examines the evolution of the continental electricity industry, with particular emphasis on the efforts to create competitive markets. The fifth section analyzes the strategies of particular firms to respond to and take advantage of these processes. The conclusion analyzes the policy implications of these processes and firm strategies, especially in the wake of California’s energy crisis, the collapse of Enron, and concerns about the future adequacy of electricity supplies to power economic growth in North America.
THE MATERIAL CHARACTERISTICS OF ELECTRICITY The material characteristics of the electric industry present a series of opportunities for and constraints on the ability of states and firms to restructure the industry and to profit from this restructuring. The combination of natural, technical and economic characteristics of electricity generation and transmission have long shaped the operation and regulation of the industry, and this section will highlight a few of these characteristics that are particularly salient for the current period of restructuring, continentalization and globalization. First, there are many potential sources for generating electricity, ranging from renewable sources such as wind and water power to nonrenewable fossil fuels such as coal, natural gas and oil, to nuclear power. This great diversity of potential raw materials for electricity generation distinguishes the industry from almost all other raw materials-based industries, providing a wide variety of options for social use of the raw materials for generating electricity. Many forms of electricity can be generated almost anywhere that the raw materials are found or can be moved to at competitive cost (coal, natural gas, oil, nuclear), although some, most notably water and wind power, are relatively fixed by nature in particular locations, reducing their degree of social manipulability. Second, the spatial location of the industry is sharply constrained by the loss of electricity in transmission and the existence (or lack thereof) of regional and more broadly interconnected power grids. Electricity cannot easily be stored and must therefore be supplied continuously, and significant amounts of electricity are lost in transit; historically, electricity grids moved electricity relatively short distances, and even the highest capacity transmission lines typically move electricity several hundred miles over regional grids with limited interconnections between regions. Transmission lines are also expensive to build (both in terms of material and for land access) and to maintain, especially at the level of retail sales to individual small consumers. Transmission losses and the characteristics of power grids
A Continental Electricity Industry
303
effectively limit how far electricity can be shipped, creating important spatial constraints on generating, transmission and consumption patterns, with a strong pull for locating generating facilities relatively close to cities and factories that will consume the electricity. Concerns about pollution from fossil fuel burning and the potential for reduced transport costs for raw materials, especially coal, pull generating facilities toward rural raw materials-supplying regions, further complicating locational decisions for plant siting. Other forms of electricity, especially hydroelectricity but also geothermal and wind power, are only available where nature has created them, constraining the ability of firms to utilize these often low cost sources of electricity even more sharply and often drawing large electricity consumers to the area of the electricity resource. Third, these material and technical characteristics of generation create wide variation in the capital and operating costs of electricity generation. In terms of capital costs, investment for competitive facilities today range from a few million dollars (natural gas on an existing gas pipeline, wind farms) to tens of millions to a few hundred million dollars (oil, coal) to multi-billion dollars (large scale hydroelectricity, nuclear). However, in terms of operating costs, the continuum is generally exactly the opposite: hydroelectricity and nuclear power are usually by far the cheapest (a few mills per kilowatt-hour produced), to wind and then to gas, oil and coal, whose operating costs can vary tremendously depending on fossil fuel prices but are typically measured in tens of mills per kilowatt-hour. Fourth, technical and economic characteristics combine to affect how generating facilities using different types of raw materials are operated by firms. Gas and oil-fired plants are the easiest and cheapest to turn on and off for peak (high demand) production, while coal, nuclear, and most hydroelectric sources (especially those without the option of pumped storage of water) are more expensive to shut down at non-peak times. As a result, firms typically maintain both baseload facilities of coal, nuclear, and/or hydroelectric plants, and a supplementary amount of peak capacity, most commonly today fired by natural gas. Fifth, these natural, technical and economic characteristics combine to create very large economies of scale in generation and in transmission, as well as other limitations and inefficiencies on efforts to create truly competitive markets. The most notable of these problems is the cost and inefficiency of multiple power lines to supply each potential customer. Efforts to deregulate and restructure electricity have sought various solutions to these problems, including separating ownership or at least control over transmission from generation. The competitive advantage resulting from these economies of scale has historically been regulated as a “natural monopoly” by states around the world because of the fear of monopolistic pricing practices by firms that are able to keep competitors out of the market.
304
PAUL S. CICCANTELL
The restructuring, continentalization and globalization of the electric industry are profoundly constrained by the material characteristics of electricity generation, transmission and consumption, requiring firms to operate subsidiaries in many political jurisdictions with often very different legal, political and economic environments in order to expand their business opportunities. These characteristics provide the framework for firm and state strategies to transform this industry, providing opportunities for and constraining social manipulation.
THE CHANGING INSTITUTIONAL ENVIRONMENT: CUSFTA, NAFTA AND DEREGULATION AND RESTRUCTURING Recessions, inflation, high unemployment, and other negative economic indicators in the 1970s and early 1980s heralded for many analysts the collapse of the Fordist labor-capital accord (Bluestone & Harrison, 1982) and stagnant real incomes simply provided further evidence of the collapse of U.S. hegemony. The fall of the Berlin Wall, the breakup of the Soviet Union, and the growing economic integration of Western Europe were seen as signs of the development of a multipolar world. What is often termed a new strategy for corporate hegemony (Mander, 1993; Moody, 1995) was also being formulated and implemented during the 1980s. Although given a variety of labels, including neoliberalism, neoconservatism, the Washington Consensus, and market liberalization, among others, this new strategy was firmly rooted in U.S. economic and political conditions and interests. NAFTA from the U.S. perspective was an effort to reconstruct U.S. economic competitiveness and hegemony. The roles of energy and more broadly of raw materials industries in the CUSFTA and NAFTA negotiations and implementations illuminate the U.S. interests driving these broader processes of restructuring the North American political economy in the context of declining U.S. economic competitiveness (Ciccantell, 2001). The origins of the CUSFTA and NAFTA in the mid-1980s during the Reagan administration were in one sense a defensive maneuver to create a secure, continental energy market that would help the U.S. compete in the new multipolar world (Dillon, 1993, p. 315; Pastor, 1990, p. 20; Shrybman, 1993, p. 274). Energy trade with Canada had long been characterized by trade disputes; in periods of abundant energy supplies, U.S. energy producers had often sought to keep Canadian exports out of the U.S. market, but during periods of shortage Canada had threatened to and sometimes had cut off exports to the U.S. (Macdonald, 1987, p. 29). As Drache (1993) has argued, “U.S. negotiators aimed for and got ‘secure and enhanced access’ to Canada’s resource sector” (Drache, 1993, p. 81).
A Continental Electricity Industry
305
Since the implementation of the CUSFTA, Canadian energy development focused on megaprojects to serve U.S. markets (Shrybman, 1993, p. 275). In comparison to the raw materials supply system obsolescence in the U.S., Japan and, to a lesser extent, Europe, had developed new raw materials supply networks in the 1960s and 1970s that had sharply reduced their raw materials supply costs, in large part due to the use of large scale transport systems that could not be utilized by U.S. competitors because of the inadequate capacity of U.S. ports (Bunker & Ciccantell, 1995). Canada and Mexico, despite their historical roles as U.S. raw materials peripheries and their own developmental efforts to use raw materials wealth as the basis for domestic industrialization, still had very large and diverse energy and other raw materials reserves that could be harnessed to supply the U.S. economy (Ciccantell, 2001). Canada and Mexico had very different goals in negotiating free trade agreements with the U.S. For Canada, “over the years Canada has run a substantial merchandise trade surplus by selling resource or resource-derived products to foreign customers. . . . it has now become apparent that the rate of growth of the resource sector in Canada will not continue. If we were to be dependent on the resource sector alone, the rate of improvement in the Canadian standard of living would drastically slow down” (Macdonald, 1987, p. 11) unless Canada negotiated a free trade agreement with the U.S. that would provide an economic alternative. At a minimum, Canada sought to protect resource and manufactured exports from U.S. trade discrimination, a growing problem during the 1980s (Grinspun, 1993, p. 105). The CUSFTA and NAFTA have protected some, but not all, of Canada’s resource exports from unilateral U.S. action, but in so doing has consolidated Canada’s role as a raw materials periphery of the U.S. (Ciccantell, 2001). The U.S.-Canada Free Trade Agreement of 1988 demonstrated the U.S. interest in access to Canadian raw materials, particularly energy, and the broader implications of global economic restructuring under U.S. leadership. Tariffs on raw materials and processed raw materials were eliminated, sharply reduced, or entered a process of phased reductions (USBM, 1991, p. 133). The energy sector was a particularly contentious issue due to the history of conflicts over oil, natural gas, and electricity trade. The CUSFTA sought to eliminate these conflicts by guaranteeing market access (Macdonald, 1987, p. 29), but guaranteed U.S. access even during periods of global shortage faced severe criticism in Canada (Ciccantell, 2001). NAFTA also contained a number of special provisions regarding energy in Chapter 6 of the agreement, as well as guaranteeing more broadly national treatment and market access of electricity produced throughout the region and investor rights. The CUSFTA largely remains in force to govern Canada-U.S.
306
PAUL S. CICCANTELL
energy relations, with some notable expansions, including restricting the ability of states and provinces to disrupt contractual relations across borders under the terms of NAFTA. NAFTA also created opportunities for U.S. and Canadian firms to supply services and equipment to the Mexican oil and electric industries and to set up electricity generating facilities in Mexico as independent power producers that would then sell wholesale electricity to Mexico’s Federal Electricity Commission (CFE). A central U.S. interest in the NAFTA negotiations was access to Mexican energy resources. Although the U.S. sought to have Mexico accept the CUSFTA energy provisions, Mexico was able to partially protect its control over energy resources, although only at the cost of opening the Mexican market to U.S. firms as subcontractors and equipment suppliers and of the promised opening of part of the petrochemicals sector (Ciccantell, 2001; Dillon, 1993, pp. 326–327; Dukert, 1999; Horlick, Schuchhardt & Mann, 2001). The special provisions for energy included in the NAFTA were due to the raw materials peripheries’ recognition of the key potential roles of these industries as generative sectors. Canada and Mexico made significant efforts to protect their own development policies rather than being thoroughly reincorporated as U.S. raw materials peripheries, but these efforts were at best only partially successful. The central significance of access to energy resources on the part of the U.S. forced Canada and Mexico to accede at least partially to this U.S. goal, ensuring low cost availability of electricity, oil, and natural gas to fuel the U.S. economy (Ciccantell, 2001). A variety of efforts to deregulate and restructure the electricity industry have been undertaken at the national and state levels in the U.S. and, to a lesser extent, at the national and provincial levels in Canada. In the U.S., one critical early step was the phased deregulation of natural gas transmission, with the goal of increasing supplies and lowering the cost of natural gas. Opening access to capital-intensive pipeline networks essential to the movement of gas to customers to other firms and the creation of open markets for natural gas sales in place of heavily regulated transactions by decoupling the stages of production and transmission was mandated by the U.S. Federal Energy Regulation Commission (FERC) in 1993. This deregulation spurred a great deal of investment in exploration, production, and pipelines, creating an increasingly interconnected continental natural gas pipeline system and market that dramatically increased the longstanding connections between the U.S. and Canada and, in recent years, led to the construction of a growing set of connections with Mexico as well (Dukert, 1999; EIA, April 2001). Electricity deregulation and restructuring in the U.S. began with the Public Utility Regulatory Policies Act (PURPA) of 1978 that required utilities to purchase
A Continental Electricity Industry
307
power from firms operating renewable fuel sources or industrial cogeneration (selling excess electricity produced by a power plant primarily used to power a company’s own operations), creating the first attempt at “unbundling,” or separating generation from transmission. This opening to other firms via unbundling was expanded by the Energy Policy Act of 1992 and the FERC’s Orders 888 and 889 of 1996 to create competition in wholesale power generation and sales by opening transmission lines to all electricity generators and facilitating the efforts of states to deregulate and restructure their electricity industries. Almost half of all states have now changed laws or regulations regarding electricity to promote the restructuring of their electricity industries (EIA, February 2003). Merger and acquisition activity has reached record levels, with fifty mergers between investor-owned utilities approved by the FERC between 1995 and 2002. In addition, these regulatory changes also opened the industry to the entry of dozens of new firms as nonutility electricity generators, typically referred to as Independent Power Producers (IPPs). These IPPs increased their electricity generation from 62 billion kilowatt-hours in 1989 to 207 billion kilowatt-hours in 1996 and to 932 billion kilowatt-hours in 2001, increasing their share of total U.S. electricity generation from 2.1% in 1989 to 6% in 1996 and to 25.1% in 2001 (Brennan, Palmer & Martinez, 2002; EIA, September 2002, February 2003, 2002, pp. 221, 1993). One of the most difficult issues involved in changing the regulatory structure governing the electricity industry has been the problem of stranded costs: investments made by utilities based on the previous regulatory environment and projections of demand and capacity growth that were negotiated under earlier regulatory systems but that are uneconomic in the new competitive environment if customers change to other electricity suppliers. The EIA, in reviewing the wide range of studies on this issue, reports a range of stranded costs estimates nationally at anywhere from U.S.$10 billion to U.S.$500 billion. The federal legal and regulatory structure has assigned repayment of these costs to customers who change to new electricity suppliers (EIA, February 2002), but this issue remains a contentious battleground at the federal and state levels between utilities, IPPs, various types of customers, political parties, and various regulatory bodies because of the large burden these costs impose on firms and other groups. California and Rhode Island were the first two states to open their electricity industries to competition and consumer choice of electricity supplier, and several other large states, including Massachusetts, New Jersey, Illinois, Pennsylvania, New York, Virginia and Texas, have implemented large scale competition and restructuring programs more recently (EIA, February 2002). In many ways, California has served as both the pioneer and as the harbinger of the many difficulties involved in making this dramatic transition from extensive regulation
308
PAUL S. CICCANTELL
to a more competitive industry; I will return to the issue of the lessons of California’s experiences in the concluding section of the paper.
THE EVOLVING CONTINENTAL ELECTRICITY INDUSTRY The CUSFTA, NAFTA, and deregulation and restructuring in the U.S. have generated incentives and a new institutional environment for creating a truly continental electricity industry. For the U.S. and Canada, cross-border electric grids and supply of power from hydroelectric plants in Ontario and Quebec have long been an area of U.S.-Canadian interdependence (Bothwell, 1992, p. 114; Dukert, 1999). Canada exported 42.9 billion kilowatt-hours of electricity to the U.S. in 1999 (EIA, April 2001). Leading Canadian energy companies, including Transcanada and Enbridge, have implemented continental strategies in electricity and/or natural gas transmission and generation, taking advantage of the new institutional environment. The massive hydroelectric potential in northern Quebec that has been developed in recent years may also prove to be a low cost complement to electricity generation and consumption in the eastern U.S. Although a major contract to supply electricity from these new hydroelectric utilities to New York state was canceled due to concern over the large scale environmental and social disruptions and dislocations that have resulted from these dams (Gitlitz, 1992, p. 1), over the long term the availability of this low cost electricity may be irresistible to U.S. electric utilities, potentially strengthening the electricity link between the U.S. and Canada. Natural gas pipeline connections from Canada to the U.S. to supply electricity generation have expanded dramatically during the fifteen years since the CUSFTA was signed, creating an integrated raw materials supply network for electricity generation, and a variety of other proposals, including linking the most remote regions of northern Canada (Krauss, 2002) and building power lines underwater across Lake Erie (AP, 2002) are likely to be built at some point in the future. Despite this long term interconnection of electricity grids between the northeastern U.S. and eastern Canada in the east and between the western states of the U.S. and the western Canadian provinces and a variety of efforts to create a continental electricity grid, the U.S. is still effectively separated into three poorly linked regional grids: the eastern half of the U.S. and eastern Canada, the western half of the U.S. and western Canada, and Texas (Dukert, 1999; EIA, February 2002). This legacy of the technical difficulties of transporting electricity and the fragmented and largely state-based system of electricity regulation in the U.S. remains a significant obstacle to complete continental integration, effective competition, and security and reliability of electricity supplies.
A Continental Electricity Industry
309
The restructuring of the U.S. electric industry and NAFTA’s steps toward opening the electricity generating and sales industries in Mexico and Canada are leading to further integration of the three nations’ electricity supply networks, particularly between the U.S. and Mexico. Integration before NAFTA between the U.S. and Mexico was limited to the Falcon Dam cooperative project (Dukert, 1999, p. 25), but the interconnections of U.S. and Mexican electricity and natural gas networks are expanding rapidly. Many U.S. and Canadian utilities are exploring or have established operations or supply agreements in Mexico (see, e.g. Dowling, 2002a,c; Monaghan, 2002; Weiner, 2002), as are foreign electricity firms, including EDF of France and Fenosa of Spain (Dowling, 2002b). These plans include a variety of types of generating plants, including importing liquefied natural gas from Asia, to supply electricity in Mexico and to the U.S. (Dukert, 1999; Royce, 2002b), bringing the continent closer to a fully integrated electricity supply system. The following section will highlight a few of the key firm strategies in the new era of restructuring, continentalization and globalization of electricity.
ELECTRICITY FIRM STRATEGIES IN NORTH AMERICA The extremely warm reception the processes of restructuring and globalization of the electricity industry received in global capital markets, especially in those based in the U.S., is quite understandable. An industry with heavily regulated, low rates of profit that provided an almost irreplaceable product for billions of people was becoming a vast new investment and profit opportunity, and many firms and investors rushed in to take advantage of this opportunity. Mimetic isomorphism (imitating the strategies and structures of the organizations in a field that are perceived as successful, whether or not they are “successful” in any measurable sense), led governments and firms around the world to follow the lead of the U.S. and California governments and of firms like Enron in their efforts to restructure the industry and benefit from this transformation. In a situation of great uncertainty created by this increasingly rapid transformation and globalization, mimetic isomorphism was a rational, relatively safe choice for firms and states. In the absence of complete information, changing and often unclear government rules, and the potential for huge increases in profit and stock market values, imitating the leading actors in the process made good political and economic sense. Existing electric utilities and other firms seeking to enter the newly competitive North American electricity industry as IPPs have pursued a variety of strategies in this new institutional environment. Many existing utilities have effectively
310
PAUL S. CICCANTELL
demerged their operations, separating into companies providing transmission and/or retail sales on the one hand and into newly formed IPPs based on the transfer of their generating assets on the other. Transfers of generating assets via such corporate divisions and sales to other IPPs between 1997 and April 2000 included 51 utilities that sold or were in the process of selling a total of 156,500 megawatts of generating capacity, with 38,300 megawatts transferred to a nonutility subsidiary and the rest sold or offered for sale to IPPs (EIA, 2000, p. 105). Merger and acquisition activity have become a critical corporate strategy in the electricity industry as well. Between 1992 and April 2000, 35 mergers or acquisitions were completed between investor-owned utilities or between utilities and IPPs, and twelve more were announced and pending approval in mid-2001, and an additional 23 mergers between utilities and natural gas companies (termed “convergence mergers”) took place or were pending between 1997 and April 2000 (EIA, 2000, pp. 91, 98). Many foreign electricity firms, including Scottish Power, Powergen, British Energy, and International Power of the United Kingdom have become major players in the U.S. electricity industry via mergers and acquisitions (EIA, August 2002). The seventy mergers in eight years must be viewed in the historical context of the regulated electricity and natural gas industries, a context that made mergers and acquisitions relatively uncommon until the deregulation and continentalization era of the past decade. At least two multibillion dollar mergers between electricity firms collapsed during the last three years (EIA, 2001), however, because of the problems of Enron and concerns over the future and profitability of restructuring. On a broader geographic scale, these processes of restructuring and continentalization are being replicated globally, as many nations open at least part of their electricity industries to foreign investment. For the largest North American utilities, globalization has been a critical complement to their efforts to generate new revenues and sales in North America. Of the largest electricity firms in the U.S. and Canada, most now have a significant international presence in Latin America, Europe, and/or Asia, including Enbridge of Canada, as well as many U.S. firms, including Duke Energy, AEP, El Paso, TXU, and PSEG, as well as the remnants of Enron. In the natural gas industry, emerging as the fastest growing source of fuel for electricity generation, deregulation and the broader process of electricity restructuring have fueled a dramatic increase in mergers, acquisitions, and investments in new facilities. Between 1990 and 1997, for example, mergers and acquisitions in the U.S. natural gas industry increased from U.S.$10.4 billion to U.S.$39 billion, and in 2001, El Paso Energy (a major natural gas and IPP firm) completed a U.S.$24 billion merger with Coastal (EIA, April 2001). These activities and relatively high and volatile gas prices (in contrast to its history as
A Continental Electricity Industry
311
a relatively stable, highly regulated industry) have, however, opened the door to increased competition from coal-fired and other types of new generating plants in the new competitive environment (EIA, April 2001). Other firms in the electricity industry have followed other strategies in the current era, including some utilities that have lobbied successfully to maintain the current regulatory structure and as a result their existing integrated business strategies, a growing number of small firms that invest in limited numbers of generating plants to serve particular deregulated markets, firms or, more often, subsidiaries of larger firms that specialize in trading of electricity, and some firms that have sought to become retail marketers of electricity without owning generating facilities.
CONCLUSION: IMPLICATIONS FOR ANALYSIS AND PUBLIC POLICY The current problems in the North American and global electric industries result directly from these firm and state strategies to restructure, continentalize and globalize the electricity industry. It is important first to put the conflicts over electricity costs and shortages in California, the collapse of Enron, and the growing public questioning of the benefits of restructuring in the U.S. and globally in historical context. Historically, when an entire new field of investment and profit is opened and firms and states seek to benefit from the potential for high rates of profit, innovation, and other hoped-for benefits, the ensuing boom is often followed by a dramatic crisis. One prominent example is the rapid expansion of the U.S. railroad industry during the second half of the nineteenth century. This new technology was profoundly rooted in space and the potential for ecosystems and social organizations in the regions it linked for producing large volumes of raw materials and agricultural products that could repay the huge investments required in infrastructure. Rail lines to major coal and copper deposits and areas with high grain and cattle production potential, for example, earned substantial profits, but lines in less favorably endowed regions often went bankrupt because their large investments were physically sunk and inflexible (Barham & Coomes, 1994) and could not be readily deployed to potentially more profitable areas, dragging many railroad firms into bankruptcy during the late 1800s. This new railroad technology, supported by large state subsidies and a flood of U.S. and foreign capital into the industry, earned large profits and saw its market value grow in a poorly regulated and understood business and legal environment and then created an economic and stock market crisis in the 1890s. The collapse of Enron and the broader crisis in the global electric industry are the products of a similar constellation of material and social characteristics. Electricity generation
312
PAUL S. CICCANTELL
and transmission are constrained by natural, technical and material characteristics that are not easily susceptible to manipulation by states and firms. At the global level, Enron’s global operations, while usually lacking the almost unbelievable financial manipulation of its trading schemes in the U.S., are being sold, including U.S.$3 billion worth of assets in Brazil alone, and other firms are trying to sell or even simply abandoning investments initially worth hundreds of millions or even billions of dollars; PSEG, for example, announced this week that it would simply abandon a U.S.$212 million investment in Argentina (Dowling, 2003; Royce, 2002a; Smith, 2002). Much of the estimated U.S.$55 billion spent by energy utilities during the first half of 2002 (Taylor, 2002) now looks in danger of becoming competitive disadvantages, rather than increases in scale and scope that provide the future of many of the world’s leading electricity firms. The severe problems faced by California, the efforts to reassert state control and funding, and the ongoing battles between the state and federal governments over the causes and the distribution of the costs of California’s pioneering experience are generating a stream of research (see, e.g. Brennan, Palmer & Martinez, 2002; EIA, July 2002; Fellmeth, 2002; Jurewitz, 2002; Munroe & Baroody, 2002; Tomain, 2002; Wasserman, 2001) and almost countless newspaper and magazine articles, many of which focus on the negative consequences of the dramatic transformation of the electricity industry in recent years. One industry consultant, for example, has concluded that “the economics of the retail electricity market limit profitability as much as regulation does. . . . So far, in states with deregulated markets, no company has achieved sustainable gross margins. Neither has any company demonstrated the ability to leverage electric commodity sales into high-margin value-added services” (Humphrey, 2002, p. 54). The problems of California and provincial problems in Ontario’s restructuring program have led effectively to its halt by the provincial government as well (Simon, 2002), and growing protests over the impacts of this process in Brazil, Peru and many other nations threaten to curtail the global process as well. These efforts to curtail or even reverse these processes of deregulation, continentalization and globalization in the electricity industry do not, however, recognize the fundamental problems that motivated these processes, particularly on the part of many governments. Most notably, electricity generation and transmission are both capital-intensive and essential to all other forms of economic growth, and the burden of funding this essential pillar of economic growth is difficult to manage and sustain in heavily regulated environments. Firms, banks and investors can find many other, potentially more profitable, and at least potentially less risky investments for their capital than in a heavily regulated industry open to various forms of political intervention for both public and private purposes. Particularly in poorer nations without large capital markets, these problems are
A Continental Electricity Industry
313
critical, but even in North America, this challenge is significant, as government agencies and some analysts argued during this process of transforming the electricity industry. Moreover, creating a new institutional environment is clearly quite difficult. It is not simply a matter of ending regulation, but instead of creating a new, stable framework, resolving problems from making the transition from the old framework (such as stranded costs), and developing the oversight mechanisms necessary to thwart the efforts of firms such as Enron that recognize the huge profits to be made from exploiting the uncertainties and gaps in the new system. Given the difficulties of many firms and states in the developing new institutional framework, we clearly do not know the best ways to transform this industry yet, but we clearly need to learn from the experiences of California, the broader U.S. and continental electricity industries, and of the firms and states globally that are now deeply involved in trying to create the new institutional frameworks and firm strategies needed for this very different industry.
REFERENCES Associated Press (AP) (2002). Power companies evaluate line to Canada. New York Times, October 20. Barham, B., & Coomes, O. (1994). Reinterpreting the Amazon rubber boom: Investment, the state, and Dutch disease. Latin American Research Review, 29(2), 73–109. Bluestone, B., & Harrison, B. (1982). The deindustrialization of America. New York: Basic Books. Bothwell, R. (1992). Canada and the U.S.: The politics of partnership. New York: Twayne Publishers. Brennan, T., Palmer, K., & Martinez, S. (2002). Implementing electricity restructuring: Policies, potholes, and prospects. Environmental and Resource Economics, 22, 99–132. Bunker, S., & Ciccantell, P. (1995). Restructuring space, time and competitive advantage in the world-system: Japan and raw materials transport after World War II. In: D. Smith & J. Borocz (Eds), A New World Order? Global Transformation in the Late Twentieth Century (pp. 109–129). New York: Greenwood Press. Ciccantell, P. (2001). NAFTA and the reconstruction of U.S. hegemony: The raw materials foundations of economic competitiveness. Canadian Journal of Sociology, 26(1), 57–87. Dillon, J. (1993). The petroleum sector under continental integration. In: R. Grinspun & M. Cameron (Eds), The Political Economy of North American Free Trade (pp. 315–330). New York: St. Martin’s Press. Dowling, J. (2002a). Transalta completes campeche financing-Mexico. BNAmericas.com, October 17. Dowling, J. (2002b). Fenosa seeks strategic partner for 3 power plans-Mexico. BNAmericas.com, October 8. Dowling, J. (2002c). Kinder Morgan to start U.S.$87 million Monterrey pipeline ops 2003-Mexico. BNAmericas.com, September 25. Dowling, J. (2003). PSEG abandons Edeersa due to economic crisis-Argentina. BNAmericas.com, April 3. Drache, D. (1993). Assessing the benefits of free trade. In: R. Grinspun & M. Cameron (Eds), The Political Economy of North American Free Trade (pp. 73–88). New York: St. Martin’s Press.
314
PAUL S. CICCANTELL
Dukert, J. (1999). The evolution of the North American energy market. Center for Strategic and International Studies Policy Papers on the Americas (Vol. X) Study 6. Energy Information Administration (EIA) (1993). The changing structure of the electric power industry, 1970–1991. Washington, DC: U.S. Department of Energy. Energy Information Administration (EIA) (2000, October). The changing structure of the electric power industry 2000: An update. Washington, DC: U.S. Department of Energy. Energy Information Administration (EIA) (2001, April). U.S.A. http://www.eia.doe.gov/cabs/usa.html. Energy Information Administration (EIA) (2002). Annual energy review 2001. Washington, DC: U.S. Department of Energy. Energy Information Administration (EIA) (2002, February). The restructuring of the electric power industry: A capsule of issues. http://www.eia.doe.gov/cneaf/electricity/chg str/booklet. Energy Information Administration (EIA) (2002, July). Status of the California electricity situation. http://www.eia.gov/cneaf/electricity/california/. Energy Information Administration (EIA) (2002, August). Foreign direct investment in U.S. energy in 2000. http://www.eia.doe.gov/emeu/finance/fdi/2000/index.html. Energy Information Administration (EIA) (2002, September). Electric power industry restructuring fact sheet. http://www.eia.doe.gov/cneaf/electricity/page/fact sheets/restructuring.html. Energy Information Administration (EIA) (2003, February). Status of state electric industry restructuring activity as of February 2003. http://www.eia.doe.gov/cneaf/electricity/chg str/tab5rev.html. Fellmeth, R. (2002). Plunging into darkness: Energy deregulation collides with scarcity. Loyola University Chicago Law Journal, 33, 823. Gitlitz, J. (1992). New York cancells $12.6 billion hydro-Quebec contract. World Rivers Review, 7(2), 1, 13. Grinspun, R. (1993). The economics of free trade in Canada. In: R. Grinspun & M. Cameron (Eds), The Political Economy of North American Free Trade (pp. 105–124). New York: St. Martin’s Press. Horlick, G., Schuchhardt, C., & Mann, H. (2001). NAFTA provisions and the electricity sector. Montreal: Commission for Environmental Cooperation. Humphrey, B. (2002). Retail competition: The toughest course. Electric Perspectives, 27(3), 54–66. Jurewitz, J. (2002). California’s electricity debacle: A guided tour. The Electricity Journal (May), 10–29. Krauss, C. (2002). A prospect delayed in Canada’s Arctic. New York Times, November 16. Macdonald, D. (1987). The politics and economics of bilateral free trade: Canadian perceptions. In: E. Fried, F. Stone & P. Trezise (Eds), Building a Canadian-American Free Trade Area (pp. 11–17). Washington: Brookings Institution. Mander, J. (1993). Megatechnology, trade, and the New World Order. In: R. Nader et al. (Eds), The Case Against “Free Trade”: GATT, NAFTA, and the Globalization of Corporate Power (pp. 13–22). San Francisco: Earth Island Press. Monaghan, C. (2002). Marathon upsizes LNG plans to U.S.$1.5 billion-Mexico. BNAmericas.com, October 2. Moody, K. (1995). NAFTA and the corporate redesign of North America. Latin American Perspectives, 22(1), 95–116. Munroe, T., & Baroody, L. (2002). California’s flawed deregulation-implications for the state and the nation. Journal of Energy and Development, 26(2), 159–179. Pastor, R. (1990). Post-revolutionary Mexico: The salinas opening. Journal of Interamerican Studies and World Affairs, 32(3), 1–22. Royce, K. (2002a). NRG to decide on LatAm divestments end 2002. BNAmericas.com, September 3.
A Continental Electricity Industry
315
Royce, K. (2002b). EDF aims for Rio Bravo III, IV completion in 2004–2005-Mexico. BNAmericas.com, November 22. Shrybman, S. (1993). Trading away the environment. In: R. Grinspun & M. Cameron (Eds), The Political Economy of North American Free Trade (pp. 271–296). New York: St. Martin’s Press. Simon, B. (2002). Outcry in Ontario over energy prices. New York Times, November 13. Smith, T. (2002). 8 Utilities in Brazil could go on the block. New York Times, August 30. Taylor, A. (2002). Energy utilities go on $55 billion takeover spree. Financial Times, August 18. Tomain, J. (2002). The past and future of electricity regulation. Environmental Law, 32(369), 435–474. United States Bureau of Mines (later U.S. Geological Survey) (USBM/USGS) (Various Years). Minerals Yearbook. Washington: USBM/USGS. Wasserman, H. (2001). Power struggle: California’s engineered energy crisis and the potential of public power. Multinational Monitor (June), 9–20. Weiner, T. (2002). U.S. will get power, and pollution, from Mexico. New York Times, September 17.
PARTICIPATION BY CIVIL SOCIETY AND NON-GOVERNMENTAL ORGANIZATIONS IN WESTERN HEMISPHERE ECONOMIC INTEGRATION Jonathan P. Doh and Barbara Kotschwar ABSTRACT Civil society, as represented by non-governmental organizations (NGOs), is exerting increasing pressure on national governments, multinational corporations, and international institutions. In this chapter we document the evolution of participation by civil society and NGOs in Western Hemisphere economic integration, focusing particularly on the NGO role in three important trade and investment agreements: the U.S.-Canada Free Trade Agreement, the North American Free Trade Agreement, and the Free Trade Area of the Americas process. We find that NGOs are having increasing influence on the trade and investment agreements in the Hemisphere, and are poised to take on a major role in multilateral negotiations and agreements.
North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 317–337 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10017-X
317
318
JONATHAN P. DOH AND BARBARA KOTSCHWAR
INTRODUCTION Globalization and economic integration have generated opportunities for economic growth and development in modern economies, but have also drawn criticism from groups concerned about potential negative externalities. As countries have codified economic integration through trade and investment agreements, citizen concerns have emerged about the overall impact and distribution of benefits from such arrangements, and demand has grown to request that civil society – often represented by non-governmental organizations (NGOs) – have a formal say in the negotiation and implementation of these arrangements. To casual observers, the picture that often emerges regarding the role of NGOs in global trade and finance is that of protesters rallying at the site of international economic policy meetings, however, many NGOs work within the international economic decision-making processes through advocacy, participation, and collaboration with both business and government. The U.S.-Canada Free Trade Agreement (USCFTA), implemented in 1989, was a milestone in that it constituted the largest bilateral free trade agreement to date. Its successor, the North American Free Trade Agreement (NAFTA), implemented on January 1, 1994, represented a further landmark in linking together the economies of developed and developing countries in a reciprocal, far-reaching trade treaty. It was also a novel experiment because, as a result of political and citizen pressure exerted during a critical phase, supplemental agreements on labor and the environment were negotiated to help mitigate some of the real or perceived negative spillovers from liberalized trade, especially those associated with integration involving a developing country. One feature of these side agreements was that they expressly provided for citizen groups and NGOs to have input into the deliberations of government officials. The current process to extend free trade to the entire Hemisphere via negotiations of a Free Trade Area of the Americas (FTAA), an ongoing negotiation geared towards eliminating barriers to trade and investment among thirty-four countries in the Americas, more explicitly takes citizen views into account, through a committee established for this purpose. These developments in international trade have opened up new opportunities for civil society. NGOs now view their input and involvement in trade negotiations as routine; governments and companies are less accepting of this new role, but generally acknowledge that NGOs and civil society deserve a “voice” in such deliberations. In this chapter, we explore how NGOs, especially those from the United States and Canada, are exerting increasing pressure on national governments, multinational corporations (MNCs), and international institutions. We focus on efforts by these groups to press governments to incorporate what were previously
Participation by Civil Society and Non-Governmental Organizations
319
considered “extra-trade” commitments in trade and investment agreements, specifically environmental and labor provisions. We begin by offering a brief review of how the emergence of NGOs has influenced debates about globalization in general, and trade and investment agreements, in particular. We develop a conceptual framework for understanding the increasing influence of NGOs in the negotiation and implementation of such agreements. We then examine the emerging role of NGOs in hemispheric integration, focusing on the influence of U.S. and Canadian environmental NGOs within three basic trade and investment agreements: the USCFTA, the NAFTA, and the FTAA process. We find that NGOs are having increasing influence on trade and investment agreements in the Hemisphere, and are poised to take on a major role in multilateral negotiations and agreements.
NGO ROLE IN GLOBAL ECONOMIC POLICY: AN ORGANIZING FRAMEWORK The Emergence of NGOs NGOs, defined as “self-governing, private, not-for-profit organizations that are geared to improving the quality of life of disadvantaged people” (Vakil, 1997, p. 2060), are increasingly active in global issues and debates. The Yearbook of International Organizations put the number of international NGOs at 26,000 in 1999, up from 6,000 in 1990. Their activity in the international economic arena has also grown dramatically: in 1948 the United Nations, the first international organization to incorporate NGO participation, listed 41 consultative groups formally accredited to the UN Economic and Social Council; by 1998 this number had exceeded 1,500 (Simmons, 1998). Over the past several decades, numerous additional groups with specific interest in trade and finance have surfaced. These groups span a wide range of interests, and include environmental, human rights, poverty relief, and other organizations. As their objectives and interests differ, so do their tactics. Some groups have specifically articulated positions that they want incorporated in the policies of governments and international organizations. Others call for a wholesale change in the global economic order, including the elimination of many of the organizations that are responsible for overseeing aspects of the global economy. The recognition of NGOs as influential actors is beginning to gain attention in the international relations and management fields (Doh, 2003; Economist, 2000, 2001; Mathews, 1997; Simmons, 1998). Along with MNCs and international governmental organizations (IGOs), NGOs are increasingly discussed as playing
320
JONATHAN P. DOH AND BARBARA KOTSCHWAR
a significant role in global affairs (see Keck & Sikkink, 1998). Using their network of members, NGOs collect and disseminate information and work collaboratively with national governments, international organizations, and other non-state actors, including MNCs, in furthering social and environmental priorities (Doh, Newburry & Teegen, 2003; Ottaway, 2001).
Understanding NGO Roles: Insiders and Outsiders There have been a number of efforts to define and understand the multiple roles NGOs play in global social and economic matters. One approach distinguishes between “operational” NGOs, which provide social services such as education, health, or human relief, and “advocacy” NGOs, which lobby governments, MNCs, and IGOs (van Tuijl, 1999). Our focus is primarily on advocacy NGOs, and within this category, we draw from the simple framework of Korzeniewics and Smith (2001) who divide NGOs into “insiders” – groups favoring cooperation and working within the system – and “outsiders” – groups that employ more confrontational strategies and wish to upset the system. Although some NGOs may engage in the two strategies simultaneously, and other may transition from one to the other (and back), this typology is helpful for understanding NGO roles in trade and economic policy matters. The fact the NGOs may engage in both cooperative and combative strategies makes interacting with them difficult for policy makers. The protests common at global policy gatherings such as the World Economic Forum and the WTO negotiation sessions exemplify that some NGOs choose not to have their views introduced as part of established institutional frameworks, but rather to take advantage of outsider status to promote their purposes. These outsiders are likely banking on their “anti-establishment” credentials to win support. Some of those shouting in the streets, however, may also deploy representatives to petition their interests within the system. All NGOs utilize their potential independence as a credibility-enhancing mechanism. Given recent scandals and accountability challenges in the private and public sectors, such a role may be particularly important to the viability of NGOs in the future. In this regard, Lindenberg and Dobel (1999) describe how NGOs are increasingly the focus of attention by governments, corporations, donors, and citizens, as they experiment with new strategies, structures, and functional activities. Brown and Moore (2001) argue that this growing scrutiny places new demands on NGOs for accountability, suggesting that international NGOs focused on service delivery will be increasingly accountable to donors and regulators, those oriented toward capacity building will be primarily accountable to clients, while
Participation by Civil Society and Non-Governmental Organizations
321
those focused on advocacy efforts, as is the case in the trade and investment arena, will be especially accountable to political constituencies and influence targets. At the same time, NGOs are increasingly gaining status as organizational institutions whose contributions are inextricably linked to public and private organizational processes, as opposed to “outsiders” providing alternatives to those processes. Recent trends would seem to indicate that NGOs are increasingly choosing to operate as institutional “insiders” in both their advocacy and operational roles as they interact with private and public sector counterparts. Taken to its extreme, NGOs may supplant the efforts of governments. This role can result from the emergence of NGOs as viable alternatives to public sector service provision, but also a result of the outsourcing behavior of governments who increasingly view NGOs as a viable option more connected to the civil societal groups that represent those receiving such services (Parker, 2003). NGOs that have chosen to use “insider” techniques use a number of methods to influence policy. These include reports and studies containing technical information, including legal briefs and opinions to be used in legal fora, Websites for disseminating information and informational campaigns, and training and assistance in technical areas. NGOs sometime pursue both outsider and insider status simultaneously. For example, NGOs have become increasingly active in pursuing boycotts, organizing protests, developing public relations campaigns, while at the same time sponsoring shareholder resolutions in order to press for changes in corporate behavior (Doh, 2002). Friends of the Earth (FOE) sponsored a resolution at the May 30, 2001 Exxon Mobil shareholder meeting that would require the company to move away from fossil fuels toward clean energy, one of hundreds of such resolutions organized by NGOs during 2001 (Economist, 2001). Although most of these proposals are unlikely to win significant support in the short term, such shareholder activism, bolstered by public relations and broader advocacy, will have an increasingly powerful impact on MNCs, especially those involved in politically and socially sensitive industries. Similar examples are evident for NGO influence on trade policy making. Given their operational and technical expertise, access to unique resources, effectiveness in communication and perceived legitimacy in citizens’ minds, private sector and public sector interests are realizing that negotiating with NGOs is imperative in an era of increased globalization and regional integration.
NGO Role in the Issues Life-Cycle In addition to the role of NGOs as “insiders” and “outsiders,” we identify another dimension for understanding the emerging role of NGOs in trade agreements and
322
JONATHAN P. DOH AND BARBARA KOTSCHWAR
negotiations – that of when and how NGOs intervene in a particular policy event or dispute. More specifically, the timing of the NGO role – whether it serves as after-the-fact critic, engages in oversight and compliance while an agreement is implemented, or is involved in the formulation and initial framing of an issue or problem – is a major factor in determining its long-term influence in that particular agreement and its substantive domain. Research in management and organization helps specify the intuitive assessment that NGOs will have greater influence the earlier they intervene in the issue cycle. Building on Vernon’s conceptualization of an evolutionary product life cycle (Vernon, 1996), more recent research has suggested that there are life cycles in which social issues emerge, evolve, and recede. Issues may follow a predictable evolutionary trajectory from policy to learning to commitment (Ackerman, 1975). Mahon and Waddock (1992) identify four “zones” of classification for issues according to the position of pressure groups, public policy groups, and corporate groups with regard to their degree of rejection, indifference, or acceptance of the issues as relevant to their affairs. Their model suggests a linear progression in which public and pressure groups “lead” corporations in their earlier acceptance and action related to specific issues, and that such leading activity has the effect of shaping the impact on corporate adoptions of the issues in questions. In addition, literature on “pioneering advantage” and order of entry suggests that early entrants have opportunities to shape market responses before the business environment has congealed (Lieberman & Montgomery, 1988). Hence, where NGOs are involved in the initial development of trade and investment agreements, they will have the most influence on both the early shape and continuing evolution of agreements. It should be noted, however, that the ability NGOs to intervene early in the development of an agreement is shaped as much by government negotiators and business lobbyists as it is by the activities of NGOs themselves. In addition, NGO influence is neither static nor linear: NGO influence may ebb and flow as the relative salience of different stakeholders shifts as issues evolve (Mitchell, Agle & Wood, 1997). More specifically, an issue or issues can serve as a catalytic event in mobilizing NGOs in pressing for change, and rapidly draw NGOs into a broader debate. International organizations responsible for guiding international economic relations have taken a number of steps to respond to the rising demands for involvement of NGOs in the area of trade and economic integration by providing for NGO involvement earlier in the process of formulation and implementation of agreements. We see insights from these two dimensions in efforts to understand NGO involvement in these bargaining exchanges over the emerging international trading and investment regime within the Western Hemisphere.
Participation by Civil Society and Non-Governmental Organizations
323
CIVIL SOCETY AND NGO INVOLVEMENT IN WESTERN HEMISPHERE INTEGRATION The record of NGO involvement in trade and investment issues and agreements in the Western Hemisphere is still relatively limited, and the mechanisms for incorporating NGO input into trade and integration policy fora are relatively new. The examples presented here are descriptive. As NGOs have increased the scope and focused the strategy of their trade-related activities, countries and international organizations have begun to take their views into account. They have also been able to work in cooperation with – and to benefit from – these organizations. Increasingly, organizations and governments have sought to use input from NGOs to gather technical information to which they themselves may not have easy access and to gauge the tenor of potential popular reaction to proposed policies. At the same time, NGOs have pressured governments and other institutions in the Western Hemisphere to respond to their request for greater input into efforts to liberalize trade and investment within the region (Table 1). Although civil society groups have been vocal on issues related to international labor and environment for some time, NGO activism in international trade and investment in the Western Hemisphere, especially regarding concerns over the relationship between trade liberalization and environmental protection, accelerated in the early 1990s. While a number of international agreements have opened up to receive NGO input and contributions directly, the primary mechanism for NGOs to wield influence is still via the national governments that constitute the main participants to the agreements. U.S.-Canada FTA Under the U.S.-Canada Free Trade Agreement, negotiated in the 1980s, public interest NGOs had little direct input into the negotiation process. Formal trade and industry associations, and informal coalitions of companies, however, had significant input, often formalized through the designation of industry advisory committees (USTR, 2003). In Canada, NGOs such as the Council of Canadians helped to mobilize significant opposition to the agreement, and it was only a split in both provincial and party support that allowed for the re-election of FTA-supporter Brian Mulroney, allowing the agreement to be ratified in Canada (see Hart, 1994, for a thorough insider’s view of the USCFTA negotiations). Another dimension of NGO involvement in the agreements stemmed from longstanding allegations by U.S. lumber producers that Canadian provinces
324
Table 1. Trade Agreements Under Negotiation or in Effect in the Western Hemisphere. Multilateral agreements Regional scope agreements Customs unions
Bilateral agreements (Economic complementation, free and preferential trade, and partial scope)
LAIA/ALADI CARICOM MERCOSUR/MERCOSUL CARICOM – Venezuela Third declaration of Tuxtla
Colombia, Mexico, and Venezuela (G3) NAFTA Canada – Chile Central America – Dominican Republic
Central America – Chile Bolivia – Mexico Costa Rica – Mexico Chile – Mexico
Mexico – European Union Mexico – Nicaragua Canada – Costa Rica Mexico – Northern Triangle (El Salvador, Guatemala and Honduras)
Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico
Nicaragua Panama Paraguay Peru United States Uruguay Venezuela
Signed by
Argentina Bolivia Brazil Canada Central America Chile Colombia Note: Adapted from OAS (2001).
CARICOM – Colombia
JONATHAN P. DOH AND BARBARA KOTSCHWAR
Temporary non-reciprocal agreements General association and cooperation agreements Free trade agreements (subregional)
WTO/GATT FTAA Andean community CACM CARICOM – Dominican Republic Association of Caribbean States
Participation by Civil Society and Non-Governmental Organizations
325
subsidized softwood lumber harvesting on provincial lands, and in so doing, posed unfair import competition to U.S. producers. According to some environmental NGOs, the removal of Canadian subsidies that support over-harvesting of timber may lead to better resource management practices (see Davey, 1996). In this case, to a large part, many U.S. environmental and business groups allied behind a common position. On balance, however, NGO participation in the U.S.-Canada FTA tended to reflect NGOs as “outsiders,” and generally occurred after agreements were reached and negotiations concluded. Hence, in Table 2 and Fig. 1, we see reflected the relatively minimal involvement of NGOs in the formulation of the agreement, and the general posture of NGOs as outsiders.
NAFTA During the NAFTA negotiations, NGOs were already playing or demanding a larger role in public policymaking, and had been inspired by several high profile examples of NGO action. The NGO-spurred boycott on South African goods during the period of apartheid was one demonstration of success in bringing NGO concerns to the international economic arena. Another high profile case that preceded, and largely influenced NGO actions during the NAFTA negotiations, was the so-called tuna-dolphin case. This case was somewhat of a landmark in launching NGO activism in the trade-environmental arena, and also became a cause celebre in the NAFTA negotiations (Deslauriers & Kotschwar, 2003). Under the Marine Mammal Protection Act (MMPA), the United States banned the importation of tuna from certain countries that were caught in a manner that also caused harm to dolphins. The United States faced a GATT challenge from Mexico and Venezuela, two of the countries from which tuna had been banned over these restrictions (see Reinhardt & Vietor, 1996). The Panel ruled against the United States in both cases, in 1991 and 1992, and the reaction from environmental NGOs constituted a rallying cry to efforts by activists to gain access to multilateral trade dispute settlement processes and to incorporate environmental commitments in government-negotiated trade agreements. The panel decision was never adopted by the GATT members (before the reform of the Dispute Resolution system in the Uruguay Round of multilateral negotiations, adoption of the findings of a panel could be blocked by a single member, including the country found to be in violation), and eventually, the matter was resolved through bilateral negotiations that led to the creation of an international dolphin conservation program.
326
Table 2. Role of NGOs in Key Trade Agreements in Which U.S. and Canada are Members. North American Free Trade Agreement
Free Trade Areas of the Americas
Agreement provision
No explicit role
Specific reference in supplemental agreements: (1) Creation of National Public Advisory Committees and Joint Public Advisory Committees (JPAC) to advice the commissioners; and (2) NGOs may initiate dispute settlement procedures;
Invitation for input from “Civil Society” Consideration of supplemental agreements
Effective impact
NGOs engage in discrete intervention on bilateral disputes
Moderate NGOs gained some degree of institutionalized input; Supplemental Agreement on the Environment allows more direct input because of autonomy accorded the Executive Secretariat of Commission and trinational JPAC
Through the submission process for input from Civil Society, NGOs appear to be having an impact on process and content of FTAA
Results to date
Mixed – no consistent NGO impact
Mixed – substantial role in driving environmental disputes
Establishment of specific committee on the participation of civil society “Liberation” of texts, providing transparency and access to negotiating documents
JONATHAN P. DOH AND BARBARA KOTSCHWAR
U.S.-Canada FTA
Participation by Civil Society and Non-Governmental Organizations
327
Fig. 1. Framework for Understanding NGO Influence in Agreements.
Following on the tuna-dolphin cases, pressure for inclusion of environment and labor considerations in international government-government agreements intensified and became part of the political environment in which NAFTA was negotiated. It was in this environment (during the 1992 U.S. presidential campaign) that Bill Clinton called for negotiation of supplemental agreements on labor and environment to improve the NAFTA. Active support of Clinton by a number of environmental groups began a process of close engagement between the incoming Administration and those groups. Around this time, Clinton appointed Jonathan Lash, President of the moderate and highly influential World Resources Institute, as co-chair of the President’s Council on Sustainable Development, a group of government, business, labor, civil rights, and environmental leaders that developed recommendations for national strategies to promote sustainable development. In addition, it was during this period that both Canada and the United States were expanding their industry- dominated trade advisory committee systems to include labor and environmental representations (Cameron & Tomlin, 2000; USTR, 2003).
328
JONATHAN P. DOH AND BARBARA KOTSCHWAR
The NAFTA agreement served as a catalyst for increasing the visibility of concerns about the impact of trade on the environment, and set precedents for subsequent trade negotiations. Broadly, environmental NGOs and trade unions expressed concerns over the agreement’s potential to result in the lowering of standards – a “race to the bottom” – over potential U.S. loss of sovereignty over environmental and labor laws, and the perceived absence of public participation in environmental policies in Mexico (Esty, 1994; Johnson & Beaulieu, 1996; Menz, 1995; Rugman et al., 1999). Three environmental NGOs (ENGOs) – Ralph Nader’s Public Citizen, Friends of the Earth, and the Sierra Club – demonstrated the strength of their opposition by bringing a case against NAFTA to a U.S. District Court (Cameron & Tomlin, 2000). The outcome of the NAFTA debate was that the “insiders” won – the NAFTA passed, however, only with the inclusion of agreements on labor and the environment (see Mayer, 1998, for a thorough analysis of the politics of NAFTA). While some NGOs criticize the strength of the language, claiming the obligations are weak, the mere fact that such side agreements were included can be seen as a victory for the environmental and labor groups. The North American Agreement on Environmental Cooperation (NAAEC), the environmental side agreement that was eventually negotiated in NAFTA, calls for each country to work toward “high and improved levels of government protection of the environment,” essentially promoting “upward harmonization” (see Johnson & Beaulieu, 1996; Menz, 1995). At the same time, the Agreement allows a country or NGO to file a petition if a party exhibits a “persistent pattern of non-enforcement” of domestic environmental law (see NAFTA, 1993). This agreement represents a compromise that the three governments could live with and that NGOs could hold up as a sign that attention is being paid to the link between trade and environmental issues. The U.S. government argued that this gives the agreement “teeth,” given the prospect of trade sanctions in the event of a consistent pattern of non-compliance, while Mexico could maintain that it had not sacrificed its right to govern internal laws and policies (Johnson & Beaulieu, 1996; Menz, 1995). The agreement also created a series of cooperative programs and established an administrative mechanism to oversee its operation. NGO participation in the NAFTA reflected both outsiders and insiders strategies and status. As a result of initial pressure from “outsiders” who opposed the NAFTA, and subsequent overtures from “insiders” (due both to their moderate position and the politically-motivated invitation from U.S. government policymakers), ENGOs ended up having a substantial impact on the development and implementation of the NAAEC. NGOs were involved, to some extent, in its formulation, and are accorded an institutional role in its implementation. Hence, in Table 2 and Fig. 1,
Participation by Civil Society and Non-Governmental Organizations
329
we see reflected the relatively greater involvement of NGOs in the formulation of the agreement, and the general posture of NGOs as both “insiders” and “outsiders.” The NAFTA experience paved the way for and has resulted in continued NGO pressure for involvement in trade and investment liberalization and economic integration in the Western Hemisphere, especially efforts that have clearly built from the NAFTA experience. For example, the free trade agreements between Canada and Chile, Canada and Costa Rica, and the U.S. and Chile, all include supplemental agreements on labor and the environment that closely parallel the NAFTA. Hence, precedents have now been set for potential inclusion of such agreements in subsequent free trade negotiations in the Western Hemisphere, for example, the U.S.-Central American free trade negotiations that were concluded in early 2004.
FTAA Negotiations The negotiations towards a Free Trade Area of the Americas have received a great deal of attention from civil society “outsiders.” The Qu´ebec Summit of the Americas saw huge anti-FTAA protests, as have Ministerial meetings in Buenos Aires and Quito; Websites with names like StopFTAA disseminate information on the shortcomings of the FTAA process, and various NGOs have mobilized public information campaigns opposing these negotiations. The FTAA process itself, building upon the experiences in NAFTA and lessons from the GATT/WTO experience, has taken steps to allow NGOs to make their views known early. The FTAA process came into being at a time during which the NGO movement had gained strength. NGO voices were insisting on being heard in the making of governmental policy, and NGOs had developed a much more sophisticated and broad knowledge base about trade issues. Governments participating in the FTAA negotiations increasingly have to consult with their civil society representatives at the domestic level, and domestic as well as regional NGOs demand participation and voice at the hemispheric level. Initiated at the first Summit of the Americas in Miami in 1994, and formally launched at the Second Summit in Santiago de Chile in 1998, the FTAA negotiations are scheduled to conclude by 2005. In preparation for launching the negotiations, FTAA ministers acknowledged that a range of views existed regarding this initiative within their own countries. In the San Jose Ministerial, they issued the following statement: “We recognize and welcome the interests and concerns that different sectors of society have expressed in relation to the FTAA. Business and other sectors of production, labor, environmental and academic groups have
330
JONATHAN P. DOH AND BARBARA KOTSCHWAR
been particularly active in this matter. We encourage these and other sectors of civil societies to present their views on trade matters in a constructive manner” (FTAA, 1999). While there is no explicit mention of NGOs, the FTAA process includes mechanisms for the participation of civil society. Specifically, within the institutional structure of the FTAA process, a Committee of Government Representatives on the Participation of Civil Society (hereinafter, the Committee) was created to collect these views from civil society, which, in the FTAA context, includes the NGOs discussed in this chapter, but also the private sector, labor, and academia. Participating governments also committed to increase transparency in the negotiating process and to broaden public understanding and support for the FTAA.
Collecting the Views from Civil Society To meet their first mandate, at their initial meeting in October 1998, the Committee extended an “Open Invitation to Civil Society,” soliciting members of the public to submit their views on the FTAA negotiations. This invitation was posted on the FTAA official website and disseminated through national sources. The submissions received were collected, and a summary of the range of views expressed was presented to Ministers in the Committee’s report. There have to date been three rounds of submissions to the Open Invitation which, initially posted for time limited periods, was made permanent at the Buenos Aires Ministerial meeting in April 2001. The process of soliciting views from civil society and disseminating them has evolved throughout the process. The first set of submissions were summarized by the Committee’s report, but were not made available on the FTAA Website, although some individual countries published them (they can be viewed at www.sice.oas.org/ftaa e.asp). The second and third sets of submissions have been posted on the FTAA Website. In addition, Ministers also instructed that the Negotiating Groups receive the submissions from civil society groups “which refer to their respective issue areas and those related to the FTAA process in general” (FTAA, 2001). Thus, the submissions that are received by the Committee are now sent to the negotiators for the various issue areas and distributed throughout the negotiating groups. While many groups hold that there is still too little room for participation by civil society in the negotiations (in fact, this is a common complaint in the submissions), NGOs and others now do have a means for sending their views directly to the negotiators. The submissions that have been received – about 200 as of June 2003 – have come from a varied array of sources. As Table 3 illustrates, contributors have
Participation by Civil Society and Non-Governmental Organizations
331
Table 3. Submissions by Civil Society Groups Under Free Trade Agreement of Americas Process Topical Distribution. Issue
2nd Round
3rd Round
Market access Agriculture Government procurement Investment Competition policy Intellectual property rights Services Dispute settlement Subsidies, antidumping and countervailing duties Smaller economies Civil society Electronic commerce Institutional issues FTAA process
31 (40%) 27 (35%) 13 (16%) 23 (29%) 16 (20%) 15 (19%) 21 (27%) 13 (16%) 18 (23%)
21 (40%) 21 (40%) 15 (28%) 16 (30%) 17 (32%) 22 (42%) 17 (32%) 18 (34%) 17 (32%)
18 (23%) 37 (48%) 5 (6%)
16 (30%) 23 (43%) 14 (26%) 17 (32%)
38 (49%)
Source: Reports of the Committee of Government Representatives on the Participation of Civil Society to the Trade Negotiations Committee (www.ftaa-alca.org).
ranged from business associations; sectoral groups (for example, groups representing airline pilots or the pharmaceutical industry); environmental and labor groups (these represent about 10% of submissions in each round); academics, and a wide range of other organizations and individuals. The opinions expressed include those who are strongly against and strongly in favor of the FTAA, and many submissions contain specific suggestions for how to address particular issues in an eventual FTAA Agreement. The submissions have come from a significant number of participating countries, with the participation rate seeming to correlate with countries whose governments have significant consultations with their own civil society. Table 3 shows that submissions address a range of issues and come from a variety of groups. Submissions ranged from those addressing specific areas of negotiation (agriculture, antidumping and countervailing duties, competition policy, dispute settlement, government procurement, intellectual property, investment, market access, services) to those addressing issues related to business facilitation, transportation, labor standards or environmental protection (FTAA, 1999). In terms of geographic distribution, as seen in Table 4, nearly half of all submissions have come from North America, reflecting the more well-developed non-governmental institutions in North America than in other regions of the Hemisphere, and the concomitant history of channels of formal input by such organizations into government processes. About one third of the submissions
332
Table 4. Submissions by Civil Society Groups Under Free Trade Agreement of Americas Process Geographical Distribution. 2nd Round (April 10, 2000 – September 30, 2000)
3rd Round (November 1, 2001 – May 1, 2002)
66
77
56
50%
47% (of which 72 from United States)
36% (89 from the United States)
17% 13%
3% 48% (of which 70% from Chile)
17% 47% (of which 52% from Argentina)
Source: Reports of the Committee of Government Representatives on the Participation of Civil Society to the Trade Negotiations Committee (www.ftaa-alca.org).
JONATHAN P. DOH AND BARBARA KOTSCHWAR
Total # of qualifying submissions Geographical distribution North America (this includes the NAFTA countries except for the first invitation which only includes Canada and the United States) Central America and the Caribbean South America
1st Round (November 1, 1998 – March 31, 1999)
Participation by Civil Society and Non-Governmental Organizations
333
originated in South America, with the majority of the second round coming from Chile and for the third round from Argentina. The smaller countries in Central America and the Caribbean have contribute, on average, slightly over 10%. The Committee on Civil Society will continue to receive input on the FTAA process, and to disseminate these views both to the FTAA negotiating groups and to the public. The Committee’s mandate has also been strengthened and it has been instructed to consider mechanisms to “foster a process of increasing and sustained communication with civil society” (FTAA, 2001).
Increasing Transparency The Committee’s second mandate is to increase the transparency of the FTAA process. This builds upon earlier efforts in the FTAA termed the “transparencyrelated business facilitation measures,” which included publicly disseminating information on issues such as business visa requirements, trade and tariff data, and publishing contact information for official contact points for each participating country. In addition, the Committee has pledged to make the official FTAA Website (www.ftaa-alca.org; www.alca-ftaa.org; www.zlea.org) more informative and user-friendly and increasing the availability of documents of the process. A major step towards this goal was the publication of the draft FTAA Agreement. Early in 2001, in anticipation of the third Summit of the Americas which was to be held in Quebec City in April, many NGOs called for the “liberation” of the FTAA text, much as they had called for the liberation of the OECD’s Multilateral Agreement on Investment (MAI). Domestic and transnational groups criticized the FTAA negotiators for meeting behind closed doors and not sharing their results. Ministers at the Buenos Aires Ministerial meeting duly recommended to their leaders that this text be “liberated” and the nearly 500-page draft text, reflecting the progress made to that date in each of the negotiating groups, was made publicly available in the four official languages of the FTAA process on July 3, 2001 (see http://www.ftaa-alca.org). Subsequently, the second draft of the Agreement was also made public.
Broaden Public Understanding The Committee has also been mandated to broaden public understanding of the process by holding regional seminars bringing together representatives from civil society and government representatives on FTAA-related issues. The first
334
JONATHAN P. DOH AND BARBARA KOTSCHWAR
seminar for NAFTA countries was held in Merida, Mexico, and a summary of this seminar is included in the Committee’s third report. In addition, the Committee will expand its meetings to include representatives of civil society, and will hold its meetings in locations throughout the Americas, rather than at the central site of the negotiations. Civil society has also strived to make its voice heard at the formal meetings of Ministers. Traditionally, the private sector has held the Americas Business Forum in advance of the Ministerial meetings, and has presented recommendations to the Ministers. A parallel Civil Society Forum was held in Toronto in 1999 and civil society groups have organized similar fora in subsequent meetings. At the last meeting, in Quito, Ecuador, Ministers were invited to this forum, which resulted largely in civil society representatives shouting at the government representatives. For the eighth FTAA Ministerial meeting, held in Miami in November 2003, a committee of civil society representatives was formed to advance an initiative for a “Americas Trade and Sustainable Development Forum” held directly in advance of the Ministerial meeting, with the objective of stimulating a dialogue between civil society representatives and trade ministries (CEIP, 2003). It is too early to gauge the effectiveness of the FTAA’s “insider approach” to civil society or to fully evaluate NGO participation in the emerging efforts to extend trade and investment liberalization throughout the Hemisphere. In particular, it is not at all clear whether the FTAA will have side agreements on issues such as labor and/or environment or other issues of importance to NGOs in the Americas. It does appear as if NGOs are taking on an increasingly active role, facilitated both by governmental willingness to include NGOs in the process, and as a result of NGO efforts themselves.
CONCLUSION: THE FUTURE OF NGO PARTICIPATION IN INTEGRATION As P. J. Simmons writes, the question facing governments and international institutions today is not whether to include NGOs in their decision-making process, but rather “figuring out how to incorporate NGOs into the international system in a way that takes account of their diversity and scope, their various strengths and weaknesses, and their capacity to disrupt as well as to create”(Simmons, 1998, p. 83). NGOs have been increasingly successful in increasing the visibility of trade and focusing attention on the actions of trade negotiators. NGO pressure has clearly brought onto the international economic agenda new issues that had not heretofore been discussed in these fora.
Participation by Civil Society and Non-Governmental Organizations
335
Dynamic Effects We have presented the two dimensions of a framework for understanding NGO influence in agreements as discrete and unrelated. Hence, we have evaluated the degree to which agreements provide for NGOs involvement as a separate construct from the question of whether NGOs approach the particular agreement or issue as “outsiders” or “insiders.” In fact, there is a dynamic interaction between these two dimensions. If NGOs pressure governments to allow them to participate as insiders, government will likely (eventually) respond and accord NGOs such status. Similarly, if governments first accord NGOs access to trade negotiation processes and mechanisms, NGOs will respond by taking advantage of this status and pursuing their role as “insiders.” The interaction between these two dimensions has undoubtedly contributed to the increasing role for NGOs in these agreements, as depicted in Fig. 1.
NGO Participation in Multilateral Trade and Investment Agreements NGOs have increasingly pushed to have greater access to the trade policy process, a system that has historically been limited to governments acting as agents of business, and, to a lesser degree, representatives of labor interests. In addition to regional efforts, NGOs have expressed a great deal of interest is in the trade policy dispute settlement mechanism under the General Agreement on Tariffs and Trade (GATT) and its successor agreement, the World Trade Organization (WTO). The Uruguay Round agreements explicitly makes provisions for cooperation with NGOs. In the Uruguay Round Agreement Establishing the World Trade Organization, Article V(2) on Relations with Other Organizations states that: “The General Council may make appropriate arrangements for consultation and cooperation with non-governmental organizations concerned with matters related to those of the WTO” (see WTO, 1995).
Future Role of NGOs in Economic Globalization Discussions NGOs bring with them resources that can be helpful to trade policy makers. They also bring a new challenge to institutions and organizations working to represent the interests of their constituencies. While in many ways NGOs open up the debate on economic policymaking and create pressure for transparency and accountability for decision makers, there is also the issue of representation
336
JONATHAN P. DOH AND BARBARA KOTSCHWAR
in the NGO community. While the most active NGOs may reflect the views of the civil society groups that they claim to represent, this claim is difficult to test. The challenge to both policy makers and NGOs is to prevent chaos and to forge a space in which the appropriate groups can access the expertise, resources and perspectives brought by NGOs. Governments and international organizations continue their own efforts to forge relations with NGOs. These mechanisms will help to determine the manner in which these actors will coexist on the new playing field of international economic relations.
REFERENCES Ackerman, R. W. (1975). The social challenge to business. Cambridge, MA: Harvard University Press. Brown, L. D., & Moore, M. H. (2001). Accountability, strategy, and international nongovernmental organizations. Nonprofit and Voluntary Sector Quarterly, 30, 569–587. Cameron, M. A., & Tomlin, B. W. (2000). The making of NAFTA. London: Cornell University Press. Carnegie Endowment for International Peace (CEIP) (2003). Civil society and the November 2003 Miami Free Trade Area of the Americas (FTAA) ministerial. Davey, W. (1996). Pine and swine: Canada-U.S. trade dispute settlement. The FTA Experience and NAFTA Prospects. Ottawa: Centre for Trade Policy and Law. Deslauriers, J., & Kotschwar, B. (2003). After Seattle: How NGOs are transforming the global trade and finance agenda. In: J. P. Doh & H. J. Teegen (Eds), Globalization and NGOs: Transforming Business, Government, and Society (pp. 35–64). Westport, CT: Praeger. Doh, J. P. (2002). Nongovernmental advocacy organizations and corporate social responsibility: Ownership activism and issues advocacy. New Academy Review, 1(3), 21–37. Doh, J. P. (2003). Nongovernmental organizations, corporate strategy, and public policy: NGOs as agents of change. In: J. P. Doh & H. J. Teegen (Eds), Globalization and NGOs: Transforming Business, Government, and Society (pp. 1–18). Westport, CT: Praeger. Doh, J., Newburry, W., & Teegen, H. (2003). Cooperative strategies in environmental NGOs. In: J. P. Doh & H. J. Teegen (Eds), Globalization and NGOs: Transforming Business, Government, and Society (pp. 65–80). Westport, CT: Praeger. The Economist (2000, 29 January). NGOs: Sins of Secular Missionaries, 25–27. The Economist (2001, 12 May). Firm Resolutions, 64. Esty, D. (1994). Greening the GATT. Washington: Institute for International Economics. Free Trade Area of the Americas (FTAA) (1999). Fourth trade ministerial: Joint declaration of ministers. San Jos´e, Costa Rica, March 19th, www.ftaa-alca.org/ministerials/costa e.asp. Free Trade Area of the Americas (FTAA) (2001). Sixth trade ministerial: Joint declaration of ministers. Buenos Aires, Argentina, April 7th, www.ftaa-alca.org/ministerials/BAmin e.asp. Hart, M. (1994). Decision at midnight: Inside the Canada-U.S. free trade agreement negotiations. Victoria: University of British Columbia Press. Johnson, P. M., & Beaulieu, A. (1996). The environment and NAFTA: Understanding and implementing the new continental law. Washington: Island Press; Toronto: Broadview Press. Keck, M. E., & Sikkink, K. (1998). Activists beyond borders: Advocacy networks in international politics. Ithaca, NY: Cornell University Press.
Participation by Civil Society and Non-Governmental Organizations
337
Korzeniewics, P. R., & Smith, W. C. (2001, September). Transnational civil society networks and the politics of summitry and free trade in the Americas. Paper delivered to the Inter-American Dialogue. Lieberman, M., & Montgomery, D. (1988). First mover advantages. Strategic Management Journal, 9, 41–58. Lindenberg, M., & Dobel, P. (1999). The challenges of globalization for northern international relief and developments NGOs. Nonprofit and Voluntary Sector Quarterly, 28, 4–24. Mahon, J. F., & Waddock, S. A. (1992). Strategic issues management: An integration of issue life cycle perspectives. Business & Society, 31, 19–32. Mathews, J. T. (1997). Power shift. Foreign Affairs, 76, 50–66. Mayer, F. (1998). Interpreting NAFTA: The science and art of political analysis. New York: Columbia University Press. Menz, F. (1995). An environmental policy for North America Post-NAFTA. North American Outlook, 4(3), 1–15. Mitchell, R. K., Agle, B. R., & Wood, D. J. (1997). Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of Management Review, 22(4), 853–886. The North American Free Trade Agreement, Texts of The Agreement, Implementing Bill, Statement of Administrative Action, And Required Supporting Statements (NAFTA) (1993). Washington, DC: U.S. Government Printing Office. Ottaway, M. (2001). Corporatism goes global: International organizations, nongovernmental organization networks, and transnational business. Global Governance, 7, 265–292. Parker, R. P. (2003). NGOs and multinational enterprise: The prospects for collaboration. In: J. P. Doh & H. J. Teegen (Eds), Globalization and NGOs: Transforming Business, Government, and Society. Westport, CT: Praeger (forthcoming). Reinhardt, F., & Vietor, R. (1996). Starkist. In: F. Reinhardt & R.Vietor (Eds), Business Management and the Natural Environment (pp. 3, 102, 122). Cincinnati: South-Western College. Rugman, A., Kirton, J., & Soloway, J. (1999). Environmental regulations and corporate strategy: A NAFTA perspective. Oxford: Oxford University Press. Simmons, P. J. (1998). Learning to live with NGOs. Foreign Policy, Fall, 82–96. U.S. Trade Representative (2003, April 17). Trade policy advisory committee system. Retrieved from http://www.ustr.gov/outreach/advise.shtml#list. Vakil, A. C. (1997). Confronting the classification problem: Toward a taxonomy of NGOs. World Development, 25(12), 2057–2070. van Tuijl, P. (1999). NGOs and human rights: Sources of justice and democracy. Journal of International Affairs, 52(2), 493–512. Vernon, R. (1996). International investment and international trade in the product cycle. Quarterly Journal of Economics, 80, 190–207. World Trade Organization (1995). Uruguay round agreement establishing the World Trade Organization, V(2). Geneva: World Trade Organization.
HAS NAFTA FULFILLED ITS PROMISE? Bernard L. Weinstein ABSTRACT Effective in 1994, the North American Free Trade Agreement (NAFTA) has substantially lowered tariffs and other trade and investment restrictions. Consequently, three-way trade expanded exponentially as did inward and outward investment in Mexico, Canada, and the United States through 2000. However, both trade and investment decreased in 2001 and 2002 because of a global recession. Critics of NAFTA argue that jobs have been destroyed and wages suppressed in both Mexico and the U.S. But these claims do not hold up to careful scrutiny. Though many points of friction remain among the three NAFTA partners, the agreement must be considered a qualified success.
INTRODUCTION The North American Free Trade Agreement (NAFTA) will celebrate its tenth anniversary shortly. This comprehensive agreement, which was negotiated in the early 1990s and became effective on January 1, 1994, has substantially lowered tariffs and other restrictions on the flow of commodities, products and capital among the United States, Mexico, and Canada. At this juncture, both academics and politicians are asking, “Has NAFTA Fulfilled its Promise?” Without doubt, the debate will be heated and lengthy. North American Economic and Financial Integration Research in Global Strategic Management, Volume 10, 339–348 Copyright © 2004 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1064-4857/doi:10.1016/S1064-4857(04)10018-1
339
340
BERNARD L. WEINSTEIN
In what follows, the achievements of the NAFTA to date are documented and a number of current challenges are discussed and evaluated.
NAFTA TO DATE: WHAT HAS BEEN ACHIEVED? Well before NAFTA’s implementation, trade among the three countries had grown substantially, especially that between the United and Mexico. For example, in the nine-year period between 1984 and 1993, U.S. exports to Mexico nearly quadrupled, rising from $12 billion to $42 billion. But U.S. exports accelerated dramatically after the commencement of NAFTA, in the case of Mexico jumping from nearly $60 billion in 1996 to about $110 billion in 2000 (Fig. 1). Export growth to Canada was also impressive, rising from $135 billion in 1996 to almost $180 billion in 2000. U.S. exports to both Mexico and Canada declined between 2000 and 2002, a reflection of the global economic slowdown, the North American recession, and the aftermath of the September 11th attacks. A similar pattern is seen with regard to imports. Between 1996 and 2000, U.S. imports from Mexico increased from $75 billion to $135 billion while imports from Canada jumped from $155 billion to $230 billion (Fig. 2). In 2001 and 2002, U.S. imports from both Mexico and Canada declined in tandem with a slowing North American economy. Prior to NAFTA going into force, the U.S. was running a small trade surplus with Mexico and a modest deficit with Canada (Fig. 3). By 2001, however, the
Fig. 1. U.S. Exports to Canada and Mexico, 1996–2002 (in Billions of U.S. Dollars). Source: U.S. Census Bureau.
Has NAFTA Fulfilled its Promise
341
Fig. 2. U.S. Imports from Canada and Mexico, 1996–2002 (in Billions of U.S. Dollars). Source: U.S. Census Bureau.
U.S. was recording sizable deficits in its trade with both Mexico and Canada. Critics of NAFTA point to these large and growing deficits as evidence of job destruction in the U.S. following NAFTA, though the deficits are more likely a result of the robust economic expansion that occurred in the U.S. during the 1990s coupled with depreciating currencies in Mexico and Canada.1 The NAFTA has also accelerated two-way investment among the three partners. Canadian direct investment in the U.S. has literally exploded since 1994, rising from less than $5 billion to more than $25 billion in both 1999 and 2000 (Fig. 4). Mexican direct investment in the U.S. also grew significantly between 1994 and
Fig. 3. U.S. Trade Balances (in Millions of U.S. Dollars). Source: U.S. Census Bureau.
342
BERNARD L. WEINSTEIN
Fig. 4. Foreign Direct Investment in the U.S.: Capital Inflows (in Millions of U.S. Dollars). Source: Bureau of Economic Analysis.
2000 – from about $1.5 billion to more than $5 billion. Not surprisingly, inward investment from both Canada and Mexico dropped dramatically in 2001. U.S. direct investment in Canada did not increase significantly between 1994 and 1998 though it surged in 1999, 2000 and 2001 (Fig. 5).2 U.S. direct investment in Mexico actually fell in the first three years of NAFTA but picked up in the late 1990s and reached an all-time high of $15 billion in 2001.
Fig. 5. Foreign Direct Investment in the U.S.: Capital Outflows (in Millions of U.S. Dollars). Source: Bureau of Economic Analysis.
Has NAFTA Fulfilled its Promise
343
HAS NAFTA DESTROYED JOBS IN THE U.S.? Detractors of NAFTA have argued consistently that NAFTA has been bad for American workers in that it has destroyed jobs and held down wages. But these claims do not hold up to careful scrutiny. A 2001 report from the Economic Policy Institute (EPI) claims that NAFTA eliminated 766,030 actual and potential jobs in the U.S. between 1994 and 2000.3 Not only is the EPI estimate derived from an incredibly convoluted methodology, it is also expressed in gross, rather than net, terms. Without question, some American jobs have “disappeared” as Mexican and Canadian imports have supplanted domestic production. But this phenomenon always occurs when the volume of trade among countries expands.4 Total U.S. non-farm employment grew by 20 million (net) between 1994 and 2000 while the nation’s unemployment rate plunged to its lowest level in 40 years. Thus, even if some jobs were “lost” to Mexico and Canada they were more than offset by America’s job generating economic engine. What’s more, total U.S. exports of goods and services jumped from about $700 billion to more than $1 trillion between 1994 and 2000, suggesting millions of the new jobs created during this period were export related. EPI also argues that NAFTA has contributed to growing income inequality and to the declining wages of U.S. production workers (EPI, 2001). Here again, the data do not support EPI’s assertions. In 1994, average weekly earnings in manufacturing were $497 but by 2000 had topped $607 per week. This 22% increase in earnings was well in excess of the 16.2% rise in consumer prices during that period. EPI further argues that NAFTA has depressed wages and incomes in Mexico, notably manufacturing wages which EPI claims fell 16% between 1994 and 1999. But this drop – assuming EPI’s methodology is correct – probably reflects the impact of the 1994–1995 peso crisis and subsequent currency devaluations rather than any insidious direct effects of NAFTA. Between 1994 and 2000, the U.S., Canada and Mexico all posted healthy employment gains. In the case of Mexico, average annual job growth exceeded 3.7%. Without question, the implementation of NAFTA helped produce these gains.
THE BATTLE OVER TIMBER AND WOOD PRODUCTS: CANADA VS. THE U.S. Implementation of NAFTA has not occurred without a number of ongoing disputes, timber being a prime example. In 2001, Canada exported $6.5 billion of softwood lumber and wood products to the U.S. The American lumber lobby,
344
BERNARD L. WEINSTEIN
asserting that Canadian lumber was unfairly subsidized, pressured the U.S. Trade Administration to impose countervailing duties averaging 27% on Canadian imports in May of 2002. Not surprisingly, the Canadian government has protested loudly to the World Trade Organization (WTO) who, in turn, has ruled that the U.S. violated trade rules in applying the duties. One of the problems with timber is that provincial governments own most Canadian stands while most American forests are privately owned. Thus, it is extremely difficult to identify just what is a subsidy and what isn’t. Nonetheless, the upshot has been pain on both sides of the border. In the face of overcapacity and falling prices, 114 American mills have shut down over the past year, as have 51 in Canada. Still, because of greater efficiencies in their remaining mills, Canadian firms have maintained their share of the American market. The irony of this dispute is that the alleged subsidies on the part of the Canadians and the countervailing tariffs imposed by the U.S. are not a consequence of NAFTA but rather a reflection of market distortions in both countries. And the true losers are consumers of both countries who are paying more for lumber and home construction than would be the case in a truly competitive marketplace.
MEXICO’S FARMERS: FLOUNDERING IN A TARIFF-FREE LANDSCAPE On January 1st, 2003, tariffs were eliminated on a host of agricultural products in accordance with NAFTA. Many Mexican farmers are angry, claiming that the treaty unfairly favored U.S. farmers (Economist, January 25, 2003). Facing a congressional election this year, Mexican President Vincente Fox has increased farm subsidies by $1.3 billion and started to erect barriers to U.S. food imports, including new tariffs on apples and chicken parts (Luchnow, 2003). The U.S. has threatened to retaliate. Here, again, politics is trumping economics in that two-way trade in food products has expanded steadily under NAFTA (Fig. 6). Mexican food imports from the U.S. more than doubled between 1993 and 2001, rising from $3.6 billion to $7.4 billion. But exports of food to the U.S. also doubled, from $2.7 billion to $5.2 billion. Mexican fruit and vegetable farmers, who benefit from longer growing seasons and cheaper labor than their American competitors, have been particularly fortunate. And food prices are cheaper than ever before for poorer Mexicans (Luchnow, 2003). The real problem is not NAFTA but Mexico’s failure to rationalize its agricultural sector, even though they had nearly a decade to implement the requisite
Has NAFTA Fulfilled its Promise
345
Fig. 6. Agricultural Trade Between Mexico and the U.S. Source: Dallas Morning News.
changes. When NAFTA took effect in 1994, Mexican officials argued that market discipline would force the nation’s farmers to modernize. In fact, many farmers did boost their efficiency, making Mexico the number two agricultural supplier to the U.S. after Canada (Case & Corchado, 2002). But millions of other Mexican farmers, because of a lack of capital and/or know-how, did not make the transition. Consequently, about eight million people – 22% of Mexico’s active labor force – work in the countryside though they generate only 4.4% of gross domestic product (Economist, November 30, 2002). The
Fig. 7. Breakdown of Subsidies to Mexican Farmers. Source: Mexican Government; U.S. Department of Agriculture.
346
BERNARD L. WEINSTEIN
Fig. 8. Farmers in Mexico are Better Subsidized than in the U.S. (Guaranteed Income in Dollars per Ton). Note: For Spring 2001 and Winter 2001–2002 crops. Source: Mexican Government; U.S. Department of Agriculture.
overwhelming majority are poor subsistence farmers who work plots as small as two acres. But they do vote. Mexican governments, federal and state, can also be faulted for failing to take advantage of the ten-year transition period while tariffs were being phased out by investing in infrastructure improvements such as irrigation and roads. High production and transportation costs have combined to make many Mexican farmers uncompetitive. As in the U.S., most Mexican agricultural subsidies go to large farmers (Fig. 7). And Mexican government data indicate many of these large farmers are more heavily subsidized than their U.S. counterparts (Fig. 8). Though this policy may make some sense in the short-term, ultimately Mexico must deal forthrightly with the root problem which is too many campesinos producing too little corn. Because corn production is so central to Mexican agriculture, utilizing about 55% of cultivated land, it was afforded special protection under NAFTA with a 206% tariff to be phased out by 2008. Can Mexico’s three million corn-growers, each with five dependants on average, become competitive in the next five years?
OTHER AREAS OF “FRICTION” AMONG THE NAFTA PARTNERS Many policy issues remain unresolved among Mexico, Canada and the United States, though most of them would have occurred even without NAFTA. Immigration is perhaps the most contentious of these issues.
Has NAFTA Fulfilled its Promise
347
Prior to the September 11 attacks, a comprehensive accord to safeguard the rights of the estimated 10–12 million legal and illegal Mexican immigrants living in the U.S. was close to ratification. But terrorism and recession have dealt a temporary blow to such an agreement. Another irritant is Mexico’s opposition to the death penalty. In January, Mexico asked the International Court of Justice to rule against the execution of 51 Mexicans on America’s “death rows.” NAFTA protocols regarding cross-border trucking have been held hostage to parochial political interests, while drug- and terrorism-interdiction programs are resulting in lengthy and costly delays at both Mexican and Canadian border crossings. And a host of environmental protocols, especially as regards air and water quality along the U.S.-Mexican border, have been widely ignored. Even the Mexican government admits that environmental degradation post-NAFTA has accelerated and is now equivalent to about 10% of annual GDP (Wise & Gallagher, 2002).
CONCLUSION Though criticism of NAFTA abounds, and trade among Mexico, Canada and the United States has declined since 2000 because of a global recession, the trade agreement must be considered a qualified success. Trade and investment among the three countries have grown at a rapid clip post-NAFTA, while each has recorded accelerated job growth, higher real incomes, and lower rates of inflation. Many challenges and obstacles must be overcome before the full economic potential of NAFTA is realized. But most importantly, business leaders and public officials in Mexico, Canada and the U.S. must work diligently to dissipate the latent protectionist sentiment that still exists in each country.
NOTES 1. See discussion below. 2. American investment in Canada was already at a high plateau and had risen significantly after implementation of the Canada-U.S. free trade agreement in 1989. 3. EPI defines “potential jobs” as positions that would have been created if the trade deficit with Mexico and Canada had remained constant, in real terms, and everything else in the economy had been held constant as well. 4. For the first 65 months of NAFTA, the number of workers certified as being dislocated by trade with Mexico and Canada and eligible for trade adjustment assistance averaged 3662 per month (Hinojosa-Ojeda et al., 2000). Extrapolating this average for the 1994–2000 period would produce about 308,000 dislocated workers, well below the EPI estimate.
348
BERNARD L. WEINSTEIN
REFERENCES Case, B., & Corchado, A. (2002). Tarriff cuts troubling to Mexico. Dallas Morning News. Economic Policy Institute (2001). NAFTA at seven: Its impact in all three nations. Washington: EPI. Floundering in a Tariff-free Landscape (2002). The Economist, 31–32. Half an Enchilada (2003). The Economist, 37–38. Hinojosa-Ojeda, R., Runsten, D., DePaolis, F., & Kamel, N. (2000). The U.S. employment impacts of North American integration after NAFTA: A partial equilibrium approach. Retrieved February 6, 2003, from http://naid.sppsr.ucla.edu. Luchnow, D. (2003). Of corn, NAFTA and ZAPATA. The Wall St. Journal. Wise, T. A., & Gallagher, K. P. (2002). NAFTA: A cautionary tale. Retrieved February 6, 2003, from LexisNexis Database (LexisNexis Current Issues, R247–117) at http://www.lexisnexis.com/ciuniv.
AUTHOR INDEX AAFC (Agriculture and Agri-Food Canada), 226 Abbott, K., 249, 260 Acharya, R. C., 37 Ackerman, R. W., 322 Adams, D. M., 241, 251 Agle, B. R., 322 Ahmed, A., 198 Akehurst, G., 104 Alexander, N., 104, 105 Allen, D. E., 139 Altshuler, R., 66 AMAD, 226 American Chamber of Commerce, 119, 131, 133 Ames, G., Anderberg, M., 158 Anderson, W. P., 196, 198, 199 Appel Molot, M., 177 Arce, D. G. M., 95 Arshanapalli, B., 139 Artis, M., 155, 156, 158, 167 Associated Press, 308 Babcock, B., 219 Banfield, J., 159 Bargas, S., 198 Barham, B., 311 Baroody, L., 312 Baxter, M., 155, 156 Bayoumi, T., 154, 155, 181 Beaulieu, A., 328 Beaulieu, E., 30 Beine, M., 177 Bellak, C., 66 Benedict, M. E., 196 Binkley, C., 253 Blackman, S. C., 139 Blanchard, O. J., 155, 181
Blank, S., 177, 178 Blonigen, B. A., 48, 57, 59, 60 Bluestone, B., 304 Bock, H., 159 Boreiko, D., 158 Bothwell, R., 308 Brain, C., 14, 70 Braudo, R., 240, 248, 253, 255 Brennan, T., 301, 307, 312 British Columbia, 257 Brown, L. D., 320 Brunelle, D., 39 Buckley, P. J., 23 Buiter, W., 154, 155 Bunker, S., 305 Buogiorno, J., 242 Burt, S. L., 104 Business Week, 98, 102, 109, 110 Byers, J. D., 139 Cameron, M. A., 327, 328 Canada Customs, 292 Canada-U.S. Partnership Forum, 118 Cantwell, J., 66 Carnegie Endowment for International Peace, 334 Carrier and 3PL Management and Driver Interviews, 289, 293–295 Case, B., 345 Cashore, B., 238, 253 Centre for the Study of Living Standards, 36, 40 Chan, K. C., 139 Chaudhuri, K., 139 Cheung, Y., 141 Choudhry, T., 139 Chowdhury, A. R., 139 Ciccantell, P., 304–306 349
350 Clegg, J., 23 Clinebell, J. M., 139, 147 Coiteux, M., 177, 178, 181, 189 Colla, E., 104, 108 Conference Board of Canada, 296 Consumer Report, 97 Cooke, J., 295 Coomes, O., 311 Corchado, A., 345 Corsetti, G., 154 Coughlin, C. C., 82 Coulombe, D., 42 Coulombe, S., 177 Council of Economic Advisers, 43 Crone, T., 158 Croux, C., 156 Crowley, P., 154, 156, 164 Crucini, M., 29 Darrat, A. F., 139, 147 Davey, W., 325 Davidson, L. S., 83 Davies, R. B., 48, 57, 59, 60 Davis, D., 204 Davis, H. W., 290 De Grauwe, P., 155 Deblock, C., 39 DeFusco, R. A., 139 Deloitte Touche Tohmatsu, 98, 101, 104, 107 DeSerres, A., 155, 181 Deslauriers, J., 325 DeVoretz, D., 42 Dickey, D. A., 140, 143, 149 Dillon, J., 304, 306 Dobel, P., 320 Dobson, 286, 296 Doh, J. P., 319–321 Donaldo, J., 143 Dorfman, J., 117 Doukas, J., 139 Dowling, J., 309, 312 Drache, D., 304 Drennen, C., 196 Drysdale, P., 177 Dukert, J., 306, 308, 309 Dunn, R., 132 Dunning, J. H., 39, 49, 50, 198
AUTHOR INDEX Dupasquier, C. R., 181 Dupuis, M., 104, 108 Dymond, W., 296 Economic Policy Institute, 343 Eden, L., 23, 177, 198 Egelton, R., 42 Eichengreen, B., 154, 155, 181 Enders, W., 95 Energy Information Administration (EIA), 306–308, 310–312 Engel, C., 27, 28, 37, 43, 118 Engle, R. F., 140, 149 Esty, D., 328 Eun, C. S., 139 Evans, T., 24 Ewing, B. T., 139, 147 Fairhurst, A., 104 Fama, E., 147 Feigenbaum, H., 246 Fellmeth, R., 312 Financial Times, 98, 104, 109–111 Flam, H., 43 Flynn, S., 86–88 Forbes, 98 Forget, C. E., 204 Forni, M., 156 Forsans, N., 23 Fortune, 98, 102 Fraley, C., 159–162 Frankel, J. A., 177 Frankel, J., 43, 155 Fratianni, M., 154 Free Trade Area of the Americas (FTAA), 330, 331, 333 Fuller, W. A., 140, 143, 149 Gagn´e, G., 239 Galbraith, J., 158 Gandhi, P. P., 196 Gaston, N., 30 Geandet, R., 181, 189 Gerlach, H., 156 Girod, S., 14, 98, 104 Gitlitz, J., 308 Globerman, S., 19, 23, 39, 42, 43 Goldstein, J., 249
Author Index Golob, S. R., 90 Gorte, R., 238, 239, 253, 258 Government of Canada, Department of foreign affairs and international trade, 42, 43 Govindarajan, V., 97 Grafton, Q., 260 Granger, C. W. J., 140, 149 Gray, S. J., 66 Green, M. B., 198 Grinspun, R., 305 Grubel, H., 43 Grubert, H., 66 Gupta, A. K., 97 Hamilton, G., 253, 257 Handelman, S., 296 Hansen Harps, L., 274 Harris, R. G., 37, 176 Harrison, B., 304 Harrison, K., 246 Hart, M., 40, 296, 323 Hartigan, J., 158 Hassan, M. K., 139 Haynes, R., 241 Hejazi, W., 5, 12, 49, 66 Helliwell, J. F., 20, 42, 117, 118 Hine, R. C., 19 Hines, J. R. Jr., 48, 59 Hinojosa-Ojeda, R., 347 Hoberg, G., 246 Hodgetts, R., 5 Honohan, P., 158 Horlick, G., 306 Hufbauer, G. C., 19, 177, 265 Humphrey, B., 312 Hung, B., 141 Hymer, S., 141, 198 IBAT (International Bilateral Agricultural Trade), 214, 215, 217–219, 231 ICF Consulting, 292 Im, E. I., 95 Imbs, J., 156 Inder, B., 158 Industry Canada, 215, 265, 271 Irland, L., 245, 251
351 Jabara, C., 228, 230 Jackson, T. P., 104 Jacquemin, A., 158 Jansson, P., 43 Jenkins, M., 118, 132 Jeon, B. N., 139 Jiaqing, L., 158 Johansen, S., 137, 140, 143, 150, 151 Johnson, D., 27 Johnson, P. M., 328 Journal of Business Strategy, 101 Jurewitz, J., 312 Juselius, K., 137, 151 Kanas, A., 139 Karolyi, G. A., 139 Kasa, K., 139 Kasoff, M. J., 196 Kass, R., 161 Kaufman, L., 158 Keck, M. E., 320 Kenen, P., 154 Kenna, K., 287 Kim, W. C., 109 Knubley, J., 198 Kojima, K., 177 Koo, J., 156 Korzeniewics, P. R., 320 Kotschwar, B., 325 Krause, W., 198 Krauss, C., 308 Krishnan, T., 160 Krolzig, H. M., 167 Kwan, A., 139 Lafrance, R., 27 Lalonde, R., 155, 181 Lambert, B., 292 Lauer, S., 196 Lawson, B., 286 Lee, D. R., 95 Lee, J., 20 Lee, K., 251 Legault, M., 198 Lieberman, M., 322 Lipsey, R. E., 50 Litvak, I. A., 195, 204 Lo, A. W., 141
352 Logistics Management and Distribution, 277 Luchnow, D., 344 Lynch, R., 260 MacDonald, D., 139, 288 Macdonald, R., 182, 304, 305 MacFarlane, S., 286 MacKinlay, A. C., 141 Macklin, A., 89, 90, 95 MacPherson, A. D., 196 Maharaj, E., 158 Mahon, J. F., 322 Makki, S. S., 235 Mander, J., 304 Mann, H., 306 Maring, G., 292 Mark, J., 274 Martinez, S., 301, 307, 312 Mathews, J. T., 319 Mauborgne, R., 109 Maule, C. J., 195, 204 Mayer, F., 328 McCallum, J., 20 McConnell, J. E., 196 McGoldrick, P. J., 105 McKeever, T., 245 McKinnon, R., 154 McLachlan, G., 160 McPherson, A. D., 199 Mellahi, K., 104 Menz, F., 328 Meyer, J. R., 196 Meyer, S. P., 198 MIT Sloan Management Review, 104 Mitchell, R. K., 322 Monadjemi, M. S., 139 Monaghan, C., 309 Monteils, A., 23 Montgomery, D., 322 Moody, K., 304 Moore, K., 103 Moore, M. H., 320 Morey, C., 274 Morgan, A., 288 Mukerjee, S., 158 Mundell, R., 153, 154 Munroe, T., 312
AUTHOR INDEX Myers, J., 195, 199 Myneni, G., 251 Myre, P., 260 NAFTA Secretariat, 267, 328 Naka, A., 139 National Post, 92, 287 Natural Resources Canada, 244, 247, 260 Nelson, H., 260 Newburry, W., 320 Newey, W. K., 150 Newlon, T. S., 66 O’Connell, P., 123 OECD (Organization for Economic Cooperation and Development), 9, 48, 224–226, 269 O’Hagan, S., 196, 198, 199 O’Neil, 257 Organization for Economic Cooperation Osterwald-Lenum, M., 151 Ottaway, M., 320 Palmer, K., 301, 307, 312 Parker, R. P., 321 Parsley, D., 120, 133 Pastor, R., 304 Pedersen, T., 156 Peel, D. A., 139 Perron, P., 140 Perry, L., 139 Phillips, P., 140 Pollard, P. S., 82 Polzin, P., 242 Prakash, S., 37 Prestemon, J., 242 Quah, D., 155, 181 Raff, H., 53 Raftery, A., 159–162 Rajan, R. S., 177 Rao, S., 198 Ratnesar, R., 110 Reardon, J., 104 Reed, L., 238 Reeder, J., 228, 230 Reichlin, L., 156 Reilly, K., 23
Author Index Reinhardt, F., 325 Richards, A. J., 139 Richardson, J. D., 132 Richardson, K., 196 Ritchie, G., 239 Rogers, J. H., 27, 28, 37, 43, 118, 132, 139 Rose, A., 43, 155 Ross, M., 248 Roubini, N., 154 Rousseeuw, P., 158 Royce, K., 309, 312 Rugman, A. M., 4–6, 9–11, 14, 19, 48, 66, 70, 82, 98, 103, 104, 175, 195, 328 Sadeque, Z., 198 Safarian, A. E., 5, 8, 66 Salmon, W. J., 104 Sandler, T., 95 Sapir, A., 158 Saunders, J., 253, 258 Schembri, L., 27 Schott, J. J., 177, 265 Schuchhardt, C., 306 Schulz, J. D., 290 Shapiro, D., 23 Shim, S., 139 Shrybman, S., 304, 305 Sikkink, K., 320 Simmons, P. J., 319, 334 Simon, B., 312 Simpson, E. M., 104, 105 Smith, K. L., 139 Smith, R., 231 Smith, T., 312 Smith, W. C., 320 Someshwar, R., 37 Sparks, L., 104 Spelter, H., 245 St-Amant, P., 155, 181 Statistics Canada, 131–133 Stockman, A., 155, 156 Strong, M., 295 Stulz, R. M., 139 Summers, L., 178 Sun, C., 241 Tavlas, S. G., 155, 181 Taylor, A., 312
353 Taylor, R., 259 Teegen, H., 320 Tellier, P., 176 Telmer, C., 29 The Economist, 53, 98, 319, 321, 344, 345 The Wall Street Journal, 98, 102, 107, 109, 111 Thorpe, D. I., 104, 105 Tomain, J., 312 Tomlin, B. W., 327, 328 Tordjman, A., 104 Toro, J., 167 Tower, E., 181 Transport Canada, 265, 272, 279 Treadgold, A., 104 Trebilcock, M., 240, 248, 253, 255 Trefler, D., 30, 40 Trickey, M., 287, 288, 296 U.S. Congress, Joint Committee on Taxation, 57 U.S. Department of Agriculture, 243, 247 U.S. Department of Commerce, 66, 257 U.S. Department of Transportation Office, 255, 266 U.S. General Accounting Office, 264, 267 U.S. International Trade Commission, 241, 242, 267 U.S. Trade Representative, 323, 327 United Nations Conference on Trade and Development, 51 United States Bureau of Mines, 305 USDA/ERS, 220 Useem, J., 98 Vakil, A. C., 319 Van Kooten, G. C., 251 van Ooyen, A., 158 van Tuijl, P., 320 Vanhavebeke, P., 155 Verbeke, A., 70, 82 Vernon, R., 198, 322 Vida, I., 104 Vietor, R., 325 von der Ruhr, M., 49, 53 Von Furstenberg, G. M., 139 von Hagen, J., 154
354 Waddock, S. A., 322 Wall, H. J., 70, 82 Waller, C., 154 Wal-Mart, 102, 107 Wasserman, H., 312 Wear, D., 251 Wei, S. J., 120, 123, 133 Weiner, T., 309 West, K. C., 150 Willett, T. D., 181 Windsor Star, 286, 296 Wojciech, S., 42
AUTHOR INDEX Wood, D. J., 322 World Trade Organization, 335 Wynne, M., 156 Yan, B., 27 Yeats, A. J., 177 Yip, G. S., 103, 108 Zachariadis, M., 29 Zhang, D., 241, 251, 253 Zhang, W., 155, 156, 158 Zhong, M., 139, 147
SUBJECT INDEX Adjustment, 5, 32, 42, 51, 119, 137, 138, 145, 186, 187, 225, 240, 347 Agriculture, 35, 41, 214, 216, 224–226, 228, 234, 247, 275, 331, 345, 346 Banking, 31, 51, 54, 66, 202, 320 Banks, 15, 221, 312 Bill, 213, 214, 216, 219, 220, 228, 231, 232, 234, 280, 327 Bonds, 186, Border, Border crossings, 41, 42, 86, 87, 91, 118, 283, 284, 287, 288, 347 Border States, 69, 72–83 Brokers, 201, 276, 279, 291, 293, 300 Budgets, 89, 90, 154, 172, 186 Business cycle, 42, 153, 154–156, 166, 167, 177, 268,
Crops, 213, 217, 219, 220, 228, 232, 234, 246, 346 Currency, 13, 17, 27, 41–43, 59, 115, 118, 120, 123, 129, 130, 141, 153–155, 157, 166, 167, 175, 176, 180, 181, 187, 188, 197, 206, 207, 343 CUSTA, 18–25, 27–30, 34–37, 40–42, 227, 229, 232 Customs, 41, 91, 92, 201, 271, 275–227, 280, 281, 283, 285, 286, 288, 292–294, 296, 324 Debt, 32, 49, 50, 54–56, 60, 62–65, 99, 154, 186, 187 Deviation, 39, 43, 120, 141, 142, 173 Distance, 27, 28, 56, 59, 61–63, 69–74, 77, 78, 80, 81, 83, 199, 201 Dollar, 20, 28, 30, 31, 36, 37, 41, 55, 59, 117, 129–131, 141, 154, 175–177, 180–182, 186, 187, 189, 201, 244, 286, 289, 295, 310
Canada, 3–15, 17–32, 34–43, 51, 54–56, Economic, 1, 3, 5, 8, 17–21, 23–26, 32, 34, 58, 66, 67, 69–83, 85–95, 101, 102, 40, 41, 48, 54, 56, 59, 66, 104, 105, 110, 117–124, 126–147, 153–157, 111, 117, 120, 137, 138, 153–155, 164, 166, 167, 170, 172, 173, 157, 166, 167, 172, 176–178, 176–183, 185–189, 194–203, 180–182, 187–189, 193, 194, 205–207, 213–215, 217–219, 221, 196–198, 200–202, 207, 208, 214, 224–226, 228–235, 237–241, 237, 240, 242, 246, 247, 249, 251, 244–248, 250–260, 263–266, 253, 256–258, 260, 264, 268, 275, 268–281, 283–289, 292–296, 301, 276, 278, 286, 299, 300, 302–305, 302, 304–306, 308–310, 317, 318, 309, 311, 312, 317–320, 322, 325, 323–327, 329, 332, 339–347 329, 334–336, 340–343, 347 Civil Society, 317, 318, 323, 326, 329–334, Efficiency, 40, 41, 91, 101, 109, 147, 303, 336 345, Clusters, 103, 158–164, 166, 167 Electricity, 299–311 Co-movement, 137–139, 141, 156 Environment, 104, 105, 107, 138, 180, 187, Competitiveness, 245, 246, 304 217, 224, 225, 246, 250, 253, 254, Concentration, 35, 195, 279, 284 258, 264, 268, 272, 274, 275, 279, Corporate Strategies, 310 355
356 280, 287, 304, 307–309, 311–313, 318–320, 322, 323, 325, 326–329, 331, 334, 347 Equation, 39, 43, 123, 125, 128, 150 Equilibrium, 93, 96, 140, 145, 149, 181, 182, 183 Equity Markets, 137–147 Euro, 153, 206 Exchange Rate, 23, 27, 28, 31, 35–39, 41–43, 59, 61–63, 120, 121, 129, 130, 132, 133, 154, 155, 176, 181–185 188, 189, 241, 255, 273 Exports, 5, 11–14, 19–24, 35–37, 42, 53, 54, 69, 70, 72–82, 87, 93, 95, 117, 177, 179, 182, 185, 189, 194, 214, 215, 217–219, 227, 229–232, 234, 237–239, 241–243, 245, 248, 250, 252, 258, 260, 269, 286, 288, 304, 305, 340, 343, 344 Farm, 213, 214, 216, 219, 220, 223–225, 227, 228, 230–234, 344 FDI, 3–15, 19, 20, 23–25, 27, 37, 39, 42, 43, 47–67, 71, 107, 198 Financial, 6, 15, 47, 48, 98, 99, 111, 115, 138, 172, 176, 185, 186, 202, 232, 255, 260, 275, 276, 312 Financial Markets, 185 Flows,18–21, 23, 24, 27, 30, 37, 41–43, 48–51, 54–56, 58–60, 62–66, 71, 83, 87, 89, 93, 117, 138, 182–188, 193, 194, 196, 237, 268, 276, 283, 288, 342 Foreign Direct Investment (FDI), 18, 47, 48, 70, 233 FTA, 4–6, 9–11, 40, 66, 89, 240, 253, 323, 325, 326 FTAA, 47, 49, 59, 66, 319, 324, 326, 329–334 Global, 47, 55–57, 63, 69–73, 78, 81, 82, 83, 86, 95, 97, 98, 100, 102–108, 178, 240, 299–301, 305, 309, 311, 312, 318–320, 339, 340, 347 Globalization, 47, 48, 70, 71, 82, 88, 103, 104, 108, 175, 300–302, 304, 309, 310, 312, 318, 319, 321, 335 Governments, 35, 188, 221, 223, 233, 240, 245, 248, 250, 251, 253, 255, 264,
SUBJECT INDEX 274–277, 279, 280, 301, 309, 312, 318–321, 323, 328–331, 334–336, 344, 346 Hegemon, 304 Immigration, 19, 25, 26, 42, 87–90, 92, 94, 264, 265, 271, 275, 276, 280, 281 Imports, 11–14, 19, 21–23, 35, 36, 42, 177, 179, 214, 215, 218, 227–230, 232, 233, 238, 240, 241, 244, 246, 250, 252, 253, 255, 257–259, 269, 278, 286, 287, 294, 295, 340, 341, 343, 344 Industry mix, 69–72, 82, 83 Integration, 1, 3, 6, 7, 12, 17–20, 25–27, 29, 30, 32, 34, 37, 39, 40, 41, 48, 59, 66, 69, 73, 81, 82, 103, 115, 118, 119, 123, 137–139, 145, 147, 154, 175, 176, 186, 193, 194, 197, 199, 206–208, 237, 263, 265, 269, 275, 278, 279, 304, 308, 309, 317–319, 321–323, 329, 334 Intercompany, 49, 50, 54, 55, 60, 65 Internal, 13, 39, 55, 58, 250, 328 Internalization, 49, 50 International Trade, 35, 36, 43, 53, 70, 71, 92, 178, 193, 206, 227–230, 238, 246, 250, 267, 269, 288, 318, 323 Internationalization, 97, 102, 104–107, Intra-Regional Trade, 3–15, 39, 70, 103, 108, 242 Inward, 5, 23–25, 37, 43, 50, 51, 57, 59, 286, 339, 342 Joint Ventures, 107, 108 Labor, 18, 19, 24–26, 29, 30, 37, 39, 41, 43, 51, 108, 109, 138, 157, 166, 172, 195, 197, 199, 202–205, 263–265, 267, 268, 273, 277, 278, 286, 304, 318, 319, 323, 327–331, 334, 335, 344, 345 Latin America, 25, 51, 56, 59, 66, 101, 107, 139, 204, 300, 310 Law of one Price, 27–29, 37, 118, 119, 132 Location, 49, 50, 58, 59, 69, 71–73, 95, 99, 103, 104, 119, 131, 193–196, 199–201, 203, 204, 207, 242, 265, 272, 296, 302
Subject Index Lumber, 233, 237–260, 323, 325, 343, 344 Manufacturing, 4, 11–13, 15, 30–32, 34, 35, 43, 50, 51, 53–55, 71, 103, 197, 198, 200, 242, 244, 248, 271, 243 Markets, 7, 8, 15, 17, 19, 25, 26, 39, 42, 51, 54, 99, 102–109, 111, 118, 119, 122, 137, 138–147, 157, 182, 185, 186, 195, 196, 198, 199, 213, 217, 227, 239, 244, 245, 246, 248, 251, 252, 258, 260, 264, 268, 279, 290, 301–303, 305, 306, 309, 311, 312 Matrix, 72, 93, 140, 150, 151, 159, 160, 162, 163, Mexico, 3, 5, 7, 9, 10, 18, 19, 21, 22, 24, 42, 54–56, 58, 66, 73, 74, 75–83, 97–99, 101, 102, 107–110, 137–147, 153, 154, 156, 157, 164, 166, 167, 171–173, 176, 178–180, 183, 189, 194, 197, 238, 263–271, 273–275, 277–280, 301, 302, 305, 306, 309, 324, 325, 328, 334, 339–347 Multinational Enterprises (MNEs), 4, 5, 11, 14, 15, 48, 49, 51, 55–57, 59, 70, 103, 105, 106, NAFTA, 3–15, 18–25, 30, 35, 37, 40, 41, 42, 47, 48, 54, 59, 61–63, 67, 71, 82, 87, 91, 92, 97, 101, 107, 109, 110, 117, 137–147, 153, 154 156, 157, 166, 167, 175, 176, 179, 180, 188, 189, 194, 196, 197, 227, 229, 232, 233, 240, 254, 255, 256, 260, 263–281, 288, 296, 299–301, 304–306, 308, 309, 318, 319, 324, 325, 327–329, 332, 334, 339–347 National treatment, 6, 13, 15, 266, 267, 305 Negotiation, 17, 65, 266, 290, 300, 318–320, 323, 324, 327, 335 Networks, 27, 118, 271, 305, 306, 309 NGO, 317, 319, 320–323, 325–329, 334–336 Optimal currency area, 153, 154, 157, 166, 167 Origin, 107, 216, 229, 231, 232, 271, 275, 276
357 Outward, 4, 5, 9, 23, 25, 48–60, 65, 339 Ownership, 4, 8, 41, 49, 50, 138, 198, 243, 247, 248, 253, 263, 265, 266, 267, 303 Perimeter, 85–95, 279, 281, 287, 296 Policies, 47, 49, 56, 59, 65, 66, 71, 85–95, 187–189, 235, 244, 245, 247–250, 255, 257, 258, 263, 275, 280, 281, 306, 319, 323, 328 Prices, 17, 19, 26–29, 37, 40, 42, 55, 66, 98, 107–111, 117–134, 140–142, 147, 155, 181, 182, 189, 213, 216, 219, 220, 222, 232, 239, 241, 242, 245, 248, 251, 257–259, 272, 273, 289, 299, 300, 303, 310, 343, 344 Production, 8, 19, 30, 40, 50, 53, 85, 156, 201–203, 213, 221, 224, 226, 227, 230, 232–235, 241–245, 254, 258, 259, 260, 269, 275, 286, 295, 303, 306, 311, 329, 343, 346 Productivity, 36, 40, 43, 98, 182, 183, 185, 187, 197, 244, 255, 258, 264, 272, 279, 292, 294, 295 Profits, 182, 300, 301, 311, 313 Program, 8, 39, 90, 213, 216, 217, 221–225, 232, 234, 259, 276, 277, 312, 325 Protection, 89, 90, 217, 222, 234, 246, 253, 323, 328, 333, 346 Provinces (by name), 156, 157, 164, 177, 239, 244–248, 252, 254, 257, 308, 323 Public good, 85, 92, 96 Purchasing power parity, 27 Regional, 3–11, 13–15, 18–20, 23, 40, 41, 48, 49, 54, 55, 59, 65, 66, 69–72, 78, 81, 82, 103–105, 108, 155, 156, 172, 175–180, 189, 190, 196, 201–203, Regionalization, 4, 65, 70–77, 79–82, Regression, 39, 62, 65, 140, 199 Regulations, 14, 43, 73, 195, 231, 264, 267, 272, 278, 288, 307, Relative prices, 27, 120, 121, 126, 129, 130, 131, Retail, 14, 98, 99, 101, 102, 104–106, 109, 117–119, 129–132, 302, 310–312
358 Retailing Industry, 97, 98, 102, 105, 107, 109–111, Retained earnings, 49, 50, 51, 54, 55, 60, 65 Sales, 14, 15, 49, 50, 69–72, 74–78, 80, 82, 83, 99–102, 104–111, 131, 200, 222, 223, 247, 250, 258, 290, 295, 300, 302, 306, 307, 309, 310, 312 Sectors, 6, 12, 13, 15, 30, 31, 35, 36, 41, 51, 57, 78, 138, 188, 214, 228, 271, 272, 277, 288, 300, 306, 320, 329, 330 Security, 41, 66, 85–95, 216, 217, 263, 264, 275–277, 280, 281, 284, 286, 296, 301, 308 Services, 4, 6, 8, 11–13, 15, 19, 22, 26, 27, 29, 41, 42, 50, 51, 53–57, 66, 71, 83, 118, 119, 129, 138, 185, 186, 194, 249, 264, 265, 268, 269, 275, 279, 281, 301, 306, 312, 320, 321, 331, 343 Similarity index, 69–73, 76–78, 80–82, Softwood lumber, 213, 237–241, 244–246, 248, 249, 251–253, 255, 258, 259, 260, 325, 343, Spreads, 141 Standards, 32, 36, 37, 40, 253, 263, 264, 266, 267, 268, 274, 276, 278, 279, 280, 328, 331 States (by name), 153, 154, 157, 162, 164, 280, 308 Stock Market, 137–144, 147, 300, 301, 309, 311 Stocks, 4, 8, 9, 10, 13–15, 19, 42, 47, 51, 53, 54, 60, 61, 66 Strategy, 60, 70, 71, 94, 101–103, 105, 108, 109, 161, 187, 201, 252, 253, 267, 296, 304, 310, 323, Tariffs, 6, 35, 36, 39, 83, 107, 110, 111, 117, 138, 195–197, 224, 226, 228,
SUBJECT INDEX 229, 232, 234, 251, 258, 275, 288, 305, 335, 339, 344, 346 Tax, 10, 47–49, 55–66, 131, 182, 198, 199, 204, 238, 239, 257 Taxation, 57, 58, 66, 195 Technology, 36, 51, 98, 104, 231, 257, 274, 278, 279, 281, 311 Terrorism, 85, 88, 90–96, 276, 284, 347 Trade, Trade Agreements, 17, 20, 23, 36, 47, 48, 65, 66, 129, 233, 240, 299, 300, 321, 324, 325, 326, 329, Trade liberalization, 17, 27–30, 39, 40, 232, 323 Triad, 7–10, 14, 15, 48, 71, 82 U.S. U.S. Congress, 234, 267, 277 U.S. Federal Reserve, 98 Unions, 156, 245, 250, 252, 264, 273, 324, 328 United States, 3–5, 7–13, 15, 17–27, 32, 34–37, 41, 47–49, 51, 53, 58, 59, 66, 67, 69, 70, 79, 85–87, 91, 92, 139, 141–144, 146, 172, 176–180, 182, 183, 185, 186, 188, 189, 193–197, 199–204, 206, 213–215, 217, 218, 222, 228–235, 265, 267, 273, 280, 318, 324, 325, 327, 332, 339, 346, 347 Variability, 17, 37, 82, 132, 293, 295 Wages, 19, 30, 37, 40, 155, 182, 272, 273, 278, 279, 339, 343, Wal-Mart, 14, 97–102, 105, 107–111, Western hemisphere, 76, 77, 78, 317, 322–324, 329 WTO, 239, 240, 254–256, 320, 324, 329, 335