Ageing And Employment Policies: The Arable Crop Sector (Agriculture, Trade and the Environment)

caecilia_histo_gen_a_16x23.fm Page 1 Thursday, August 25, 2005 2:52 PM "HSJDVMUVSF 5SBEFBOEUIF &OWJSPONFOU 5)&"3...

Author: Organisation for Economic Co-Operation a

28 downloads 1473 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

caecilia_histo_gen_a_16x23.fm Page 1 Thursday, August 25, 2005 2:52 PM

"HSJDVMUVSF 5SBEFBOEUIF &OWJSPONFOU 5)&"3"#-&$3014&$503

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT

caecilia_histo_gen_a_16x23.fm Page 1 Thursday, June 23, 2005 9:57 AM

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

5IJTXPSLJTQVCMJTIFEPOUIFSFTQPOTJCJMJUZPGUIF4FDSFUBSZ(FOFSBMPG UIF 0&$%5IFPQJOJPOTFYQSFTTFEBOEBSHVNFOUTFNQMPZFEIFSFJOEPOPU OFDFTTBSJMZSFGMFDUUIFPGGJDJBMWJFXTPGUIF0SHBOJTBUJPOPSPGUIFHPWFSONFOUT PGJUTNFNCFSDPVOUSJFT

"MTPBWBJMBCMFJO'SFODIVOEFSUIFUJUMF "HSJDVMUVSF ÏDIBOHFTFUFOWJSPOOFNFOU -&4&$5&63%&4(3"/%&4$6-563&4

¦0&$% /P SFQSPEVDUJPO DPQZ USBOTNJTTJPO PS USBOTMBUJPO PG UIJT QVCMJDBUJPO NBZ CF NBEF XJUIPVU XSJUUFO QFSNJTTJPO "QQMJDBUJPOTTIPVMECFTFOUUP0&$%1VCMJTIJOHSJHIUT!PFDEPSHPSCZGBY 1FSNJTTJPOUPQIPUPDPQZB QPSUJPO PG UIJT XPSL TIPVME CF BEESFTTFE UP UIF $FOUSF GSBO¡BJT EhFYQMPJUBUJPO EV ESPJU EF DPQJF SVF EFT (SBOET"VHVTUJOT 1BSJT 'SBODF DPOUBDU!DGDPQJFTDPN

Foreword

T

he objective of this study is to improve understanding of the linkages between agriculture, trade and the environment in OECD countries by examining how they relate to the arable crop sector (grains, rice and oilseeds). Three of the main issues involved are: the environmental impacts of agricultural support measures and the consequences of further trade liberalisation; the trade impacts of policy measures to address environmental issues in agriculture; and the characteristics of policies that can best achieve environmental objectives in ways that are compatible with multilateral trade and environmental agreements. This study continues the analysis of agriculture, trade and environment linkages by the OECD Joint Working Party on Agriculture and the Environment. It is one of three studies carried out under the Analysing Approaches towards a more Sustainable Agriculture component of the 2003-04 Programme of Work of the OECD’s Committee for Agriculture. Two earlier studies examined the pig sector and the dairy sector (OECD, 2004a; 2003f). The arable crop sector provides a good case study for an analysis of these linkages. Farming systems vary. In some cases crops are grown on extensive areas of land, while in others the land is used much more intensively, with varying levels of input use, mechanisation and monoculture, and, consequently, with different environmental effects. There is wide variation in the form and level of support, including trade measures, provided to arable crop producers across OECD countries, between different arable crops, and over time. In addition, a number of OECD countries are reforming their policies and the arable crop sector features prominently in such reforms. Moreover, arable crops farmers are affected by a plethora of agri-environmental policies. This diversity of policy experience provides a rich variety of material to be examined and compared. The study was carried out in the Policies and Environment Division of the Directorate for Food, Agriculture and Fisheries, with Dimitris Diakosavvas as the principal author. Valuable contributions were provided by consultants Alison Burrell (Chapter 5), Marino Tsigas (Chapter 6) and Dave Ervin (Chapter 7). The US Economic Research Service, and Agriculture and Agri-Food Canada undertook the quantitative analysis reported in Section 6.3. Françoise Bénicourt and Theresa Poincet provided secretarial support, while Véronique de Saint-Martin assisted with statistical work. Colleagues in the OECD Secretariat, particularly Wilfrid Legg, Hsin Huang and Joe Dewbre (GTAP analysis), and Peter Kearns from the Environment Directorate (transgenic crops section), and Delegates from member countries provided many useful comments. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

–3

This page intentionally left blank

Table of Contents FOREWORD........................................................................................................3 ACRONYMS AND ABBREVIATIONS.............................................................9 TECHNICAL TERMS .......................................................................................10 HIGHLIGHTS ....................................................................................................11 SUMMARY AND CONCLUSIONS .................................................................13 INTRODUCTION ..............................................................................................27 CHAPTER 1. ECONOMIC AND STRUCTURAL ASPECTS OF THE ARABLE CROP SECTOR ..........................................................33 1.1. The arable crop sector in OECD countries ..........................................33 1.2. Developments in farm structures .........................................................37 1.2.1. Changes in number and size of farms........................................37 1.2.2. Regional concentration ..............................................................40 1.2.3. Sources of growth in production................................................41 1.2.4. Chemical inputs .........................................................................46 ANNEX 1.A. Selected Data ...............................................................................49 CHAPTER 2. ENVIRONMENTAL IMPACTS ASSOCIATED WITH PRODUCTION ............................................................................57 2.1. Soil-related impacts .............................................................................57 2.1.1. Soil erosion ................................................................................58 2.1.2. Nutrients ....................................................................................62 2.1.3. Waterlogging and salinisation ...................................................63 2.2. Water-related impacts..........................................................................65 2.2.1. Water use ...................................................................................65 2.2.2. Water pollution ..........................................................................66 2.3. Air quality............................................................................................68 2.4. Biodiversity .........................................................................................72 2.5. Management practice approaches to reduce environmental impacts of arable crop production.....................................................................73 2.5.1. Soil management and conservation systems..............................74 2.5.2. Nutrient Management ................................................................78 2.5.3. Integrated Pest Management......................................................79 2.5.4. Organic farming practices..........................................................80 2.5.5. Factors influencing adoption of environmentally benign farming practices .......................................................................83 AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

–5

2.6. Transgenic crops and the environment ................................................84 2.6.1. How widespread are transgenic crops?......................................84 2.6.2. What are the environmental implications? ................................87 2.6.3. Environmental impact assessments ...........................................91 2.6.4. Current and future trends ...........................................................93 ANNEX 2.A. Selected Data ...............................................................................94 CHAPTER 3. AGRICULTURAL POLICIES AFFECTING THE ARABLE CROP SECTOR ..........................................................99 3.1. Introduction .........................................................................................99 3.2. Main policy instruments ......................................................................99 3.3. Levels of support ...............................................................................104 3.4. Composition of support policies........................................................110 3.5. Developments in market price support ..............................................111 3.6. Developments in domestic support policies ......................................112 3.6.1 Payments based on output.......................................................112 3.6.2. Payments based on area planted .............................................113 3.6.3 Counter-cyclical payments in the United States .....................115 3.6.4. Payments based on historical entitlements ..............................116 3.6.5. Payments based on input use ...................................................118 3.6.6. Payments based on input constraints .......................................119 3.6.7. Payments based on overall farm income .................................120 3.7. International trade measures ..............................................................120 3.7.1. Import measures ......................................................................120 3.7.2. Export measures ......................................................................123 3.8. Summary of agricultural policy reform in the arable crop sector ......124 ANNEX 3.A. Selected Data .............................................................................126 CHAPTER 4. POLICY MEASURES ADDRESSING ENVIRONMENTAL ISSUES IN THE ARABLE CROP SECTOR ............................143 4.1. Introduction .......................................................................................143 4.2. Economic instruments .......................................................................143 4.2.1. Payments based on farm fixed assets (excluding land retirement)......................................................144 4.2.2. Payments based on resource retirement...................................145 4.2.3. Payments based on farming practices......................................149 4.2.4. Environmental taxes ................................................................154 4.2.5. Tradeable rights/quotas............................................................155 4.3. Regulatory measures..........................................................................156 4.3.1. Regulations ..............................................................................156 4.3.2. Cross-compliance mechanisms................................................161

6 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

4.4. Advisory and institutional measures..................................................166 4.4.1. Research and development ......................................................166 4.4.2. Technical assistance and extension..........................................167 4.4.3. Product information .................................................................170 ANNEX 4.A. Selected Data .............................................................................172 CHAPTER 5. ENVIRONMENTAL EFFECTS OF AGRICULTURAL SUPPORT POLICIES FOR ARABLE CROPS.........................175 5.1. Introduction .......................................................................................175 5.2. Environmental effects of agricultural support policies......................176 5.2.1. Links between high arable support and negative environmental effects...............................................................176 5.2.2. Assessing the environmental effects of lower support.............180 5.2.3. Environmental effects of shifting from price support to direct payments .......................................................................187 5.3. Cross compliance...............................................................................191 5.3.1. Background..............................................................................191 5.3.2. Advantages and disadvantages of red ticket environmental cross compliance......................................................................193 5.3.3. Design of cross-compliance provisions ...................................195 5.3.4. Various options for linking income transfers and environmental objectives ........................................................199 5.4. Efficiency and cost effectiveness of cross compliance and alternatives ..................................................................................202 5.4.1. Efficiency and cost effectiveness of various programmes.......202 5.4.2. Participation, monitoring and non-compliance........................214 5.5. Assessment and conclusions..............................................................215 CHAPTER 6. ENVIRONMENTAL IMPACTS OF MULTILATERAL AGRICULTURAL TRADE LIBERALISATION ON ARABLE CROPS................................................................225 6.1. Introduction .......................................................................................225 6.2. Cross-country analysis.......................................................................228 6.2.1. The liberalisation scenarios .....................................................228 6.2.2. Methodology............................................................................229 6.2.3. Simulated environmental impacts of multilateral agricultural trade liberalisation ...................................................................230 6.2.4. Sensitivity analysis ..................................................................234 6.2.5. Caveats.....................................................................................234 6.3. Regional environmental impacts of agricultural trade liberalisation ...........235 6.3.1. Canada .....................................................................................235 6.3.2. United States...........................................................................240 ANNEX 6.A. The Applied General Equilibrium Trade Framework ................244 AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

–7

ANNEX 6.B. Regional Models ........................................................................251 6.B.1. 6.B.2.

The US Regional Agricultural Programming Model (USMP) 251 The Canadian Regional Agricultural Model (CRAM) ............255

ANNEX 6.C. Selected Data..............................................................................260 CHAPTER 7. AN ANALYSIS OF THE TRADE EFFECTS OF AGRIENVIRONMENTAL PAYMENTS AND REGULATIONS ON ARABLE CROPS................................................................263 7.1. Introduction .......................................................................................263 7.2. Overview of agri-environmental policies for arable crop agriculture264 7.2.1. Payment programmes ..............................................................264 7.2.2. Regulatory approaches.............................................................265 7.2.3. Other measures ........................................................................266 7.3. Agri-environmental programmes and trade: theory and models .......267 7.3.1. Welfare theory .........................................................................268 7.4. Effects of agri-environmental programme payments on trade...........270 7.4.1. Trade and agricultural policy context ......................................270 7.4.2. Previous analyses.....................................................................272 7.4.3. Simulating potential trade effects of agri-environmental payments..................................................................................274 7.5. Effects of agri-environmental regulations on factor costs and trade .275 7.5.1. Previous analyses.....................................................................276 7.5.2. Simulating potential trade effects of agri-environmental regulations ...............................................................................280 7.6. Suggestions for enhancing the effectiveness of agri-environmental policies on arable crops .....................................................................284 7.6.1. Reactive or proactive policy approach?...................................285 7.6.2. Some lessons from analysis and experience ............................286 ANNEX 7.A. Equations Used to Estimate the Trade Effects of Agri-environmental Programmes...............................................291 7.A.1. 7.A.2. 7.A.3. 7.A.4.

Small country import impact of agri-regulation on factor that increases the factor price (marginal cost) .........................293 Large country imports..............................................................294 Small country trade impact of agri-environmental regulation that increases average variable cost .........................................294 Product regulation case............................................................295

BIBLIOGRAPHY.............................................................................................299

8 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

ACRONYMS AND ABBREVIATIONS AP 2002 ARP AAFC AAPS ABARE AUDIT CAP CRP CSP CCP DEFRA DP EFTA ERS ENS ESAS EQIP EC EEA EU FSRI FAIR FAO IEEP LEI LDP MLAP NRI NAFTA PFCP PROCAMPO RFISP REPS

Federal Agricultural Law 2002 (Politique agricole 2002), Switzerland Acreage Reduction Program, United States Agriculture and Agri-Food Canada Arable Area Payments Scheme, EU Australian Bureau of Agricultural and Resource Economics National Land and Water Resources Audit, Australia Common Agricultural Policy, European Union Conservation Reserve Program, United States Conservation Security Program, United States Counter-cyclical Payments, United States Department of Environment, Food and Rural Affairs, United Kingdom Direct Payments, United States European Free Trade Association (Iceland, Liechtenstein, Norway, Switzerland) Economic Research Service of the USDA Environmental News Service Environmentally Sensitive Areas Scheme, United Kingdom Environmental Quality Incentives Program, United States European Commission European Environment Agency European Union Farm Security and Rural Investment Act, United States Federal Agricultural Improvement and Reform Act, United States Food and Agriculture Organization of the United Nations Institute for European Environmental Policy, London Agriculture Economics Research Institute (Landbouw Economisch Instituut), the Netherlands Loan Deficiency Payments, United States Market Loss Assistance Payments, United States National Resources Inventory, United States North American Free Trade Agreement Production Flexibility Contract Payments, United States Direct support for the countryside (Programa de Apoyos Directos al Campo), Mexico Rice Farming Income Stabilisation Programme, Japan Rural Environment Protection Scheme, Ireland

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

–9

SAPARD

UNFCCC USDA USITC URAA WRP WHO WTO

Special Accession Programme for Agriculture and Rural Development (European Union, Czech Republic, Hungary, Poland, Slovak Republic) United Nations Framework Convention on Climate Change United States Department of Agriculture United States International Trade Commission Uruguay Round Agreement on Agriculture Wetland Reserve Program, United States World Health Organization World Trade Organization

TECHNICAL TERMS AEI CRAM ESA ESU GMO GTAP GFP GHG HEL LFA LMO NPC PSE SFP TRQ USMP UAA

Agri-Environmental Indicators Canadian Regional Agricultural Model Environmentally Sensitive Areas European Standard Unit, EU Genetically Modified Organisms Global Trade Analysis Project Good Farming Practices Greenhouse Gas Highly Erodible Land Less Favoured Areas, EU Living Modified Organisms Nominal Protection Coefficient Producer Support Estimate Single Farm Payment, EU Tariff Rate Quotas US Regional Agricultural Programming Model Utilised Agricultural Area

10 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

HIGHLIGHTS

x

Policy concerns over the environmental effects of arable crop farming (grains, rice and oilseeds) have increased over the last two decades, due largely to more intensive use of land; and a rise in the value placed on many environmental services provided by agriculture.

x

The main environmental issues associated with the production of arable crops include: soil (erosion, nutrients, waterlogging and salinisation); water (use and pollution); air quality (greenhouse gas emissions); landscape and biodiversity (such as pasture conversion to cropland or land abandonment).

x

Environmental impacts vary across countries and regions, depending on the specific resource base and on prevailing farming practices and policies. Significant problems occur in many regions, but their scope and severity vary and tend to be greater where production pressure is concentrated and natural resources are vulnerable.

x

Agri-environmental indicators related to arable crops present a mixed picture of improvements and degradation in different countries. Soil erosion in the United States has decreased, while water-use issues continue to be a cause of serious concern in parts of Australia, the United States and some EU countries. Lack of crop biodiversity is a concern in certain countries, although some have diversified and produce a greater variety of crops. Arable crop farming is a less important cause of air pollution than livestock production, accounting for only 6% of greenhouse gas emissions from agriculture.

x

A plethora of policy approaches has been adopted, reflecting the diversity of agri-environmental conditions in OECD countries. Most agri-environmental measures are not targeted at a particular arable crop sector or at a specific environmental outcome, but focus mostly on controlling the quality and quantity of production inputs, as exemplified by temporary or permanent land retirement.

x

Payments based on: (i) farm fixed assets; (ii) resource retirement; and (iii) farming practices currently have the largest potential to influence production and trade, based on the level of support afforded to the arable crop sectors, although in certain cases some regulations also exert significant effects.

x

Support for arable crops is high relative to other agricultural sectors, varies greatly between countries and crops, and is mainly provided through policy instruments that are the most production and trade distorting.

x

Although the cause-effect linkages between support levels and environmental pressures are complex, correlation does not necessarily imply causation.

x

At the aggregate country level, the environmental effects of further multilateral agricultural trade liberalisation are likely to be small. Only under the full trade liberalisation scenario would chemical intensity in certain arable crop sectors in Australia and New Zealand increase by more than 10%.

x

The production and trade effects of overall support, agri-environmental payments and regulations warrant further empirical analysis.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 11

This page intentionally left blank

SUMMARY AND CONCLUSIONS Trade and environment issues in agriculture have gained increasing prominence at international and national levels alike. The present report analyses the linkages between agriculture, trade and environment in OECD countries for the arable crop sector (grains, rice and oilseeds). The report first provides some background material on economic and environmental aspects associated with arable crop farming and discusses the policies – both agricultural support and environmental policies – affecting the arable crop sector. It then analyses some of the cause-effect linkages between policies, including trade policies and regulations, and the environment. What are the main environmental impacts associated with arable crop farming? Environmental impacts on soil, water, air quality, biodiversity and landscape are important …

Arable cultivation systems are among the most important factors influencing soil quality. While chemical inputs, such as fertilisers, herbicides and other pesticides, make a major contribution to arable crop productivity, they also create environmental problems in some regions across OECD countries. Nitrates and phosphates are the main nutrient pollutants of water courses resulting from arable farming. Increased monocultures and the reduction of mixed arable and livestock farms also have an adverse effect on biodiversity. Although recent empirical studies in some countries suggest that farm productivity losses due to erosion are relatively small, soil erosion is a widespread problem in several OECD countries. Arable crop production is an important user of water, particularly in the case of rice, which can increase diversity of habitats and farmland species. In parts of Australia, the United States and in some EU countries water-use issues are of particular concern. Nitrous oxide, carbon dioxide and methane are the main greenhouse gases arising from arable crop farming. Nitrous oxide originates from fertilised agricultural soil, while methane emissions are caused primarily by wetland rice cultivation. The impacts on biodiversity of the ecosystem are diverse. Although in some countries lack of crop

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 13

… and crop-mix, farm practices and structures, and government policies are the main driving forces.

Adoption of transgenic crops has been rapid …

… and their cultivation could have positive environmental effects …

biodiversity is a matter of concern, agri-environmental indicators related to arable crops suggest that the number of new crop varieties has been increased in several countries. The environmental impacts of arable crop production vary across OECD countries for at least three reasons. First, they depend on the quality and quantity of natural resources used in, or affected by, arable crop production. For example, growing wheat in a semi-arid region may cause wind-induced soil erosion and particles in the air. In a country that relies heavily on irrigation, however, the primary effects are likely to concern water use and quality. Second, the impacts vary according to the technologies employed to produce crops. Reduced tillage systems, for example, decrease erosion and greenhouse gas emissions, but may require an increased use of pesticides, which can cause degradation in certain situations. Third, the impacts will depend upon the country’s relative demands for different types of environmental quality. If the demand and willingness to pay for a particular environmental outcome are high (e.g. mixed use landscape), then measures may be needed to ensure its provision. Regional concentration and increased specialisation of arable crop production due to economies of scale have in some regions encouraged monocultures and reduction in the number of mixed farms, with important implications for land use, landscape and biodiversity. Since their commercialisation in the mid-1990s, the area grown with transgenic crops worldwide has witnessed a remarkable increase. Seven countries (the United States, Argentina, Canada, Brazil, China, India and South Africa), four crops (soybeans, cotton, maize and rapeseed) and two traits (resistant to certain insects and tolerant of certain herbicides) account for almost the totality of the global transgenic area. Almost two-thirds of the area under transgenic crops is in the United States. Cultivation of transgenic crops could have positive environmental effects, depending on the crop and trait under consideration. These include gains in using environmentally benign methods of managing certain weeds and pests, and reducing the need for chemical inputs.

14 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

… but there are environmental concerns as well.

All OECD countries have a system of regulatory oversight in place for assessing the environmental safety of transgenic crops, taking into account the risks associated with the transgenic behaviour of the same crop species and other related crops. Introducing insect-resistant substances could encourage insects to develop resistance. Transgenic crops may increase or decrease crop genetic diversity and the diversity associated with other crops, depending on: the diversity of transgenic crops; the diversity of the transgenic crops into which the gene is inserted; and on the production methods used.

Adoption of environmentally benign production practices is slow.

Various production practices have been developed over the past fifteen years to minimise the environmental effects of agricultural production. Among the foremost of those concerning arable crop farming are Soil Management and Conservation Systems, such as rotational cropping systems and tillage practices, Integrated Plant Nutrient Systems and Integrated Pest Management practices. These practices and technologies are interrelated and complementary, seeking to attain the dual goals of increased productivity and reduced environmental impact. Yet experience suggests that, despite their higher rate of returns, wide-scale adoption has not yet occurred across OECD countries. Farmers operating larger arable crop holdings usually adopt new technologies more rapidly than farmers operating small farms, and have a larger incentive and capacity to deal with environmental problems. Evidence from some countries suggests that the higher productivity per hectare has not been accompanied by a corresponding increase in environmental damage. There are several barriers hindering the adoption and diffusion of environmentally benign production practices, including a lack of knowledge of ecological systems and of the way in which agricultural practices impact on them, and structural factors such as the level of management skill required in order to use these systems appropriately.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 15

What are the key economic and structural characteristics of arable crop farming? The arable crop sector uses significant amounts of land and chemical inputs …

The arable crop sector occupies approximately one-third of the OECD’s agricultural area, and contributes to around one-half of the OECD’s total agricultural output. OECD countries are responsible for approximately 80% of global trade in cereals. For the arable crops analysed in this report, maize and rice are the most intensively fertilised in terms of applications per hectare, while sunflower is the least fertiliser-dependent arable crop. Maize and rice are also high users of water. The most frequently applied nutrient in arable crop farming is nitrogen. In the majority of OECD countries, most of the nitrogen consumed by the agricultural sector as a whole is used by arable crops.

… is becoming larger in size, more capitalintensive and specialised …

Available data underscore the great diversity in area allocated to arable crop per holding across OECD countries, as well as the differences in rates of change over time. Average farm size in 2000 ranged from 0.8 hectares per farm in Japan to 1 654 hectares per farm in Australia, with the EU15 average around at 14 hectares per farm. Arable crop production is also characterised by regional concentration within countries, reflecting local resource endowment, climate, soil types and policies. The number of arable farms and the area used for arable crop farming have declined over the last two decades in OECD countries as a whole, but average farm size has increased, as the number of farms has fallen by more than farmland. In several OECD member countries, however, the number of larger, more capital-intensive and specialised arable crop farms has increased in absolute terms.

… with higher yields and greater cropping intensity largely accounting for greater production.

Notwithstanding the diversity between countries, arable crop production in OECD countries increased, on average, by 0.5% per annum over the 1985-2002 period. Overall, most of this growth was derived from an increasingly intensive use of land already under crops rather than expansion of the harvested area, although the latter was the main source in some countries.

16 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

The contribution of increases in harvested area to production growth can be broken down into the effects stemming from increases in arable land expansion and to effects arising from increases in cropping intensities, such as multiple cropping and shorter fallow periods. In most of the cases where harvested area increased, the contribution of increases in cropping intensity was more important than the contribution of arable land expansion in explaining expansion of harvested area. How extensive is the agricultural support affecting arable crops? Despite some progress in policy reform, support remains high, with associated production and trade distortions …

Support to arable crop producers in OECD countries amounted to USD 62 billion in 2001-03, accounting for 39% of farm receipts from crops. Reflecting overall trends, the average support levels decreased over time for all arable crops, except rice, for which support levels have changed little since 1986-88. The rice sector is the most-supported arable crop and oilseeds the least-supported. Although rice is produced in relatively few OECD countries, the price gap between domestic and world prices is the highest for any commodity in the OECD area. In 2001-03, prices received by rice producers and paid by consumers were, on average, more than four times higher than world rice prices.

… but support varies considerably across countries and arable crops.

Overall, the level of support for the arable crop sector, as measured by the share of support in gross farm receipts (%PSE) are highest – over 70% – in Japan, Korea, Norway and Switzerland. However, arable crop producers in Australia, the Czech Republic, New Zealand and the Slovak Republic are the least supported, with less than 10%. In Canada support ranges between 10-20%, in the EU between 30-40%, while for the United States arable crops support is around 30%. Over the 2001-03 period, the majority of support was concentrated in a few countries. It is not only the level of support, but also the form in which it is provided, that is important in terms of the impacts on resource allocation and on the environment. Many governments utilise a complex array of

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 17

measures –including tariff rate quotas and preferential trade agreements – that directly or indirectly affect production, consumption, trade, prices and the environment. For arable crops as a whole in the OECD area, market price support and output-related support – which are the forms of support with the greatest potential to stimulate production, exacerbate environmental pressures and distort trade – accounted for about half of the support to the sector in the 2001-03 period. There has been a significant shift from market price support to area payments.

For the OECD area, on average, payments based on area planted constitute, to an increasing extent, the main form of support provided to grain producers, while for rice producers market price support (tariffs and export subsidies) dominates. Oilseed producers are mainly supported through payments based on output. Area payments are particularly important in the EU, where they represented almost three-quarters of producer support in 2001-03 and, following the 2003 CAP reform, are now less linked to production. Set-aside has been an important element in policies for supply control and, increasingly, for environmental reasons. Land retirement programmes are currently being pursued in the EU, Japan and the United States. In the EU, the two long-term land diversion schemes introduced as part of the 1992 CAP reform are specifically aimed at achieving environmental objectives. The payment rates of several of these land diversion schemes are intended to compensate farmers for the cost increases and/or revenue losses associated with abandoning conventional production on part of their land. In Japan, environmental provisions have gradually been incorporated into programmes aiming to divert land from rice production to other crops and activities. In the United States, the 2002 FSRI Act maintains and extends the programmes that retire environmentally sensitive land from crop production.

18 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

What are the effects on the environment of agricultural support policies for arable crops? Price support maintained over time is one of the most important factors causing the intensification of production and resulting environmental harm …

Price support and input subsidies both provide incentives for output expansion and intensification of input use, as they stimulate farmers to change their management practices and rates of input use. Commodity-linked support will also alter the mix of crops grown, which may not be neutral for the environment. If higher levels of support are given to high-performance crops that are more input-intensive, then the impacts on input use and crop mix will be even greater. Further, when high levels of support are maintained over time, this may impede structural change in the sector and may stimulate the development of new yield-enhancing and cost-reducing technologies, which could be biased in favour of those crops receiving the highest support, and which may result in variable environmental outcomes. At the same time, capitalisation of support into land prices may enhance the underlying pressures for farm consolidation and production intensification. However, the link between production changes and environmental outcomes is sitespecific.

… but environmental effects of shifting to payments will depend on the degree to which they are decoupled from production and targeted to specific goals.

The reduction in price support is likely to lead to a reduction in output and variable input use. Replacing market price support with more decoupled payments at fixed rates is expected to decrease input use and encourage a reduction of cropped area, thus producing potentially beneficial environmental effects. When payments are coupled to current area, producers are more likely to maintain or, if possible, increase their cropped area in order to qualify for payment entitlements, whilst still reducing input intensity. The evidence available on the environmental impact of the shift away from arable price support to budgetary payments that followed the 1992 CAP reform suggests that some land was released from cereal and oilseed production due to set-aside. Production intensity for these crops declined, although the extensification effect was less pronounced than expected. The total environmental impact of these changes varies depending on, among other things, on how the released land was used and on how arable crop producers adjusted their variable inputs.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 19

How large are the impacts of further agricultural trade liberalisation on the environment? Environmental impacts of trade liberalisation at the country level are small …

This study analyses the potential environmental impacts of further agricultural trade liberalisation using a multi-country, global trade model and indicators of environmental quality, such as changes in the scale and intensity of input use, pesticide use, nitrogen uptake and off load, and emissions of greenhouse gases from crop production. The analysis provides an illustration of the potential implications for the environment of multilateral agricultural trade liberalisation. It does not consider the counterbalancing influence of existing environmental policies and regulations. Two hypothetical multilateral agricultural trade liberalisation scenarios are considered. The first scenario assumes an extension of the WTO Uruguay Round Agreement on Agriculture. The second scenario involves the elimination of all agricultural policy measures in all countries. The latter scenario can be viewed as an upper bound of potential outcomes of multilateral agricultural trade liberalisation. In most cases, the simulated liberalisation impacts for the aggregate arable crop sector do not suggest significant environmental implications: the percentage changes in land and chemical use, aggregate output, and the rate of chemical application are small. This conclusion applies to both the partial and full agricultural trade liberalisation scenarios. Under the partial liberalisation scenario, the impacts on production and chemical intensity are less than 10% in all countries and country groupings. Only under the full trade liberalisation scenario are the impacts in some cases higher. Under the full trade liberalisation scenario, environmental pressures associated with the degree of intensity of arable crop farming are likely to increase in New Zealand, and to a much lesser extent, in Australia and Canada. Intensity of chemical use would decline the most in Korea, Norway and Switzerland. In the EU15, Japan and the United States, the simulated impacts of full trade liberalisation suggest that arable crop output and use of chemicals would decline. For the new EU10 members, output of arable crops, land and chemical use would increase at about the same rate.

20 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Overall, the estimated changes in arable crop production, even in the extreme scenario of full agricultural trade liberalisation, are within the bounds of average seasonal variations witnessed over the last twenty years in the OECD area. The simulations also suggest that trade liberalisation would cause global methane and nitrous oxide emissions to decline. … but there may be important regional differences.

The cross-country quantitative analysis is supplemented with some country-specific disaggregated analysis. The results from the global model are used as inputs into spatial, regional and environmental models to assess the environmental impacts of trade liberalisation at the regional level for Canada and the United States. The results suggest that the estimated changes in crop production and subsequent environmental impacts are not uniform across the regions in each country, with increases in crop production and environmental quality in some regions and decreases in others.

What are the main policies addressing environmental issues in the arable crop sector? Environmental policies mainly include non-commodityspecific payments and regulations.

The diversity of programmes across OECD countries and regions is vast. A plethora of measures affects arable crop farmers, encompassing economic instruments, direct regulation, technical assistance and conservation, research and extension. Notable trends in payment measures include the growing use of land retirement payments to promote environmental objectives; payments to support the adoption of less-intensive farming practices, such as organic farming; and transitional payments based on farm fixed assets, such as assistance for water, soil and land conservation.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 21

While European countries and the United States rely heavily on the use of payments to address environmental issues, other countries, such as Australia and New Zealand, place greater emphasis on community-based approaches. The scope of regulatory policy measures has generally expanded in OECD countries over the past two decades. These measures range from broad prohibitions to very prescriptive details for the adoption of environmentally benign farm management practices. Most regulations are implemented at the local level, and environmental legislative responsibilities usually rest with sub-national level governments. Regulations to protect groundwater quality and control soil erosion are often used, with the most severe restrictions applying to pesticide use. Taxes and charges, and tradeable quotas and rights are seldom used…

In the few OECD countries where a tax on pesticides or fertilisers is imposed, the tax is relatively small. Tradeable rights are used in only a few countries at regional/local level for water extraction for irrigation. The arable crop sector is also affected by various eco-labelling schemes, particularly those dealing with organic production and other non-crop-specific measures on research, technical assistance and extension.

… but crosscompliance is becoming important in several OECD countries.

While in the United States, cross-compliance has been used as a mechanism for seeking to control soil erosion, the ploughing of fragile rangeland and the drainage of wetlands, in European countries some of these objectives are being pursued primarily through regulations. All direct payments affecting arable crops in Switzerland, area payments in Norway and area payments for paddy field farmers in Korea are subject to cross-compliance. In the EU, following the 2003 CAP reform, cross-compliance became compulsory and the single farm payment to farmers will be linked, inter alia, to the respect of environmental, food safety, animal welfare and plant health standards, as well as the requirement to keep all farmland in good agricultural and environmental condition.

22 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

What are the production and trade effects of agri-environmental payments and regulations on arable crops? Evidence on the effects of agrienvironmental payments and regulations is limited …

The recent growth in agri-environmental regulatory and payment programmes raises concerns about the possible negative effects on trade, including arable crop imports and exports. Correcting for missing markets for environmental externalities, or reducing government policy distortions improves social welfare, despite having trade impacts. However, if agri-environmental programmes are not implemented in cost-effective ways, there is a risk that national and global welfare will be lower. Effects of regulations on factor costs and trade depend on the particular regulatory, country and crop conditions. The regulations may cover erosion, fertiliser and pesticide use, as well as land maintenance requirements under compliance schemes. Research generally has not shown that environmental regulations have significant impacts on trade competitiveness and firm location. However, the vast majority of the research did not cover agriculture and did not investigate specific product markets. Recent developments in modelling provide an empirical approach to estimate the potential impacts of environmental regulations on a country’s crop production and trade. Simulation analyses of arable crop and country combinations were conducted to gauge the sensitivity of production and trade to different forms and intensities of regulation. The analyses suggest that the trade effects can be significant, 10% or more, depending upon the particular regulatory, country and crop conditions. As an illustration, Australian wheat exports were estimated to decrease by approximately 18% if the price of a pollutive input is increased by 200% in order to discourage its use. The simulation analyses do not capture potential offsetting effects if other countries adopt similar regulatory programmes and standards in their arable crop sectors. Despite the rapid growth of agri-environmental payments, there have been few similar advances in modelling their production and trade impacts. Depending upon the programme objectives and the ways in which they are implemented, the payments may be designed to maintain particular types of land use that provide the desired environmental services, with differing effects on production. Two recent studies estimated the impacts of such payments on production and trade and arrived at different conclusions that may reflect the manner in which

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 23

other agricultural payments reinforce or offset the effects of payments. A simulation analysis suggests that agrienvironmental payments could have modest effects on production, and larger impacts on trade flows, in certain country-crop situations. … and crosscompliance measures are not sufficient to reconcile the inconsistencies between support and environmental policies.

Cross-compliance measures can improve environmental performance and lead to better harmonisation of agricultural and environmental policies. But when cross-compliance requirements are attached to direct income payments with the aim of achieving income support and environmental objectives, they are not necessarily the most cost-effective measures. If the income support payments are high enough and the cross-compliance conditions are sufficiently modest, all producers will find that the programme improves their income. Either compliance cost in income support payments will be very high or the environmental benefits will be small, or both. On the other hand, if the cross-compliance conditions are set to aim for a significant impact on environmental targets, then some producers will either suffer an income loss (if remaining in the scheme is compulsory), or these producers will leave the programme (if participation is voluntary). Gross environmental benefits will be lower, and net environmental benefit could also be lower.

Some policy conclusions Further agricultural policy reform and trade liberalisation should reduce environmental pressure in countries with high support and environmental pressure …

A comprehensive analysis of the linkages between trade and environment requires a thorough knowledge and understanding of both agricultural support policies, including trade policies, and environmental policies with a bearing on agricultural production. The levels of support, and the ways in which support is delivered, together with the dispersion of support and protection across commodities, are important causes of distortion in resource allocation between commodity sectors, input use and distortion in environmental outcomes. Production-linked agricultural support for crops has hindered the adoption of environmentally benign farming systems. Decoupling of agricultural support from production decisions, provision of information and investments in human capital would facilitate the adoption and diffusion of such systems.

24 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

… but reductions in price support alone are unlikely to redress the environmental harm caused by decades of such support, unless accompanied by targeted agrienvironmental policies…

With lower levels of support, producers are unlikely to retrace the pathway already taken when adjusting to high support. The shift from high price support is likely to be characterised by new technological choices. There are also asymmetric or irreversible processes involved and it may take years of reduced environmental stress to produce any discernible improvement in environmental conditions. In this case, complementary policies that provide explicit signals and appropriate incentives through targeted agri-environmental policies could promote sustainable production methods.

… and improving their costeffectiveness is necessary.

The theory of the economics of trade and environment shows that so long as “optimal” environmental policies are in place, open trade is nationally and globally superior to no trade. However, while allocation of resources that achieves production and trade efficiency may not be optimal if environmental effects are not considered, it is a challenge to achieve in practice. A crucial consideration in assessing the cost effectiveness of an agri-environmental programme, taking into account its production and trade impacts, is whether, or to what extent, crop production and the environmental services are joint outputs and, therefore, whether the agri-environmental policies can or cannot be decoupled from production. Even if the environmental services and production are joint, measures to improve the cost effectiveness of the programme will lessen potential trade impacts. There is a need for a coherent institutional framework in order to rationalise local and regional environmentally inspired initiatives. The level of government involvement that is appropriate, i.e. local, state/provincial, national or international, is the one that is the most cost-effective and which involves the lowest transaction costs for the particular environmental problem concerned.

Payments should be targeted to reflect different compliance costs and environmental benefits.

Cross compliance attached to direct payments can achieve some objectives at low incremental cost, but the income support and environmental objectives are sometimes in conflict. A crucial limitation of cross compliance is that those farmers who receive payments with cross-compliance conditions are not necessarily

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 25

those farming the most environmentally sensitive land or highly valued landscape. Improved environmental outcomes at lower cost could be achieved through targeted environmental measures such as taxes and regulations to deter the use of specific damaging inputs, and payments to foster certain environmental services. An impediment to discovering and implementing more cost-effective programmes is improved environmental research related to agriculture. Past research on environmental issues in agriculture has largely been reactive to a problem and sometimes in approach. Proactive research that integrates biophysical and socioeconomic sciences into a systems view is needed to develop programmes that achieve more environmental benefits and avoid emergent environmental damage from arable crop production.

26 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Introduction Over the last two decades, trade and environment issues in agriculture have gained increasing prominence at international and national levels alike. At the global level, this interest is principally expressed in the on-going WTO negotiations and the UN World Summit on Sustainable Development, whilst at the micro level, local government and agencies are concerned about the impacts of policies on production and trade, as well as on the local environment. In the context of OECD’s work on Analysing Approaches towards a more Sustainable Agriculture, further analysis exploring the linkages between agriculture, trade and environment has been undertaken through in-depth sectoral studies. Studies on the pig sector and on the dairy sector have already been completed (OECD, 2004a; 2003f). The present study analyses the impacts of these linkages on the arable crop sector. There are a number of reasons for undertaking this study:

x

There is great variation in the levels and types of support, including trade measures, provided to arable crops among OECD countries; between arable crop sector; and over time. In addition, a number of OECD countries are reviewing their policies and implementing new ones and the arable crop sector features prominently in such reforms.

x

A wide disparity exists between policy approaches taken and measures introduced across OECD countries to address the environmental impacts of arable crop production.

x

As is the case in other agricultural sectors, the arable crop sector is witnessing significant structural and technological changes. Technological advances, such as the introduction of new seed varieties, pesticides and larger-scale machinery, have enabled the spread of arable agriculture onto environmentally fragile land in certain regions, but not without entailing some negative consequences for the environment, such as the destruction of semi-natural habitats and increased risks of contamination, resulting from the use of pesticides and fertilisers. At the same time, some arable areas have been abandoned. On the other hand, Global Positioning Systems and the development of precision farming methods are helping to improve the efficient use of inputs such as fertilisers, pesticides and seeds.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 27

x

A study by the FAO projects that by 2030 an additional one billion tonnes of cereals will be needed each year to satisfy expected growth in global demand (FAO, 2003). Although, according to the study, there is enough unused potential to meet this demand, in terms of land, water and yield improvements, the expansion of cereal supply, particularly in some developing countries, could entail environmental risks such as deforestation, desertification, resource degradation of cultivated lands and loss of biodiversity.1

x

Production systems for arable crops range from the relatively extensive to the highly intensive, with varying levels of input use, mechanisation, and monoculture and, consequently, have varying environmental effects. On the one hand, a more concentrated, modern and highly intensive farming system may result in the loss of non-crop habitats, such as grassland, and so have negative environmental implications in terms of the elimination of biodiversity within the immediate production area. On the other hand, lower-yield, extensive farming systems may require more land at the expense of natural areas. While drawing general conclusions is difficult and speculative, extensive and intensive methods of production are associated with different types of externalities.

Obviously, it is not possible for the analysis to be comprehensive in all domains because of the complexity of the issues involved and the inadequacy of some data, particularly on agri-environmental indicators (AEI). Moreover, arable crops include a wide range of annual crops (e.g. wheat, barley, maize, rye, rapeseed, sunflower, peas, etc.). The following guiding criteria and considerations were considered when selecting which specific arable crops would be most suitable for the analysis:

x

The importance of the sector in production, trade and in terms of domestic and trade-related policies;

x

The relevance of the sector in terms of environmental impacts;

x

The potential for further trade liberalisation in the sector; and

x

The availability of relevant data and quantitative models.

For the purpose of this report, analysis of the arable crop sectors in relation to the aforementioned criteria has been limited to the case of cereals and oilseeds. Sugar is not part of the study. In particular, the focus of the analysis is on grains, rice, soybeans, rapeseed and sunflower. Table 1 provides a summary of the varying significance of the economic, environmental and policy relevance of the sectors.

28 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 1. Relative qualitative characteristics of different arable crops Economic and Trade p %PSE Tariffs and TRQs Export subsidies

Maize

Rice

Rapeseed

Soybeans

Sunflower

*** *** ***

** ** ***

*** *** *

** * *

** * *

** * *

US, Brazil, Argentina

Russia, Argentina, US, Hungary

US, Canada, US, China, Australia, EU, Brazil, Argentina France, Argentina, Mexico

Main exporters

Brazil, Egypt, EU, Japan, Japan, Korea, Indonesia, EU, Russia Korea

Main importers

Environmental Issue

Wheat

Thailand, Canada, Vietnam, Australia, EU China, US, India, Pakistan, Uruguay, EU Indonesia, EU, Brazil, Iran, Japan

China, Japan, EU, Mexico

EU, China, Japan, Mexico

EU, Turkey, Morocco

1

Soil (Erosion)

*2

**2

*

*2

***

*

** ***

*** ***

*** **

* *

** *

* *

** **

** ***

** ***

* **

* **

* *

**

**

***

**

**

**

*

*

***

*

*

*

Water 1) Irrigation 2) Nutrient loss 3) Nutrient demand 4) Pesticides Nature conservation, biodiversity and landscape Air Quality (nitrous oxide, methane)

Notes: *** = high; ** = moderate; * = low. 1. Assuming the crop is dominant in a given area. The comparisons are made among the arable crops considered in the study in terms of their relative importance of the respective issue. 2. Assuming conservation tillage practice. Source: OECD Secretariat.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 29

The report is organised as follows:

x

Chapter 1 provides an overview of the relative importance of the arable crop sector in production and trade, examines changes in farm structures and discusses the main driving forces that affect such changes, including changes in area, yields and chemical inputs.

x

Chapter 2 addresses the main environmental issues and impacts associated with arable crop farming, as well as management practice approaches aimed at reducing the adverse environmental impacts of arable crop production. A brief review of the main environmental issues associated with transgenic crops is also provided.

x

Chapter 3 examines the agricultural support measures provided to arable crop farmers in OECD countries and reviews their evolution over the last fifteen years. The analysis draws extensively on the OECD’s PSE/CSE database.

x

Chapter 4 discusses agri-environmental and environmental policy measures designed to address environmental issues associated with arable crop farming. The classification of the various types of policy instruments used in the OECD Inventory of Agri-environmental Measures is also adopted here.

x

Chapter 5 endeavours to analyse the environmental effects of agricultural support policies for the arable crop sector, including shifting support from market price support to direct payments. It also provides an assessment of the cost-effectiveness and efficiency of cross-compliance measures.

x

Chapter 6 explores some of the environmental impacts of further multilateral trade liberalisation and reduction in support on arable crops, using a multi-country, global trade model and indicators of environmental quality. The model is based on standard economic theory and it allows consideration of the general equilibrium impact of food and agricultural policies by accounting for inter-sectoral linkages and inter-sectoral competition for land and other resources. Environmental impacts include changes in the scale and intensity of input use for crop production, changes in pesticide use, nitrogen uptake and off-load, and impacts on emissions of greenhouse gases from crop production.

30 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

x

Chapter 7 provides a cross-country analysis of the trade effects of agri-environmental payments on the arable crop sector and examines the extent to which environmental regulations affect the factor costs for arable crop producers. It also offers some practical suggestions for enhancing the effectiveness of agri-environmental policies related to arable crops in achieving their environmental objectives without “distorting” trade flows.

Note

1.

The FAO projections suggest that, over the next 30 years, developing countries will need an additional 120 million ha for growing crops, an overall increase of 12.5%, and that land expansion will mainly take place in sub-Saharan Africa and Latin America. Overall, land expansion is expected to account for 20% of growth in crop production in developing countries, yield improvements for about 70%, and increased cropping density for the remainder.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 31

This page intentionally left blank

Chapter 1 ECONOMIC AND STRUCTURAL ASPECTS OF THE ARABLE CROP SECTOR 1.1.

The arable crop sector in OECD countries The arable crop sector occupies approximately one-third of the OECD’s Utilised Agricultural Area (UAA),1 and contributes to around a half of the OECD’s total agricultural output. China, the United States, the European Union (EU), Russia, Argentina, Australia and Canada are the world’s main cereal producers. The OECD countries are responsible for approximately 80% of world cereal trade. In Canada, the arable crop sector covers around 60% of UAA, while in the United States the share is approximately just over 40%. In the EU, arable crops cover 40% of UAA [around 140 million hectares (ha)] and are grown in all member states, although there is high regional concentration. Cereal production is concentrated in five member countries (France, Germany, Spain, Italy and the United Kingdom) and about half of all cereal production in the EU is concentrated in 20 out of its 127 regions. In the EU, cereal production represents the most important use of arable land. According to the 1999/2000 Farm Structures Survey, around 37 million ha – 52% of the total arable area of 72 million ha – were allocated to the production of cereals, and 5.4 million – 0.1% of the total arable area – to oilseeds. Wide variations appear between EU member states, with three countries (France, Spain and Germany) together accounting for some 61% of the total area under cereal production. France had the largest cereals area (9 million ha) in 1999/2000, followed by Spain (almost 7 million ha) and Germany (6.6 million ha). Italy (4 million ha) and the United Kingdom (3.3 million ha) also had relatively important areas allocated to the production of cereals in 1999/2000. In terms of harvested area, wheat is the world’s largest cereal crop. Global production is estimated at approximately just under 600 million tonnes, with international trade at just over 100 million tonnes annually (Annex Table 1.A1). Maize is grown in more countries than any other cereal and it is the third most important cereal crop in the world, after wheat and rice. In OECD countries, maize production ranks second, after wheat. Six countries (the United States, China, Brazil, Mexico, France and Argentina) AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 33

produce 75% of the world’s maize, with the United States alone producing around 40% of the total. The United States is the largest maize (corn) producer and exporter, accounting for approximately 40% of the grain produced in the world, followed by China, Brazil and Mexico. The majority of the world’s maize production is used for animal feed or industrial input, with only approximately 20% going to human consumption. Mexico is an exemption, with 68% of all maize grown being used for human consumption. Approximately 94% of maize exports from the United States are destined for Latin America, in general, and Mexico, in particular (11% of US exports). Since 1996, US maize exported to Mexico has increased at the same time that exports to Europe have decreased. The decline in European markets coincided with the nascent production of transgenic maize in the United States. Maize cultivation is believed to have originated in Mexico, has particular cultural, social and economic significance. It is the country’s most important crop in terms of land area and the second in terms of gross production volume. There is a great diversity of the varieties of maize and of its wild species (i.e. the teosintes), although their population sizes and distribution have been affected by general land-use practices, intensive agriculture and urbanization (Dyer-Leal and Yúnez-Naude, 2003). Much of the crop is grown by subsistence farmers on small plots under rain-fed conditions, where yields are typically low. Maize draws more heavily on soil nutrients than other grains and oilseeds, and substantial amounts of fertiliser and water are needed to maintain yields. Maize is often planted in rotation with other crops. Rice is the main source of food for about half of the world’s population. It is cultivated in more than 100 countries, but around 90% of the world’s rice is grown and consumed in Asia (China, India and Indonesia), and 96% in developing countries. In OECD countries, rice is produced in Australia, Japan, Korea, the United States and Italy and, to a smaller extent, in Spain, France, Greece and Portugal. Thus, rice, relative to grains and oilseeds which are produced over a more diverse area, is more dependent on narrower climatic conditions. Worldwide trade in rice expanded at an average rate of 7% a year during the 1990s, to reach about 25 million tonnes in 2003. Despite such growth, the world rice market remains a “thin or residual market”, as only a small proportion of production is traded. Data indicate that only around 5% of global rice production is traded, as compared with 18% for wheat, 12% for coarse grains and nearly 25% for soybeans. This thin world market, in tandem with low price supply and demand elasticities, implies that a small production shortfall in an important

34 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

rice-producing country could trigger a sharp rise in world rice prices. In order to protect producers and consumers from such price fluctuations, countries where rice is an important staple food have traditionally insulated their domestic rice markets from international rice markets and pursue a high degree of rice self-sufficiency. Another stylised feature is that the international rice market is segmented by type and quality, with little substitution in consumption and production. Market segmentation makes the international rice market even thinner, further contributing to price volatility. Of the 27 million tonnes of rice traded internationally in 2002, about 80% is indica and around 10% is japonica, with speciality varieties such as Indian “basmati” rice making up the rest. The bulk of the world’s rice trade occurs among developing countries. Thailand, Vietnam, the United States, China, India and Pakistan are the largest exporters, typically accounting for 75% of global exports. The EU and Mexico are large importers of high-quality indica rice, while Japan is the largest importer of japonica rice, followed by Turkey and Korea. Australia, although not a major rice producer, is an important rice exporter (about 80% of the harvest is exported). The various methods of cultivation used in different ecological conditions have, over time, led to the evolution of different types of rice. There are now four recognised ecosystems under which rice is grown: irrigated, rain-fed lowland, upland and flood-prone. Worldwide, irrigated rice accounts for almost three-quarters of total rice production. In Japan, all rice production is irrigated and sometimes, depending on location and climate, rice cultivation is rotated with wheat, barley and soybeans. In Korea, 79% of the harvested rice area is under irrigation, while 20% is on rain-fed lowlands. In the United States, all rice production is irrigated. In Australia, production is highly advanced and mechanised, and rice is generally planted in rotation with pasture. All rice production is irrigated. In Europe, rice is cultivated with permanent flooding. Oilseeds and oilseed products occupy a prominent place in world agriculture and play an important role in the agricultural sectors in all OECD countries, either through production or utilisation. Oilseeds rank third after livestock products and cereals, in terms of value of both world production and world trade. Virtually all oilseeds are crushed and processed to produce oil and meal. Most of the vegetable oil is used for human consumption, although relatively small but growing quantities are utilised for industrial purposes. Meal is used predominantly for animal feed, although in some countries it is also used as a fertiliser.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 35

Oilseeds span a range of agricultural crops, ranging from arable crops grown solely in temperate zones to tree crops grown solely in the tropics. Correspondingly, the nature and extent of any environmental effects associated with oilseed production vary considerably depending on the type of crop cultivated. Almost all types of oilseeds are joint product commodities, with the seed price determined jointly by the value of the oilseed meal and vegetable oil. In this study, the discussion and analysis will be primarily limited to the three major oilseeds (soybeans, rapeseed, sunflower seed) produced and used in the OECD region. OECD member countries currently account for close to 40% of the total world oilseed production (the United States, 29%; EU, 5%; Canada, 3%) and a similar level of world consumption. Total use of oilmeal within the OECD area accounts for about 55% of total world utilisation (the United States, 19%; EU, 27%; Japan, 5%), reflecting the higher meat production in OECD countries compared with non-OECD countries. About 70% of all trade in oilseeds and oilmeal occur within the OECD region. The two major importers of oilseeds are the EU and Japan. In contrast to the cereals markets, most OECD oilseed exports are destined for the OECD area, whilst the majority of OECD oilseed imports originate outside the OECD region. Soybeans are the world’s dominant oilseed crop and their price has a major influence on that of other oilseeds. Soybeans account for almost 55% of world oilseed production and about three-quarters of the oilseed trade. OECD member countries account for over 50% of the world’s soybean production, although almost all of this production occurs in the United States, which is the world’s largest soybean producer. Rapeseed accounts for about only 10% of world oilseed production, but is the second most important oilseed crop traded on world markets. OECD countries accounted for approximately 45% of rapeseed production and approximately 90% of rapeseed trade during the 2001-03 period. Production is concentrated in the EU (26%), China (24%) and Canada (20%). However, Canada is the principal exporter of rapeseed (canola), accounting for 60% of world exports, with almost half of this quantity going to Japan. World sunflower seed production is ranked a close third, after rapeseed, in terms of oilseed production and trade. Sunflower production takes place predominantly in Russia, Argentina and the EU. The OECD region accounts for 25% of world sunflower production and consumption, with most of the production occurring in the EU (14%). The United States accounts for 7% of world production.

36 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

1.2.

Developments in farm structures Farm structure and environmental concerns are closely related. Farm structures and their evolution over time are crucial for understanding the way the sector functions and for assessing the likely effects of agricultural policies on the environment. Farm households in different farm-size categories organise their production practices, financing and marketing strategies in different ways. Consequently, a predominance of small- to medium-sized farms would have different environmental implications from a predominance of large farms, both of which might not be homogeneous across regions (OECD, 1998a).

1.2.1.

Changes in number and size of farms

The evidence suggests that there are fewer arable farms at the beginning of the 21st century than in the latter part of the 20th century (Figure 1.1). The total number of arable crop farms (cereals and oilseeds) in the EU has declined consistently over the last fifteen years, with an annual average rate of decline of around 4% for the EU-12 as a whole during the 1987-2000 period. In 1999/2000, there were around 2.6 million farms producing cereals and 441 000 farms producing oilseeds. This decrease was driven by the large declines in the number of farms in Italy (3.5%), Spain (5.6%), France (4%), Portugal (4%) and Germany (4%). The decline in arable crop farms in OECD countries has been associated with either a reduction, or only small increases in the area used for arable crops (Annex Table 1.A2). Between the mid-1980s and 2000, the area planted for arable crops in OECD countries as a whole decreased by an average annual rate of around 0.2%. Because of the rates of decline in area devoted to arable crops were lower than the rates of declines in the number of farms, the arable crop area per farm has risen over time. In the EU, the average size of cereal farms increased by more than 4% per annum in Germany (6%), Spain (4.3%), Ireland (7%) and Finland (8%). In Australia, the number of grain-producing farms fell by one-third between the late 1970s and 2001-02, while the average area operated by grains farms increased by 34% (ABARE, 2003). This has enabled an increase of more than 25% in the average area planted to crops per farm. In Japan, the number of rice farms fell by 0.9 million over the 1985-2000 period, while the area harvested declined by 0.5 million ha. Figure 1.1 and Figure 1.2 also underscore the great diversity in the amount of area allocated to arable crop per holding across OECD countries, as well as the differences in rates of change over time. Average size in 2000 ranged from 0.8 ha per farm in Japan to 1654 ha per farm in Australia. The average size of arable crop farms varies considerably within the EU (51 ha in the United Kingdom compared to 3 ha in Portugal). AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 37

Figure 1.1. Number of cereal farms in selected countries

3500

Thousand farms

3000 Germany

2500

Spain 2000

France

1500

Italy Japan

1000

Australia

500 0 1985

1987

1990

1993

1995

1997

2000

Notes: Japan: farms cultivating rice. Australia: 1999-2000 to 2001-02 average. Sources: OECD Secretariat based on EUROSTAT; Statistical Yearbook, MAFF, Japan; ABARE (2003).

The magnitude of the structural change is reflected in the proportional share of the different size categories. In the EU, the majority of arable crop farms are relatively small in size, with 58% of all farms utilising less than 5 ha. In Greece Italy, Italy and Portugal, the proportion of small farms in the national cereals total is even more pronounced, with over three-quarters of farms using less than 5 ha. On the other hand, farms using more than 100 ha per farm account for some 3% of cereal area for the EU as a whole, with the largest proportion being in the United Kingdom (4%). Also in Denmark, France and Luxembourg the percentage of large cereal farms using more than 100 ha accounts for 10% or more. Large farms (i.e. over 100 ha per farm) and specialist cereal farms can also be found in the new German Länder. In Finland, between 1995-2002, the share of arable farms receiving agricultural support with less than 20 ha has fallen from 56% to 44%, while the share of farms with more than 50 ha has more than doubled from 7% to 17% (MTT, 2003).

38 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 1.2. Average size of cereal farms in selected countries

30

ha

25 20

Germany Spain

15

France Italy Japan

10 5 0 1985

1987

1990

1993

1995

1997

2000

Notes: Japan: farms cultivating rice. Australia, with an average area of 1 654 ha in 2000, cannot be shown on this graph. Sources: OECD Secretariat, based on EUROSTAT; Statistical Yearbook, MAFF, Japan.

In the United States, grain farms (including soybeans) in 1997 accounted for about 24% of total farms. Small farms, those with 100 ha or less of maize, comprised 75% of all US maize farms and produce 29% of US maize production. At the other extreme, fewer than 4% produced just under 20% of US maize (Foreman, 2001). Similar trends are found for other arable crops. For example, for soybeans in 1997, small farms comprised 75% of soybeans farms and accounted for 30% of total US soybean production. On the other hand, very large farms, those with 300 ha or more, comprised 9% of soybeans farms and produced 39% of soybeans using 22% of the soybean acreage. For rice in 2000, small farms accounted for 43% of rice farms, but just 16% of the rice production (Livezey and Foreman, 2004). In contrast, very large farms comprised 11% of all rice farms and 33% of rice production. Farms also tend to be specialised, rather than diversified. In the EU, the share of farms specialising in mixed crops and livestock declined from around 10% in 1987 to 6.5% by 2000, and the share of farms specialising in mixed cropping declined from more than 12% to around 7% over the same AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 39

period. Moreover, specialisation differs by farm size. In the United States, over half of farms produce just one commodity. Specialisation in grain production is more frequent for farms with sales between USD 100 000 and less than USD 499 999 (Hoppe, et al., 2001). These trends have encouraged in many regions monocultures and loss of mixed farming, with important impacts on land use, landscapes and biodiversity. An overview of the distribution of cereal farms by economic size2 suggests that, despite the diversity across member countries, the number of larger, more capital-intensive and specialised cereal farms has increased over time (Table 1.1). The increase in the number of large and very large cereal farms is more evident in Denmark, Germany and Spain. Table 1.1. Distribution of cereal farms by size for selected EU countries (%) 1990

1995

Small farms

Large farms

Very large farms

29

9 20 0 3 36 4 4 41 13

2 1 0 1 7 1 0 22 3

Denmark Germany Greece Spain France Italy Portugal UK EU

84 54 0 69 80 0 44

2001

Small farms

Large farms

Very large farms

79 41 0 63 68 0 36

11 19 1 7 39 4 6 35 14

3 9 0 1 10 1 1 27 4

Small farms

Large farms

Very large farms

77 50 0 57 56 0 33

13 27 1 9 42 5 9 35 9

4 23 0 2 19 1 2 35 2

Notes: Small farms = farms with less than 8 ESU. Large farms = farms with 40-< 100 ESU. Very large farms = farms with >= 100 ESU. Source: EC, RICA Database, 2003.

1.2.2.

Regional concentration

The trend towards fewer but larger cereal farms applies throughout the OECD area, although to varying degrees. However, aggregate national statistics conceal divergent trends within farms of different size and in different regions across countries. Arable crops production is also characterised by regional concentration, reflecting the resource endowment, climate, soil types and policy changes in diverse regions. Approximately half of the EU’s cereal production comes from 20 of the 127 regions of EU15. This concentration is particularly high in Denmark, Germany (Bayern, Niedersachsen and Nordrhein-Westfalen), France (Centre, Picardie, Champagne-Ardenne, PoitouCharentes and Midi-Pyrenées), Spain (Castilla-Leon and Castilla-la Mancha)

40 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

and the United Kingdom (the south-east). In Finland, most crop production is located in southern and south-western parts of the country, due to natural conditions more than half the crop farms are located in the main region of southern Finland (MTT, 2003). In the United States, over half of maize farmers and maize acreage are found in the Heartlands and about one-fifth is located in the Northern Crescent.3 There is also a great diversity in the average size, ranging from 27 ha per farm in the south-east to 133 ha per farm in the Prairie Gateway (Foreman, 2001). The Prairie Gateway has the fewest maize producers, but much larger farms (an average of 574 ha), producing nearly 15% of US maize production.

1.2.3.

Sources of growth in production

Notwithstanding the diversity between countries, arable crops production in OECD countries increased, on average, by 0.5% per annum since the mid-1980s. The fastest growth was observed for: wheat in Denmark (4.2% per annum), coarse grains in the Netherlands (3.6% per annum.), wheat in Ireland (3.3% per annum) rice in Australia (3.6% per annum); and soybeans in France (3.9%) (Table 1.2 and Annex Table 1.A3). There are two sources of growth in crop production: harvested land expansion and yield growth. As illustrated in Table 1.2, arable crops production in the OECD area has derived most of its growth from an increasingly intensive use of land that was already under crops, rather than expansion of the harvested area, although area expansion remains the main source of growth in several countries. Growth in wheat, coarse grains production and rice production in a number of OECD countries stems primarily from gains in yield (more than 80%), while expansion of harvested land was a major contributor to production growth of soybeans (Annex Table 1.A4). In several OECD countries, expansion in harvested area was an important source of growth in arable crop production over the 1985-2002 period. The contribution of increases in harvested area to production growth can be decomposed into the effects stemming from increases in arable land expansion and to effects due to increases in cropping intensities, such as multiple cropping and shorter fallow periods. As shown in Table 1.3, in most of the cases the contribution of increases in cropping intensity was more important than the contribution of arable land expansion in explaining expansion of harvested area. Arable crop yields for the OECD area have increased, on average, at 0.7% per annum. For the OECD area as a whole, the rate of yield growth is fairly similar between arable crops, ranging from 0.6% per annum for rice, to 0.9% per annum for coarse grains. Technological advances, including AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 41

better farming practices and improved varieties, as well as policies all contributed to increasing yields. Policy developments, for example, such as the 1992 CAP reform in the EU, are an important factor in influencing arable crop production, yield and area. The evidence so far shows that the arable crop yields in the EU for the post-CAP reform period (1995-2002) increased at a higher rate than in the pre-1992 period, suggesting that technological advances more than offset the effects of reduction in market price support for arable crops brought about by the 1992-CAP reform. Differences in average arable crop yields between countries are considerable. Figures 1.3 and 1.4 display the wheat and maize yields (threeyear averages 2000-02) in the major wheat- and maize- producing countries of the OECD. Yields vary from a high range of 6.0-7.8 tonnes per ha in four EU countries (the United Kingdom, Denmark, Germany and France); through an upper middle range of 3.0-4.0 tonnes per ha (Hungary, Poland and Italy) to a lower middle range of 2.4-2.7 tonnes per ha (the United States, Spain and Canada); and down to the low-yield range of 1.022.2 tonnes per ha of Australia and Portugal. Analogous wide differentials exist for maize and other arable crops. The reasons why country average yields differ from one another are many, including agro-ecological diversity and socio-economical factors. Irrigation, for example, is an important factor in the achievement of high yields in several countries. Moreover, agro-ecological and demand factors influence the mix of varieties of the same crop grown in each country; for example, low-yielding durum wheat versus higher-yielding common or soft wheat. In addition, agricultural support policies, including input subsidies, encouraged farmers to use more fertilisers and pesticides in order to maximise yields. However, further increases in yields might be restrained by increased environmental concerns which restrict fertiliser applications. A study by FAO (2003), has distinguished between contributions resulting specifically from agro-ecological factors and other factors. The agro-ecological attainable yields can be used to draw inferences about the potential environmental risks associated with increasing intensification. The higher the gap between actual and agro-ecological attainable yields, the larger the potential environmental risk. As shown in Table 1.4, France, Sweden, the United Kingdom and Germany have actual yields close to, or even higher, than those attainable for their agro-ecological endowments under rain-fed, high-input farming.

42 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 1.2. Sources of growth in arable crop production, 1985-2002 (%) Contribution of:

Wheat Australia Denmark France Germany Ireland Norway Sweden Coarse grains Australia Italy Mexico Netherlands Turkey Rice Australia United States Soybeans France Spain United States Arable crops Australia Czech Republic France Germany Netherlands

Production growth

Harvested land expansion

Yields

1.4 4.2 1.5 2.4 3.3 2.1 2.2

36 75 16 49 38 96 57

64 25 84 51 62 4 43

2.0 2.2 1.6 3.6 0.1

42 18 15 79 29

58 82 85 21 71

3.6 2.5

-139 -132

239 232

3.9 1.0 1.9

80 60 62

20 40 38

1.9 0.7 1.2 1.5 1.8

51 81 2 17 64

49 19 98 83 36

Note: Calculations are based on three-year averages: 1985 = 1985, 1986 and 1987; 2002 = 2000, 2001 and 2002. Source: OECD Secretariat calculations based on FAOSTAT, January 2005.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 43

Table 1.3. Sources of growth in harvested area for arable crops, 1985-2002 (%) Arable land expansion

Increases in cropping density

Australia

34

66

Czech Republic

-29

129

Hungary

-214

314

Netherlands

37

63

Poland

-95

195

Note: Calculations are based on three-year averages: 1985 = 1985, 1986 and 1987; 2002 = 2000, 2001 and 2002. Source: OECD Secretariat calculations based on FAOSTAT, January 2005.

Ita ly Sp ain US A Gr ee ce Tu rk ey Ca na d Au a str ali Po a rtu ga l

8 7 6 5 4 3 2 1 0 U Ge K rm a De ny nm ar k Fr an c Hu e ng ar y Po lan d OE CD

(tonnes/ha)

Figure 1.3. Wheat yields (average 2000-02)

Source: FAO, FAOSTAT, January 2005.

44 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

ex i

co

ey M

ry ga

Tu rk

l ga H

un

rtu

a

la nd

Po

Po

ad

D

an C

O

EC

SA U

ce an Fr

an y

ly G

er m

ai Sp

re ec G

Ita

n

10 9 8 7 6 5 4 3 2 1 0 e

(tonnes/ha)

Figure 1.4. Maize yields (average 2000-02)

Source: FAO, FAOSTAT, January 2005.

Table 1.4 Potential environmental risk from increasing intensification of rain-fed wheat production, selected countries, 1996/2000 Area suitable for rain-fed wheat Total

ha mill. Australia Canada France Germany Hungary Italy Japan Poland Sweden Turkey UK US

24.3 42.2 24.6 16.9 6.1 7.6 6.4 17.6 4.3 7.6 11.9 230.4

% of area by suitability class

VS 17.5 10.7 26.0 42.5 11.6 31.0 31.0 26.6 0.0 8.2 4.0 18.8

Yields attainable

Actual yields

Tonnes/ha

Area Yield (mill. ha) (tonnes/ha)

S

M

VS

S

M

38.0 35.0 45.6 39.2 51.5 46.9 39.7 51.0 54.8 31.3 70.6 54.1

44.5 54.3 28.4 18.3 36.9 22.2 29.3 22.5 45.2 60.4 25.4 27.1

6.2 6.3 8.4 9.0 8.5 8.6 8.9 8.7 0.0 5.7 8.4 6.5

4.5 5.6 6.7 7.1 6.8 6.2 7.0 7.2 5.7 5.9 7.2 6.1

3.2 3.1 4.7 5.2 5.2 4.0 5.1 5.1 4.2 4.0 4.8 4.6

Avg. all classes 4.2 4.3 6.6 7.6 6.4 6.5 7.1 7.1 5.0 4.8 6.7 5.8

Area Yield (mill. ha) (tonnes/ha) 11.1 10.9 5.2 2.7 1.1 2.4 0.2 2.5 0.4 9.1 2.0 23.7

2.0 2.4 7.1 7.3 3.9 3.2 3.4 3.4 6.0 2.1 7.8 2.7

Notes: Area suitable for rain-fed wheat indicates land that – given soil and climate characteristics and taking into account physical and chemical requirements for growing wheat – could potentially produce wheat. Countries with predominantly rain-fed wheat with over 5 million ha of land in the wheat suitability classes: VS (very suitable), S (suitable) and MS (moderately suitable) under high input. Source: FAO (2003), Table 11.1. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 45

1.2.4.

Chemical inputs

Increased use of agro-chemicals such as fertilisers and pesticides has been one of the most prominent factors increasing arable crop yields. Historically, around one-third of the increase in cereal production worldwide has been attributed to increased fertiliser consumption (FAO, 2003). Fertiliser and pesticide use vary significantly by arable crop, by country and by region. Their rate of application depends on a variety of factors, including soil type, crop mix, crop rotation, irrigation, climate, technology and government policies. Table 1.5 and Annex Table 1.5 show use across the OECD area, as a whole and by country, of the major nutrient elements (nitrogen (N), phosphate (P2O5) and potash (K2O)), respectively. Table 1.6 shows fertiliser use by nutrient and pesticide application by class, for selected crops in the United States. Table 1.5. Average application rates of fertilisers by nutrient and by crop in the OECD area, 1992-2000 N

P2O5

K2O

Rate (Kg/ha) Wheat

83

37

27

Maize

135

56

66

Barley

80

32

23

Rice (Paddy)

121

74

72

Rapeseed

101

33

41

Soybeans

30

39

71

Sunflower

44

31

37

Notes: Four-year average. Australia is not included due to data limitations. Source: IFA/IFDC/FAO, various issues.

Overall, these data reveal that the most frequently applied nutrient in arable crop farming is nitrogen. In terms of fertiliser intensity (as measured by the average nutrient use per ha), maize and rice are the most intensively fertilised crops, while sunflower is the least fertiliser dependent arable crop. There are also significant variations between countries in consumption of nutrients per ha of harvested area. For example, consumption of nitrogen varies between 11 kg/ha in Canada for soybeans, to 178 kg/ha in Italy for maize (Annex Table 1.A5). The United States allocates almost half of its fertiliser use to wheat and maize, while France devotes three-quarters of its fertiliser use to rapeseed.

46 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 1.6. Chemical use for selected crops in the United States, 1999 Percentage of area treated and total fertiliser applied Nitrogen

Fertiliser use, by nutrient: Phosphate Potash

Percentage of area receiving applications and total pesticides applied Herbicides

Pesticide use, by class: Insecticides Fungicides

(%) 1 000 t

(%) 1 000 t

(%) 1 000 t

(%)

1 000 t

97

21

91

14

90

18

39

13

Corn

98

4 064

82

1 381

67

1 696

98

Soybeans

18

116

26

369

28

672

Sunflower

90

79

43

15

8

Cotton

86

445

59

177

52

Wheat

1

(%)

1 000 t

70

30

5

96

32

2

9

1

95

1 283

33

55

252

97

11

84

18

(%) 1 000 t

7

329

Note: 1. Includes both spring and winter wheat; but winter wheat only for Indiana State. Source: USDA (2000).

Arable crop farming in the United States relies on intensive application of fertiliser, herbicides and insecticide. Of the 15 states included in the USDA’s survey in 1999, 98% of the planted maize acreage received nitrogen, 82% phosphates and 67% potash (USDA, 2000). Herbicides were applied to 98% of the maize acreage. Soybeans producers applied nitrogen fertiliser to 18% of the area planted, phosphate to 26% and potash to 28%. Soybeans growers applied insecticide to only 2% of the soybeans area planted. They also reported few fungicide or other chemical applications. For sunflower, nitrogen was applied to 90% of the total sunflower area. In the states surveyed, 43% of the planted sunflower acreage received phosphates, and potash was applied to 8% of the acreage. Herbicides were applied to 95% of the sunflower acreage and 33% of the area was treated with insecticides. For wheat, nitrogen fertiliser was applied to 97% of the area planted for 1999 in Indiana. Phosphate fertilisers were to 91% of the wheat acreage. Fertiliser use and pesticide application have slowed in recent years compared to the mid-1980s, with rates of decline varying across OECD countries (IFA/IFDC/FAO, 2002). Many factors have contributed to the overall reduction in fertiliser and pesticide volumes applied, including weather and seasonal conditions, fertiliser and pesticide prices, government policies, including set-aside and taxes in some countries, as well as environmental pressures.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 47

Other factors, such as the impact on soil fertility of more intensive cultivation practices and the shortening of fallow periods, are important determinants of increased yields. There is some empirical evidence suggesting that increases in production and yields can be achieved with a less than proportional increase in fertiliser nutrient use. A study on maize in the North America, for example, found increased nutrient-use efficiency was achieved through the adoption of improved and more precise management practices (Frink, Waggoner and Ausubel, 1998). Socolow (1998) suggests that management techniques, such as precision agriculture offer abundant opportunities to provide information on fertiliser management. The current trend of increasing nutrient use efficiency through better nutrient management and by improving the efficacy of nutrient balances and the timing and application of fertilisers, will continue to increase and accelerate in the future (FAO, 2003).

48 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Annex 1.A. Selected Data Table 1.A1. Area, yield, production and exports, 2000-02 C om m odity W hea t OECD A ustra lia C a na da EU Fra nce G e rm a ny U n ited S tate s A rg en tin a W o rld M a ize OECD C a na da EU Fra nce Ita ly S pa in M e xico U n ited S tate s A rg en tin a B ra zil W o rld R ic e OECD EU Ita ly Ja pa n K orea U n ited S tate s C hin a Ind ia Th aila nd W o rld S oybe ans OECD C a na da EU Fra nce Ita ly U n ited S tate s A rg en tin a B ra zil W o rld

A rea H arvested (’0 00 ha) 74 11 10 17 5 2 19 6

17 3 59 4 09 0 55 2 08 2 96 0 92 2 39 8

2 14 59 9

Yie ld

P ro du ction

E xp orts

kg/h a

(’00 0 ton ne s)

(’0 00 ton ne s)

3 1 2 5 7 7 2 2

17 7 62 5 07 4 73 0 06 0 35 7 63 4 28 1

2 16 10 16 1 04 38 20 44 12

444 059 198 376 939 818 063 300

89 15 16 27 15 5 25 10

E xpo rt sh ares (% )

03 1 98 8 21 1 73 8 77 8 38 4 95 3 28 7

76 14 14 24 13 5 22 9

1 17 10 8

10 0

2 71 8

5 73 967

56 4 21 3 41 3 83 7 10 6 46 9 35 4 40 4 78 0 89 9 67 3

7 6 9 8 9 9 2 8 5 3 4

32 8 67 4 11 5 87 3 50 0 52 8 58 3 47 3 68 5 06 8 34 9

3 09 8 40 16 10 4 19 2 28 15 35 6 01

597 999 821 440 824 463 299 805 000 933 994

58 91 7 22 6 9 10 1 7 79 1 19 6 11 8 59 47 86 7 10 42 1 28 8 81 92 0

72 0 11 10 0 0 0 58 13 0 10 0

4 75 7 39 9 21 9 1 72 1 1 07 0 10 00 2 1 29 0 29 31 8 43 24 8 1 51 11 7

6 6 5 6 6 2 7 6 2 3

79 8 36 6 90 0 64 0 63 3 61 4 22 9 20 1 89 3 90 0

31 2 1 11 6 26 9 1 76 1 08 5 71

763 607 371 111 687 057 569 342 900 076

5 01 7 1 39 7 60 8 20 9 0 7 05 5 2 87 5 2 38 3 2 92 7 26 00 5

19 5 2 1 0 27 11 9 11 10 0

31 13 5 1 05 1 32 4 91 21 3 29 38 3 10 15 1 14 66 0 76 69 6

2 2 3 2 3 2 2 2 2

57 4 12 0 30 9 64 5 70 4 59 3 52 2 56 2 25 5

78 655 2 336 816 209 566 74 825 30 180 42 125 1 80 910

30 07 1 63 7 1 56 3 16 13 27 85 3 5 88 4 14 38 8 52 99 0

57 1 3 0 0 53 11 27 10 0

43 1 4 1 1 7 28 2 11 1 38

Source: FAO, FAOSTAT, January, 2005.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 49

Table 1.A2. Growth of area, by arable crop, 1985-2002 (%)

Country Australia Austria Belgium Canada Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy Japan Korea Mexico Netherlands New Zealand Norway Poland Portugal Spain Sweden Switzerland Turkey United Kingdom United States OECD

Arable crops 0.9 -0.9 -1.2 -0.7 0.5 -0.6 -0.1 0.0 0.3 -1.0 0.2 -1.4 -0.7 -1.4 -1.5 -0.3 1.1 -3.1 -0.1 0.2 -3.0 -0.8 -1.5 -0.2 0.0 -1.0 -0.3 -0.2

Wheat 0.5 -0.6 -0.2 -1.7 0.9 3.0 0.1 0.2 1.2 -0.2 -1.0 1.3 -1.5 -1.4 -2.6 -2.8 0.6 -2.9 2.0 1.3 -2.0 0.5 1.2 -0.3 0.0 -0.2 -1.1 -0.5

Coarse grains 0.8 -1.7 -2.8 -0.8 -1.0 -1.5 -0.1 -0.8 -1.1 -1.9 1.0 -2.2 0.4 -1.6 -4.8 0.3 2.7 -3.1 -0.6 -0.2 -4.0 -1.5 -2.0 -0.3 0.1 -2.5 -0.9 -0.6

Rice

Soybeans

2.0

-2.4 5.5 5.3 8.4

2.8 0.8 -8.2

3.2 -5.7 2.0 -1.1

0.8 -1.5 -0.8 -5.6

-1.3 -0.2 -3.2 -10.0

-1.4 2.3

0.6

0.3

1.1 -7.3

1.6 -0.4

1.2 1.1

Note: Calculations are based on 3-year averages: 1985 = 3-year average of 1985, 1986 and 1987; 2002 = 3-year average of 2000, 2001 and 2002.

Source: OECD Secretariat calculations based on FAOSTAT, January, 2005.

50 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 1.A3. Growth of production, by arable crop, 1985-2002 (%)

Country Australia Austria Belgium Canada Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy Japan Korea Mexico Netherlands New Zealand Norway Poland Portugal Spain Sweden Switzerland Turkey United Kingdom United States OECD

Arable crops

Wheat

Coarse grains

1.9 -0.5 0.7 -0.6 0.7 0.7 1.2 1.2 1.5 -0.4 -0.9 0.1 0.6 -1.2 -0.8 0.8 1.8 -1.2 0.1 0.1 -0.4 0.5 -0.2 0.8 0.4 -0.2 0.4 0.5

1.4 -0.2 1.3 -1.4 1.3 4.2 1.2 1.5 2.4 -0.1 -1.9 3.3 -1.2 -0.9 -2.9 -2.0 1.2 -0.2 2.1 1.2 -2.4 1.2 2.2 0.4 0.4 0.9 -0.7 0.3

2.0 -0.9 -0.3 -0.6 -0.9 -0.9 1.3 0.8 0.4 -0.5 -0.2 -1.1 2.2 -2.5 -4.4 1.6 3.6 -1.7 -0.4 -0.4 0.3 0.2 -1.2 1.2 0.5 -1.9 0.3 0.3

Rice Soybeans 3.6

-1.9

3.3

3.9

1.4 -1.4 0.9 -1.2 -0.5 -4.8

-0.7 0.3 -3.3 -10.9

3.1

1.0

1.5

-6.6

2.5 0.1

1.9 1.8

Note: Calculations are based on 3-year averages: 1985 = 3-year average of 1985, 1986 and 1987; 2002 = 3-year average of 2000, 2001 and 2002. Source: OECD Secretariat calculations based on FAOSTAT, January, 2005.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 51

Table 1.A4. Growth of yields, by arable crop, 1985-2002 (%)

Country Australia

Arable crops

Wheat

Coarse grains

Rice

Soybeans

0.9

0.9

1.1

1.6

0.5

Austria

0.4

0.4

0.9

2.0

Belgium

1.9

1.4

2.6

0.0

Canada

0.0

0.3

0.1

-1.1

Czech Republic

0.1

0.4

0.1

2.2

Denmark

1.3

1.2

0.6

Finland

1.4

1.2

1.4

France

1.2

1.2

1.6

Germany

1.2

1.2

1.5

0.4

1.0 -2.4

Greece

0.6

0.1

1.4

0.7

-1.8

Hungary

-1.1

-1.0

-1.2

0.5

-0.4

1.5

2.0

1.1

Ireland Italy

1.3

0.3

1.8

0.0

0.7

Japan

0.3

0.4

-0.9

0.3

0.5

Korea

0.7

-0.3

0.4

0.3

-0.1

Mexico

1.1

0.8

1.4

0.9

-1.0

Netherlands

0.7

0.6

0.8

New Zealand

1.9

2.9

1.5

Norway

0.2

0.1

0.1

Poland

-0.1

-0.1

-0.3

Portugal

2.6

-0.5

4.5

1.4

Spain

1.3

0.7

1.7

0.8

Sweden

1.3

1.0

0.8

Switzerland

1.0

0.7

1.5

Turkey

0.4

0.4

0.4

United Kingdom

0.8

1.0

0.5

United States

0.7

0.4

OECD

0.7

0.8

0.4 1.9

1.2

1.2

1.2

0.8

0.7

0.9

0.6

0.7

Note: Calculations are based on 3-year averages: 1985 = 3-year average of 1985, 1986 and 1987; 2002 = 3-year average of 2000, 2001 and 2002. Source: OECD Secretariat calculations based on FAOSTAT, January, 2005.

52 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 1.A5. Average area and application rates of fertilisers by nutrient and crop, 1992-20001

Wheat Canada France Germany Italy Poland Spain United Kingdom United States (spring) United States (winter) Maize Canada France France (silage) Germany (silage) Italy Mexico United States Barley Canada France Germany Spain Turkey United Kingdom United States Rice (Paddy) Japan Korea United States Rapeseed Canada France Germany United Kingdom United States Soybeans Canada Mexico United States Sunflow er 2 France Hungary 2 Italy Spain Turkey United States

Area (1 000 ha)

N

P 2O 5 Rate (Kg/ha)

K 2O

12260 3735 2682 2345 2510 2166 1993 19573 13138

46 157 144 99 76 86 186 71 75

25 54 31 67 34 43 47 37 41

5 49 83 36 44 26 49 36 54

1023 1752 1442 1182 990 7859 30286

152 166 48 79 178 82 145

51 66 31 31 90 24 63

95 53 56 31 42 3 84

4450 2250 2164 3556 3439 1206 3144

63 118 133 73 43 120 45

25 45 27 38 25 49 19

10 32 35 23 0 57 7

2030 1090 1249

82 149 150

95 71 40

82 82 44

4271 729 844 383 437

68 151 143 189 150

20 54 48 49 120

15 92 98 51 84

807 301 25727

11 30 32

34 34 40

99

810 520 516 1043 591 1179

56 50 55 19 96 23

53 13 63 12 46 8

102 15 46 11 24 6

72

Notes: 1. Average for four years. 2. Includes sunflower, soya and linseed. Source: IFA/IFDC/FAO, various issues.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 53

Table 1.A6. Arable crop production data for selected countries, by type of land, 1997/1999 Rainfed Land Crop

Irrigated Land

Area

Yield

Production

1000 ha

kg/ha

174 37 6 230 61 1 250 2 0 3 7 773 16 170 210 48 15 7 208 526

Total

Area

Yield

Production

1000 MT

1000 ha

kg/ha

1000 MT

1000 ha

kg/ha

1000 MT

2 733 3 455 2 142 1 307 2 480 1 235 0 875 2 206

474 129 13 342 80 3 099 2 0 3 17 148

553 62 1 249 181 665 0 99 0 2 760

5 148 5 038 3 843 2 046 4 477 0 1 570 0 4 126

2 847 312 4 800 370 2 977 0 156 0 11 388

727 99 7 479 242 1 915 2 99 3 10 534

4 571 4 442 2 426 1 860 3 174 1 235 1 570 875 2 709

3 321 441 18 142 449 6 077 2 156 3 28 537

1 455 1 735 1 817 1 778 909 1 000 903 1 680

24 295 381 85 13 7 188 883

53 420 366 10 0 0 94 849

2 409 4 200 2 615 2 723 0 0 1 696 3 388

127 1 764 957 27 0 0 159 2 875

69 590 576 58 15 7 302 1 374

2 182 3 490 2 324 1 942 909 1 000 1 149 2 735

151 2 059 1 338 112 13 7 347 3 758

23 691 29 103 1 350 2 266 28 597

2 806 8 261 6 511 3 179 25 656

66 477 240 433 8 670 7 202 73 333

Area

Yield Production

Mexico Wheat Rice Maize Barley Sorghum Rapeseed Soyabeans Sunflowers Cereals Korea Wheat Rice Maize Barley Millet Sorghum Soybeans Cereals United States Wheat Maize Rice Barley Soybeans

22 192 24 891 0 1 861 26 788

1 499 4 212 1 350 405 1 809

Source: FAO (2003), ESDG database; USDA (2003d).

54 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Notes 1.

The UAA is the total area taken up by arable land; permanent pasture and meadow; and land used for permanent crops and kitchen gardens.

2.

Economic size is based on the European Standard Unit (ESU). A farm has an economic size of 1 ESU if its total “standard gross margin”, that is, production minus certain variable costs, has a certain value in euros.

3.

These are regions depicting geographic specialisation in production of US farm commodities as defined by the Economic Research Service (ERS) of the United States Department of Agriculture (USDA).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 55

This page intentionally left blank

Chapter 2 ENVIRONMENTAL IMPACTS ASSOCIATED WITH PRODUCTION With the widespread intensification of arable farming, environmental consequences have become apparent throughout the OECD area. Such environmental impacts include damage to, and removal of, soil thereby threatening agricultural sustainability, and water pollution. Modern arable systems also impact upon biodiversity within the system itself, and on associated non-cropped habitats such as grassland, field boundaries and watercourses, as well as on the aesthetic quality of the arable landscape. There is a high degree of integration between the various environmental impacts of arable farming because crop production affects multiple environmental services through complex ecosystem linkages. For example, the conversion of grassland to an intensive form of arable crop production will reduce certain wildlife habitat and landscape values formerly provided by the grassland, increase erosion and release carbon emissions from tillage, increase the potential for nutrient and pesticide residue run-off and infiltration into surface and ground waters, and could increase surface or ground water withdrawals if supplemental irrigation is used. In this chapter, as far as possible, environmental impacts will be treated separately. Arable systems are also often highly integrated with livestock and forestry, and therefore references are made as appropriate. Generally, such multiple land use tends to be associated with higher biodiversity and landscape value compared with purely arable systems. The following environmental impacts of arable farming systems are discussed in this chapter:

2.1.

x

soil-related impacts;

x

water-related impacts;

x

air quality; and

x

biodiversity.

Soil-related impacts Soils under arable crop cultivation may be susceptible to erosion; declining organic matter resulting mainly from frequent cultivation; AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 57

pollution by pesticides and, to a lesser extent, heavy metals (Stoate, et al., 2001).1 These processes are highly interrelated. Farming practices are important driving forces influencing soil properties.

2.1.1.

Soil erosion

Soil erosion is widespread throughout OECD countries. Soil erosion can adversely affect crop productivity and damage the environment in a variety of ways. Impacts of soil erosion are felt both on-farm and off-site.2 Moreover, there is a direct link between the magnitude of soil erosion and loss of soil biodiversity (OECD, 2003c). There are two distinct, but related, facets of the on-site decline in productivity caused by soil erosion: short-term reduction in agronomic yield and long-term decline in soil productivity, resulting from a lessening in soil quality due to reduced rooting depth water-retaining capacity, soil organic matter and soil biodiversity. The two most important off-site impacts of erosion on the environment are, respectively, degradation of surface water by sediment and sediment deposition, and emission of greenhouse gases into the atmosphere (Heimlich, 1991). The risk of soil erosion from wheat cultivation is normally low, with soybean, sunflower and maize cultivation systems generally being associated with higher levels of soil erosion.3 For rice, soil erosion is constrained by the ground coverage offered by irrigation water during the early stages of growth and through the widespread use of terracing in upland rice cultivation. The system of terracing can prevent soil erosion and landslides. On the other hand, irrigated rice production systems may cause problems of soil salinisation and waterlogging, particularly in regions where irrigation water is often of poor quality and paddy fields are provided with inadequate drainage (van Tran, 1998). Expansion of upland rice farming systems may increase soil erosion and deforestation. The draining of coastal wetlands for rice cultivation can lead to the dehydration of soil, often causing sulphur to rise to the surface, with consequent acidification (Barbier and Mouret, 1998). Soil erosion is caused by wind and water. The rate of erosion is influenced by a combination of physical factors such as climate, topography, soil texture, crop type and management factors such as cultivation practices, dates of seeding and harvest and post-harvest residue levels. Higher rates of erosion can result from devoting larger areas to autumn cultivation, increasing field size, with the associated loss of hedges, and continuous arable cropping, all of which increase the exposure of soil to wind and water in space or time (Evans, 1996). Soil erosion is partly related to crop rotation. Available studies, mainly in the European context, seem to

58 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

suggest that, generally, lack of crop cover tends to increase erosion rates on arable land (Boatman, et al., 1999). Late-harvested spring-sown crops such as maize also increase the exposure of soils to erosion. Moreover, rainfall, slope and soil type can all have a major influence on erosion risk (Brouwer and Ervin, 2002). The capacity of farmland to produce, and the link between agricultural production practices and soil erosion, has been studied extensively. Recent research shows that on-site productivity losses from erosion are relatively small. Erosion-induced productivity decline is estimated to generate a potential annual loss of 0.3% in the value of the global production of selected crops, ranging from 0.04% per year in Europe, to 0.61% per year in Australia (den Biggelaar, et al., 2003). A USDA study found that average annual water-induced soil erosion rates vary widely by crop production area, soil, and region, but, in most cases, range between 12 and 17 t/ha/year (Eswaran and Reich, 2001). Estimates of annual production losses to erosion in the United States range from USD 40 million, to over USD 100 million (Crosson, 2004). Den Biggelaar, Wiebe and Breneman (2001), taking into account differences due to regional variations in soil and climate, but assuming unchanged farmer management practices, estimated the erosion-induced losses for wheat, maize, soybeans and cotton at only USD 56 million. The same study found that, although the erosion-induced yield loss varies widely by crop and region, there is, on average, a 0.3% per year loss in the value of global crop production, ranging from 0.04% in Europe, to 0.6% in Australia.4 This average yield loss ranges from 0.03% for wheat on Alfisols (fertile soils that occur primarily in the Corn Belt) in the United States, to 0.3% for wheat on Alfisols in Canada, and for soybeans on Ultisols (fertile but acidic soils that occur primarily in the Southeast) in the United States (Figure 2.1).5 The total loss in production for the United States was estimated at 229 000 tonnes for maize; 54 000 tonnes for wheat; 61 000 tonnes for soybeans; and 2 000 tonnes for cotton.6 While the estimated costs of erosion in terms of lost output are insignificant at the national level, there may be important regional and local impacts in terms of resource depreciation and off-site costs of crop production. For example, in the United States, Faeth (1993) found negative net economic value per hectare, after accounting for soil degradation and off-site costs, for Pennsylvania’s best maize-soybean rotation over 5 years.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 59

Figure 2.1. Annual yield losses due to erosion for selected crops in Canada and the United States, 1939-99

Percent per year

0.3 0.25 0.2

Alfisols

0.15

Mollisols Ultisols

0.1 0.05

he at .S .S oy be an s U .S .C ot to Ca n na da M ai ze Ca na da W he at U

U .S .W

U .S .M

ai ze

0

Source: USDA/ERS.

In Australia, rates of soil erosion associated with arable cropping are similar to those of native pastures. Of the land uses, sugarcane has the highest erosion rate (20.3 t/ha/year, as compared to 4.3 t/ha/year for oilseeds and 3.3 t/ha/year for cereals excluding rice) (Lu, et al., 2003). The same study found that although acceleration of current erosion rates above natural rates is relatively evenly distributed across Australia, there is a great diversity across various land uses: cereals (excluding rice) 18 times the natural rate; oilseeds 33 times; sugarcane 33 times – while grazing lands have rates typically 2-5 times the natural rate. Soil erosion from cropland is an issue of concern regionally in Canada, particularly in the arable plains of the Canadian wheat belt. Soil erosion is widespread in the EU, although levels of severity vary across countries, and between regions within countries (EEA, 2003a). Major causes are unsustainable agricultural practices, over-grazing, large-scale farming, construction activities, and poor water and irrigation management. Estimates of soil loss by erosion range from 3.6/t/ha to 40 t/ha/year (Boatman, et al., 1999). The European Soil Bureau and the Pan-European Soil Erosion Risk Assessment programme show that the south European region is the most prone to soil erosion – with most erosion linked to the occurrence of high rainfall in short periods during storms. There is also evidence of significant erosion occurring in other parts of Europe (e.g. Austria, Belgium, the Czech Republic, France and the United

60 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Kingdom). A study found that half of the arable fields surveyed in England and Wales showed signs of soil erosion at least every other year (Evans, 1996). Erosion by water is exacerbated by intense rainfall, steep slopes and sandy soil, late-harvested, spring-sown crops, such as maize and sugarbeet. It is lower where crops are drilled in early autumn and where minimum cultivation or direct seeding practices are used. Annual economic losses are estimated at around EUR 53 per hectare, while the costs of off-site effects on the surrounding civil public infrastructures reach EUR 32 per hectare (Torress, et al., 2001). In the United Kingdom, in grassland and arable regions, the timing of agricultural activities is as significant as considerations of cultivation practice, crop cover and soil type, in determining the scale and extent of soil erosion (McHugh, 2004). Approximately 70% of crops on arable soils are winter-grown, and therefore planted between August and December, when rainfall duration and intensity greatly increase the risk of erosion. For arable crop farming in Korea, soil erosion by water is mainly due to the concentrated rainfall in the summer season. Annual soil loss is only 0.02 t/ha in paddy fields, as compared to 32 t/ha in uplands (on slopes greater than 15%) and 0.9 t/ha for forest. Annual total soil loss in paddy fields is 22 768 tonnes, while in upland and forest areas it is 26 and 488 million tonnes, respectively (Hur, et al., 2004). Water-induced soil erosion is an important by-product of cereal production in Norway (Oygarden and Gronlund, 2004). Erosion occurs mainly in autumn or winter as the result of rainfall, snow melt and partly frozen soil conditions. Since 1993, threshold values for soil loss were 2 t/ha/year. In Switzerland, average soil losses during the 1998-2001 period decreased by 6% compared to the 1987-89 period (Prasuhn and Weisskopf, 2004). Between 1998-2001 around 20% of the arable land was affected by soil erosion every year, with an average soil loss of 0.7 t/ha/year. Significant damage associated with erosion was estimated for winter wheat. Threshold values were 4 t/ha/year. The 2001 National Resources Inventory (NRI) showed that soil erosion on cropland in the United States declined from between 2.8 billion tonnes per year in 1982, to 1.6 billion tonnes per year in 2001 (NRCS, 2003). Sheet and rill erosion dropped from 8.9 t/ha/year, to 6.1 t/ha/year, and wind erosion dropped from 7.4 t/ha/year, to 4.7 t/ha/year. Water-caused erosion dropped by almost 41% during this period, while wind erosion dropped by 43%. Between 1982 and 2001, cropland acreage eroding at excessive rates dropped by 39%.7 In 2001, 42 million ha of cropland were experiencing excessive erosion, down from 69 million ha in 1982. In 2001, about 72% of total cropland was eroding at, or below, the soil loss tolerance rate, up from AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 61

60% in 1982. Highly Erodible Land (HEL) cropland acreage declined from 50 million ha in 1982, to 41 million ha in 2001. The decline occurred in HEL acreage eroding at excessive rates, while HEL acreage eroding at acceptable soil loss tolerance rates increased slightly. Heavy concentration of both HEL cropland and high average erosion rates was located in cereal and oilseed producing areas such as the western plains, the western Corn Belt and in the Mississippi delta (Claassen, et al., 2004a).

2.1.2.

Nutrients

Loss of nutrients and organic matter from the soil can represent a loss of fertility which ultimately can affect crop yields and also pollute water bodies. Losses of phosphates from the soil are largely due to soil erosion. Nitrates originating from organic and inorganic fertilisers are particularly prone to leaching and the degree of losses resulting from arable crop production depends on the type farming system operated as well as on specific site characteristics. The quantity of nitrate loss from a particular farming system is determined largely by the balance between nitrogen inputs in the form of fertilisers, and nitrogen outputs from the farm in terms of harvested crops. It also depends on whether the farming system protects the soil from leaching during winter, by avoiding spreading of nitrogen fertilisers (organic or inorganic) on the land in this period and ensuring vegetation cover. Leaching of nitrogen can result from applications of mineral fertilisers at very early stages of crop growth, so that little is taken up by plants, or from the excessive application of fertilisers. However, in some regions much of the nitrogen lost from soil is associated with mineralisation of soil organic matter, normally during the period following the harvest or the ploughing of pasture for planting arable crops (Bloem, et al., 1994). Hoffmann, et al. (2000) estimated long-term changes in nitrogen leaching from cereals, grass and bare fallow for three different soil types in nine Swedish agricultural regions, covering a range of climatic conditions. They found that leaching of nitrogen was approximately the same in the 1860s as it was in mid-1980s. For cereals, in particular, both N input and N-uptake efficiency have exhibited upward trends. To gauge whether nutrients from arable crops pose an environmental risk, nitrogen balances for arable crops were calculated. A negative balance indicates that the amount of nitrogen removed from the soil through the harvested crop exceeds the amount of nutrient applied. Continued negative balances deplete nutrients in the soil, disrupt the soil ecosystem and can damage productivity (USDA, 2003b). Positive balances occur when farmers over-apply nitrogen.

62 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 2.1 displays application rates of nitrogen (N) on arable crops, the share of N consumed by arable crops and the potential environmental risk from nitrogen loss in arable crop farming, as measured by the nitrogen balance (e.g. greater than 50KgN/ha). According to these results, countries where both the risk and the application rates of N are estimated to be highest include Korea, Belgium and Denmark. The Netherlands has the largest potential risk, but is ranked fourteenth in terms of application rates. Poland, Canada, Australia and Turkey are estimated to have both the lowest risk and application rates.

2.1.3.

Waterlogging and salinisation

Waterlogging and soil salinisation have become important environmental concerns in some OECD regions. Waterlogging occurs as a result of a rise in the level of the water table, commonly caused by inefficient irrigation practices, such as inadequate drainage. The rise of the water table may also increase salinisation by drawing salt upwards from the lower soil horizons. Most arable crops do not tolerate salt and are seriously affected when salts concentrate within the root zone. The main impact of increasing soil salinity is loss of production, yields and income. Other on-farm effects include the decline in the capital value of land, salinisation of water storage, loss of farm flora and fauna, and loss of shelter and shade. These effects are propagated at the regional level, where they could have a significant impact on biodiversity, water supplies and infrastructure. It is estimated that moderate-to-severe salinity on agricultural land can reduce the annual yields of most cereal and oilseed crops by about 50% (McRae, Smith and Gregorich, 2000). In Australia, the incidence of soil salinisation is high on dry and irrigated land, predominantly in the Murray-Darling Basin and the south-western part of the country. In these areas, production of wheat is particularly affected. Around 30% of the grain farms in the west and 10% in the south of Australia are affected by significant dryland salinisation (AUDIT, 2001). It is estimated that in 2000 4.6 million ha of agricultural land in Australia were under a high risk of salinity hazard, and is projected that, unless effective solutions are implemented, the area could increase to 14 million ha by 2050. In the United States, some 5% of the cropland and pasture is affected by soil salinisation. Salinisation is also a problem in Turkey where it is associated with poor irrigation practices in some regions (OECD, 2001a).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 63

Table 2.1. Potential environmental risk from nitrogen in arable crops1

Country

United Kingdom Germ any Switzerland Ireland France Korea Belgium Norway Denm ark Italy Austria Czech Republic Portugal Netherlands Greece Sweden Spain New Zealand Hungary United States Japan Finland Slovak Republic Mexico Poland Canada Australia Turkey

Application rates of N on arable crops 2,3 (kg/ha)

Share of N consum ed by arable crops as % of the total am ount of N consumed by total 2,3 agriculture

Nitrogen balance (kgN/ha) 1995-97

156 140 135 124 114 112 111 104 104 103 103 101 94 94 91 90 89 86 82 80 78 74 71 64 59 56 37 32

48 63 57 8 32 46 37 33 72 62 80 87 60 12 41 61 59 6 87 88 35 53 86 58 55 86 72 31

87 61 61 79 54 253 181 73 115 30 27 54 63 262 33 34 44 6 4 32 135 64 45 n.a. 29 14 7 12

n.a. = not available. Notes: 1. Environmental risk is indicated where nitrogen surplus is greater than 50 kgN/ha. 2. As time series data for N by crop are not available, the most recent data from IFA/IFDC/FAO (2002) were used. Nitrogen balance data are from OECD (2003b) 3. Caution should be exercised in interpreting these results due to a number of data and methodological problems. Data on fertiliser use by crop types should be taken to reflect the general magnitude rather than the exact measurement. Mixed-cropping, for example, makes it difficult to estimate the amount used for each crop. On the other hand, with double-cropping, although the fertiliser is applied to one crop, both crops benefit. Moreover, some countries (e.g. Australia) make estimates for a group of crops (e.g. cereals, oilseeds) rather than individual crops. Sources: OECD Secretariat calculations, based on IFA/IFDC/FAO (2002), OECD (2003b); FAOSTAT.

64 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

2.2.

Water-related impacts Arable crop production can have environmental impacts on water through extraction for irrigation and pollution of watercourses with nutrients and pesticides.

2.2.1.

Water use

Agriculture is a significant user of water resources in many OECD countries. Large volumes of water are used annually in some regions for the irrigation of arable crops. Non-irrigated crop systems generally use significantly fewer inputs of fertilisers and other agro-chemical inputs. Different crops are subject to irrigation at varying levels of intensity. Wheat requires little irrigation except in arid and semi-arid regions. Maize requires relatively high levels of water during the early stages of growth, and in some regions cultivation relies heavily on irrigation. Water also plays a prominent role in the cultivation of rice. Paddy rice consumes more water than any other crop, but much of it is recycled and put to other uses. Certain rice cultivation practices develop water storage capacity and help to control flooding during heavy rains. In Australia, the agricultural sector is the most intensive user of water per unit of value created. Approximately 75% of Australia’s water is used in irrigated agriculture (AUDIT, 2001). The intensity of water use varies within and across states, due to climate, soil crop type and method of application. Generally, rice is the most water-intensive crop sector, with application rates varying between 11.9 and 13.9 ml/ha, followed by grapes. The intensity of water use for cereals and oilseeds is, on average, 3 ml/ha, as compared to 7 ml/ha for all irrigated land uses. In Europe, agriculture accounts for around 30% of total water use. The scale and importance of irrigation is significantly greater in southern areas of the EU, accounting for over 60% of water use in most countries. Within the EU, the main irrigated arable crops consist of maize and rice, particularly in France, Greece, Italy and Spain (IEEP, et al., 2000). In Portugal, the application rate for maize varies between regions from 3.9 m3/ha, to 6.6 m3/ha (Plano Nacional da Água, 2002). On the other hand, in the United Kingdom, in the mid-1990s cereals accounted for only 12% of total area of irrigated crops and around 5% of the volume of water used for crops (potatoes, sugarbeet, cereals, other crops grown in the open) (DEFRA, 1997). In Mexico, the total area planted for soybeans is irrigated, and for wheat and barley more than two-thirds of the area planted is irrigated. For maize, available evidence seems to suggest that the decline in maize production and AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 65

yield observed since the mid-1990s was in the higher-yield irrigated sector. Between 1995 and 2000, production on irrigated land declined by 31% from its 1994 peak, whilst average rain-fed maize was 18% higher than the average rain-fed production of the previous six years. Likewise, the area cultivated by the irrigated sector, which applies more pesticides, has significantly declined, while the rain-fed maize sector, which uses significantly less pesticides, has expanded (Dyer-Leal and Yúnez-Naude, 2003). In Korea, paddy fields take up about 77% of total water use in agriculture, 58% of which is used in irrigated paddy fields. Even though large areas of irrigated paddy fields have been converted for non-agricultural uses, the share of irrigated paddy field in total paddy field has increased steadily since 1970s (Hong-Sang, 2004). Irrigated agriculture accounts for an important part of the United States cropland sector, contributing almost half the total value of crop sales on just 16% of total cropland harvested. Over time, the mix of irrigated crops has changed. From 1969 to 1982, irrigated area increased for almost all crops, with the biggest gains in the major export grains (maize, soybean and wheat). Since 1982, there has been a general trend towards crops with higher value per hectare irrigated. Acreage of irrigated soybean, maize, horticulture and mint has doubled, while declines occurred in irrigated areas of sorghum, wheat, oats, barley, dry beans, pasture and un-harvested cropland. In 2000, around 280 000 farms irrigated 22.4 million ha of crop and pastureland (USDA, 2003b). Irrigated acreages in 2000 were substantial for several crops, including maize for grain (4.1 million ha, or 18% of all irrigated crops), wheat (1.3 million ha, or 6% of all irrigated crops), barley (0.4 million ha, or 2% of all irrigated crops), rice (7.7 million ha or 6% of all irrigated crops) and soybeans (2.1 million ha, or 9% of all irrigated crops). All of the rice-growing area is irrigated.

2.2.2.

Water pollution

For most OECD countries, nutrients, pesticide and soil sediments are the principal sources of water pollution associated with arable crop production. Inputs such as pesticides and nutrients can enter ground and surface waters, seriously affecting the quality of drinking water, and the cost of its treatment. Their presence in surface water can also have serious consequences for aquatic life. Greater impacts are associated with simplified, high-input arable systems. Nutrients, especially phosphates, cause eutrophication of water, which changes the ecological balance and can result in undesirable effects such as fish death and algal blooms. Problems are greatest where farming is intensive (Stoate, et al., 2001).

66 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Nutrient pollutants from arable crop production are comprised primarily of nitrogen and phosphates which reach water courses from the soil by leaching, surface run-off, sub-surface flow and soil erosion. Both nutrients can cause severe eutrophication of water. Arable farm systems are smaller sources of phosphate pollution than livestock systems. In the United Kingdom, for example, the UK Environmental Agency has estimated that agriculture is responsible for 43% of phosphates in surface water - 29% from livestock and 14% from fertiliser. In the United States, nutrient pollution is the most important cause of water quality impairment in lakes and the third-largest cause of river pollution. Phosphate pollution from arable crops production may be important in regions with low absorption-capacity soils, such as sandy soils, and in areas where phosphorus-demanding crops (e.g. maize) are grown. For example, in the United States, some evidence shows that the Corn Belt has a high potential for nitrate contamination of both groundwater and surface water from commercially applied fertiliser, and for phosphorus contamination of surface water the same source (USDA, 2003b). Whether nitrogen actually contaminates surface or groundwater depends on the amounts of nitrogen applied to agricultural land, the leaching characteristics of the soil, precipitation, crop type, timing of cultivation and on farming practices. Early ploughing of rape residues can lead to nitrogen leaching. Nitrates are particularly prone to leaching during the autumn, when nitrate passes through the root zone faster than the crop is able to exploit it, and also following the ploughing of grassland, when organic nitrogen is mineralised (Young, 1986). Leaching is greater under cereals than under permanent grass (Croll and Hayes, 1988), but can also be high under rotational set-aside (Meissner, et al., 1998). The likelihood of nitrate leaching is higher for spring-sown of cereals in northern Europe, unless cover crops, under-sowing or stubble regeneration are adopted. In contrast, nitrate leaching for autumn sowing is similar to winter cover crops (Boatman, et al., 1999). Pesticides reach water via surface run-off, through soil cracks and drains. Spray drift and acute pesticide pollution incidents can adversely affect aquatic organisms, as can the silt burden from eroded soil particles, which may also have phosphates and pesticides bonded onto their surfaces. Inappropriate cropping and cultivation techniques can exacerbate these problems. Pesticides may enter water from point-source contamination or from diffuse sources, following application to crops. The risk of pesticide pollution depends on its solubility, mobility in soil and rate of degradation. As with nutrients, rates of pesticide use over much of southern Europe are lower and pesticide pollution of water is less of a problem than in northern AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 67

Europe, but it does occur where intensively managed, irrigated crops, such as maize are grown. Some evidence suggests that in the Po Valley in Northern Italy in early 1990s, use of the herbicides atrazine and molinate on irrigated maize and rice caused contamination of local drinking water and led to a ban on their use in vulnerable areas (Boatman, et al., 1999). Ground- and surface-water vulnerability to pesticides varies geographically, depending on soil characteristics, pesticide application rates, and the persistence and toxicity of the pesticides used. Areas with sandy, highly leachable soils and high application rates of toxic or persistent pesticides generally have high vulnerability ratings for pesticide leaching. Areas with heavy soils and high application rates of toxic or persistent pesticides generally have higher vulnerability ratings for pesticide run-off. The relatively high levels of inorganic fertiliser used in the cultivation of rice may lead to the contamination and eutrophication of water. However, nitrogen leaching into surface water and groundwater from paddy fields is low compared to dryland crops and orchards, due to denitrification. Both lowland and upland systems make heavy use of pesticides. The draining of coastal wetlands for rice cultivation leads to the dehydration of soil, often causing sulphur to rise to the surface, with consequent acidification.

2.3.

Air quality Although arable crop production is not in itself a major source of air pollution, it can contribute to air pollution and climate change in a multitude of ways. Air quality concerns arising from arable crop farming include emissions into the air of greenhouse gases (GHGs), ammonia, wind-borne soil and other particulates (e.g. from burning crops). The focus of this report is on GHGs. The main arable crop activities which lead to airborne emissions include emissions of GHGs arising from the use of fertilisers, fossil fuel combustion (primarily through long-distance transport of arable inputs and products), wetland rice cultivation and the burning of crop residues. Burning crop residues in fields produces methane and nitrous oxide, while, of all arable crops, wetland rice cultivation is the principal source of methane (UNFCCC, 2003). On the other hand, production of biofuels from crops such as wheat and maize (for ethanol) and soybeans and rapeseed (for biodiesel) provide significant benefits for GHG reductions and air quality improvements (OECD, 2004d). Notwithstanding considerable uncertainty and lack of data, it is generally accepted that agriculture is an important contributor to emissions of three GHGs: carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4). Carbon dioxide emissions from agriculture occur primarily in areas

68 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

where land-use changes have taken place, or fuel use occurs; nitrous oxide, where there is crop cultivation using organic and inorganic fertilisers; while methane emissions are generally related to livestock and rice production. Most of the greenhouse gases result from intensive livestock rather than arable farming. As shown in Table 2.2, the contribution of arable crop production in agricultural GHGs is, on average, just over 10%, with considerable variation among countries (ranging from 1% in Switzerland to 49% in Japan). Table 2.2. Contributions of agriculture and arable crop farming to GHG emissions, 2001 Share of Share of Share of agriculture arable crops Share of wetland rice in total in agricultural agriculture cultivation in in total CH4 agricultural GHG GHG emissions emissions emissions CH4 emissions Australia Greece France Italy Japan Poland Portugal Spain Switzerland UK US

Share of N2O emissions from Share of arable crops agriculture in soil in total N2O agricultural N2O emissions emissions

%

%

%

%

%

%

20 8 20 8 3 8 14 12 10 7 8

2 26 2 10 49 4 9 5 1 2 11

62 33 68 50 67 24 55 58 66 42 27

1 4 0 8 43 0 3 1 0 0 5

81 59 68 55 57 68 73 66 72 64 74

26 17 20 24 7 43 10 24 24 19 16

Source: OECD Secretariat calculations, based on UNFCCC (2003).

Farming practices associated with arable crops such as tillage methods, soil protection, crop timing and rotation, crop selection and land use can all play a role in the emissions of CO2, N2O, and CH4 (OECD, 1998b). Agricultural soil is a major source of nitrous oxide emissions mainly originating from inorganic and organic fertilisers, while incorporation of crop residues, biological nitrogen fixation and cultivation of some soil also generate nitrous oxide emissions. Crop practices often affect the carbon content in the soils. Extreme differences can be found between wetlands and sandy soils. Wetlands can contain far more carbon than other types of soils. Changes in land use can affect the exchange of carbon between the soil carbon and atmospheric carbon dioxide. Arable crop production is the most important source of nitrous oxide emissions from agricultural soil in Switzerland, the United Kingdom, Poland and France, while emissions of methane from rice cultivation are AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 69

the most important source in Japan (Table 2.3 and Figure 2.2). Over 90% of methane produced by the cultivation of arable crops is caused by rice cultivation in all the countries listed, with the exception that of Poland. Emissions from burning arable crop residues in the field are important in Australia, Greece, Japan, Portugal, Spain and the United States (Annex Table 2.A1). Figure 2.2. Gross emissions of GHGs from arable crop farming, 2001 CH4 wetland rice cultivation

CH4 cereals residue burning

N2O emissions from arable crops soil

N2O cereals residue burning

NOx cereals residue burning

CO cereals residue burning

NMVOC cereals residue burning

100% 80% 60% 40% 20%

Sp ain Sw itz erl Un an ite d dK ing do Un m ite dS tat es

Po rtu ga l

Po lan d

Ja pa n

Ita ly

Fr an ce

Au str ali a Gr ee ce

0%

Source: OECD Secretariat calculations, based on UNFCCC (2003).

Overall, recent estimates show that rice cultivation accounts for a much smaller share of methane emissions than was previously believed. Although in most of the countries listed, methane emissions from rice cultivation increased during 1990-2001, they represent only a small share of CH4 emissions from agriculture, if Japan is excluded. In 2001, methane emissions from rice cultivation represented 43% of the methane emitted from all agricultural sources in Japan, although on average in OECD rice-producing countries, methane from rice represented less than 5% of agricultural methane emissions (Table 2.3).

70 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 2.3. Methane emissions from agriculture, 1990-2001 (1000 tonnes) 1990 Australia Agriculture Rice Share (%) France Agriculture Rice Share (%) Italy Agriculture Rice Share (%) Japan Agriculture Rice Share (%) Portugal Agriculture Rice Share (%) Spain Agriculture Rice Share (%) United States Agriculture Rice Share (%)

1995

2001

Annual Growth Rate (%)

3579.0 23.4 0.7

3413.0 30.9 0.9

3707.9 35.1 0.9

0.3 3.5

2185.0 8.6 0.4

2102.0 10.9 0.5

2087.5 8.5 0.4

-0.4 -0.1

913.8 73.3 8.0

901.2 81.4 9.0

871.1 74.0 8.5

-0.4 0.1

741.4 336.9 45.4

737.1 342.9 46.5

651.3 281.3 43.2

-1.1 -1.5

302.1 12.2 4.0

278.4 7.8 2.8

279.8 8.6 3.1

-0.6 -2.8

912.4 10.8 1.2

957.6 6.5 0.7

1120.6 14.0 1.3

1.7 2.2

7473.4 339.1 4.5

7972.4 362.8 4.6

7717.7 363.7 4.7

0.3 0.6

Source: OECD Secretariat calculations, based on UNFCCC (2003).

The amount of methane released from the cultivation of paddy rice depends on a number of factors, including water management during the growing season, soil characteristics – such as soil temperature and type – application of inorganic and organic fertilisers, and also other cultivation practices (Yagi, 1997). Long periods of submersion promote the aerobic decomposition of organic material and reduce the amount of oxygen in the AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 71

soil. As the oxygen is depleted, anaerobic decomposition by methanogenic bacteria begins. The resulting methane is partially released into the air through evaporation of water and transpiration of the rice plants. Other agricultural practices conducive to GHG reduction from rice cultivation include lowering the levels of organic fertiliser used, reducing the amount of crop residue left in the paddy fields and increasing the use of varieties of rice that emit low levels of methane.

2.4.

Biodiversity Arable crop farming can affect biodiversity and landscape in several ways. In particular, factors such as cropping systems, field size, use of agro-chemicals, drainage and irrigation can influence habitat and farmland species. Increased intensification and specialisation of arable cropping is characterised by significant economies of scale, which could trigger declines in diversity of habitats and in farmland species. In some regions, particularly in Europe, with farm amalgamation, many rotations have been simplified so that crops such as wheat or maize may be grown continuously without any breaks, which often requires higher applications of fertilisers and pesticides and increases the erosion risk. Moreover, increased monocultures and reduction in the number of mixed arable and livestock farms have led to loss of biodiversity and created a less diverse landscape (Baldock, Dwyer and Vinas, 2002; Boatman, et al., 1999). Increased drainage and irrigation have also caused habitat degradation in many areas where irrigation of crops (e.g. maize) is usually associated with increased fertiliser and pesticide applications. In contrast, the cultivation of rice can increase the local diversity of birds and the aquatic invertebrates on which they feed. Paddy fields can play a particularly valuable role in the conservation of wetland wildlife, including breeding, wintering and migratory birds, where rice is grown close to estuary habitats. The seasonal wetland habitat provided by flooded paddy fields also supports a number of ecosystems, including many species of birds and small mammals. Rice fields also host many natural enemies or predators, which provide a mechanism to control harmful insects and pests, and thereby reduce the need for pesticides. In some OECD countries such as in Japan, rice production is considered the single most important factor of “multifunctional” agriculture (Nakashima, 2001). On the other hand, the introduction of upland rice production can result in deforestation on marginal, steep hillsides, whilst lowland systems are often extended at the expense of coastal wetlands and mangrove swamps, with the consequent loss of habitats and destruction of ecosystems. Further, chemicals,

72 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

agricultural run-off, sedimentation and other forms of pollution could accumulate in rice fields and cause environmental damage and loss of plant and animal species. A number of biodiversity indicators has been established by OECD within the general framework of genetic, species and ecosystem diversity (OECD, 2001a). According to OECD work on Agri-environmental Indicators, the number of new crop varieties has increased between the mid-1980s and the mid-1990s. This work also suggests that the trend in the share of one of the top five dominant varieties in the total marketed production for certain arable crops (i.e. wheat, barley, maize and soybeans) increased for many OECD countries. Wetterich (2003) reports increasing diversity in Germany in terms of the number of registered varieties of maize and wheat over the 1992-2000 period. McRae and Weins (2003) found positive trends in Canada for wildlife habitat in cropland (land used for grains, oilseeds, fruits, nuts, vegetables, tames hay). Scott (2003) calculated changes in stock and condition of habitats for the United Kingdom over the 1990-98 period. It was found that for arable crops and horticultural farming there has been little change in the stock, but unfavourable trends have appeared in the condition of wildlife habitat. Some other studies have calculated changes in biodiversity using farmland bird indicators as a proxy. These studies, which are mainly for Europe, found declining trends, especially in the United Kingdom. Heath and Rayment (2003), for example, report that although the number of common birds has remained stable in the United Kingdom since 1970, the variety of farmland species has declined.

2.5.

Management practice approaches to reduce environmental impacts of arable crop production The improvement of arable crop yields described earlier stems to a great extent, from changes in agricultural practices and techniques. Few practices have remained unaltered by the increased intensification and modernisation of arable crop production. Tilling, sowing and harvesting have become increasingly mechanised, and application of chemicals has become more sophisticated. Contemporary agricultural practices – such as monoculture or the continuous production of row crops, fewer rotations with forages, shorter rotations, intensive tillage, inappropriate fallowing and crop residue management, and the cultivation of marginal lands – are often held responsible for many of the adverse environmental effects of arable crop farming discussed in the preceding section. This section will endeavour to

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 73

discuss those farming practices which are deemed to be benign to the environment. Various approaches have been developed over the past 15 years to minimise the environmental effects of agricultural production. Among the foremost of those concerning arable crop farming are Soil Management and Conservation Systems, Integrated Plant Nutrient Systems and Integrated Pest Management. These practices are interrelated and may be substitutes or complements, but they are treated separately here, as far as possible.

2.5.1.

Soil management and conservation systems

Awareness of the need for protection of the soil resource is increasingly on the research agenda and also the wider political agenda. Different combinations of crops, rotations and tillage practices may have different impacts on soil and water quality. Decisions on crop selection, rotation and tillage can affect the risk of erosion, compaction, salinisation and nutrient loss (OECD, 1994). These choices are also likely to affect water quality. Concentrations of wildlife may also be affected, as different crops and tillage methods provide different levels of habitat. Large shifts in crops and tillage practices can also affect emissions. Research on a wide range of agricultural husbandry systems and techniques has revealed direct beneficial implications in mitigating impacts on water quality. For example, the use of contour cultivation, or minimum tillage, silt traps, cover crops, the technical application of fertiliser, and riparian buffer zones can significantly reduce sediment and fertiliser run-off losses from arable cropping activities. A United Kingdom survey shows that nitrogen surpluses for winter wheat have dropped from 70 kgN/ha/year in the early 1980s, to around 25 kgN/ha/year in the late 1990s due to improvements in crop protection, plant breeding and agronomy (DEFRA, 2002).

2.5.1.1. Rotational cropping systems Different land uses have different effects on natural resources. Generally speaking, annual cropping is the most disruptive type of land use and, depending on local soil conditions, it may reduce surface and groundwater quality. It also tends to provide less wildlife value. On the other hand, perennial forages, improved pasture, and native grassland or woodland are less disruptive. Cropping systems which involve crop rotation could reduce the environmental risk posed by arable crops because they affect soil conservation, soil fertility and pest control. For example, row crops on erodible soils can be rotated with soil-conserving crops to reduce soil loss.8

74 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Closely sown row grain crops such as wheat, barley and oats provide additional vegetative cover to reduce soil erosion and add organic matter. As such, these crops tend to be less erosive than rapeseed, which is in turn less erosive than wide-row crops, such as maize and sunflowers (AAFC, 1996). Wide-row crops are associated more with soil degradation, silt and nutrient infiltration to surface water, and the leaching of nutrients and pesticides to groundwater (USDA, 2003b). Rotations that include forages, green manure and winter cover crops tend to erode less and improve soil quality. Rotations that include tilled summer fallow may raise the risk of salinisation and erosion. In the United States, rotational cropping of arable crops is predominant with soybeans and maize. Most rotational cropping of maize and soybeans alternates, while winter wheat rotates with a row crop and small grains, and fallow. About 60% of the acreage in maize and soybeans and 40% of winter wheat were rotated in 1999 (USDA, 2003b). Because maize production leaves more residue after harvesting than soybeans, a maize-soybeans rotation reduces soil erosion to a greater extent than continuous soybeans (although to a lesser extent than continuous maize). Over time, rotating maize with other crops, particularly soybeans, has increased. Empirical studies in the United States found that crop rotations were associated with higher yields than those achieved with continuous cropping under similar conditions. For example, in 1996 returns to maize averaged 5% to 51% higher, depending on the region, when in rotation with soybeans rather than in continuous maize production (USDA, 2003b). However, agricultural support policies could be an important impediment to the adoption of crop rotational cropping systems. For example, while farmers may be able to increase nitrogen to crops and decrease susceptibility to pests and diseases through crop rotations with leguminous crops, they may be able to earn greater profits through monocultures of crops. For example, in the United States maize grown in rotation with soybeans received deficiency payments was generally less profitable for farmers than continuous maize production in Iowa and Nebraska (Hrubovcak, et al., 1999). The amount of cover and residue left on the soil also affects soil quality and productivity and alters the effects of the soil on environmental quality. Cover crops are a management option to reduce nitrate leaching under cereal grain production. Soil organic matter in agricultural topsoils, derived from crop residues, organic manures, microbial biomass and soil microflora and fauna, plays a key role in maintaining soil quality, structural stability, and water-holding and buffering capacity. Crops that provide a high level of ground cover tend to have lower erosion rates compared to other crops. A AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 75

cover crop of small grains, meadow, or hay planted in the autumn after harvest of a row crop provides vegetative cover to reduce soil loss, hold nutrients and add organic matter to the soil. Except for winter wheat, the cover crop is usually not harvested, but is sometimes grazed by livestock. A study undertaken in Sweden on the effects of rye-grass cover crops on nitrate leaching in spring barley found that rye-grass cover reduced leaching by two-thirds in the first year and by more than 50% over a two-year period (Bergström and Jokela, 2001). Soil residue cover provided by arable crops depends on tillage practices. For example, in Canada, the highest soil cover is provided under no-till and the lowest is produced under conventional tillage. Conservation tillage is associated with medium soil cover for maize, rapeseed and soybeans, and high for wheat, barley and oats (AAFC, 1996).

2.5.1.2. Tillage practices Tillage systems are defined by the amount of crop residue remaining on the soil after the previous crop has been harvested. In the United States, conventional tillage leaves a maximum of 15% of the previous crop residue covering the soil, whereas conservation tillage maintains a maximum 30% of the previous crop residue covering the soil. The adverse effects of conventional tillage practices (such as ploughing) on farm productivity and on the environment are being increasingly recognised (EEA, 2003b). The recurring disturbance of topsoil buries any soil cover and may destabilise the soil structure so that rainfall can cause soil dispersion, sealing and crusting of the surface. It often results in compacted soil which, in turn, negatively affects productivity. In response to these problems, conservation tillage practices have been developed in a number of OECD countries. Conservation tillage reduces soil erosion and the risk of soil salinisation, and has the potential to improve surface-water quality (Derpsch, 2000; Pieri, et al., 2002). It maintains and improves crop yields and resilience against drought and other hazards, while at the same time protecting and stimulating the biological functioning of the soil. Studies in the United States found that pesticide use on maize, soybeans and wheat differs among tillage systems and it is difficult to distinguish the effects related to tillage systems from differences in pest populations between areas and from one year to the next, and from use of other pest control practices (USDA, 2003b). The study by Caswell, et al., 2001, which is based on a detailed field-level survey across the US, found that tillage choice had no effect on yields for soybeans and maize.

76 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Almost half of the area worldwide where conservation tillage practices have been applied is in the United States, although a considerable share of this is under monoculture. Adoption of conservation tillage has also increased over time. In the United States, for example, farmers employ conservation tillage practices on over 36% of planted area to maize and 56% of planted area to soybeans in 2000, compared with 30% in 1990 (Table 2.4). Table 2.4. Adoption of alternative tillage practices in the United States, 1990-2000 1990

1995 (%)

1997

2000

Conservation tillage No-till Ridge-till Mulch till Non-conservation tillage Reduced-till Intensive-till

32.3 8.7 2.6 21.0 67.7 24.4 43.3

41.3 18.1 3.1 20.1 58.8 22.6 36.2

41.5 17.5 3.1 20.9 58.5 24.2 34.3

36.5 17.9 2.1 16.5 63.5 23.2 40.3

Conservation tillage No-till Ridge-till Mulch till Non-conservation tillage Reduced-till Intensive-till

30.4 9.6 1.4 19.4 69.6 24.2 45.4

50.4 30 1 19.4 49.6 20.8 28.8

53.6 30.5 1 22.1 46.4 20.2 26.2

56.1 32.8 0.9 22.4 43.9 18.8 25.1

Conservation tillage No-till Ridge-till Mulch till Non-conservation tillage Reduced-till Intensive-till

24.4 3.0 0.0 21.4 75.5 30.4 45.1

31.2 6.6 0.0 24.6 68.7 33.7 35

32.2 8.3 0.1 23.8 67.9 35 32.9

30.4 9.8 0.1 20.5 69.6 27.1 42.5

Corn

Soybeans

Small grains

Source: USDA (2003b).

The trend towards adoption of conservation tillage, and the corresponding decline in intensive tillage, is attributable to many factors including the prospect of higher economic returns with conservation tillage and by government policies and programmes promoting tillage for its conservation benefits. Higher economic returns resulting from conservation tillage stem primarily from increased or stable crop yields and an overall reduction in input costs, with both heavily dependent on the characteristics of the resource base and appropriate management. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 77

Farm size and cropping practices affect the likelihood of farmers’ adopting soil conservation and tillage practices. According to the ERS/USDA study (Caswell, et al., 2001) farm size and cropping practices, especially crop type and use of crop rotations proved to be important determinants in the adoption of till conservation practices. However, the most important determinant was the influence of policies concerning areas such as conservation compliance and technical assistance.

2.5.2.

Nutrient Management

Any method of crop production – extensive or intensive, conventional or organic – removes plant nutrients from the soil. Nutrient uptake varies according to the type of soil and the intensity of production. An increase in biomass production results in a higher plant nutrient uptake. As mentioned in earlier sections, the major nutrients required by arable crops are nitrogen, phosphate and potash. Enhanced nutrient management aims to optimise the uptake of plant nutrients by the crop and thereby increase productivity. It involves efficient use of nutrients from commercial fertilisers and animal wastes. Enhanced nutrient management practices include improving existing practices in regard to assessing nutrient needs and the timing of applications, placing fertiliser closer to the seed, using alternative products, changing crop and irrigation management, and using manure and organic wastes. Nutrient management practices may have a significant effect on nitrogen fertiliser use and crop yields. OECD countries use a wide range of nutrient management practices to enhance fertiliser use efficiency and reduce nutrient losses into environment. These practices, inter alia, include: assessing nutrient need through regular soil and crop tissue testing before applying nutrients; timing nutrient application to tailor feeding to crop-growth needs; applying nutrients close to the root zone; selecting the nutrient product according to the soil’s chemical stability; rotating nitrogen-using with nitrogen-fixing crops; using nitrogen inhibitors and other products to retard the release of nitrates from ammonium fertilisers until later in the growing season; and applying manure and organic waste based on nutrient management plans. Soil nutrient tests are carried out in almost all OECD countries. In Australia, the focus has shifted from broad regional fertiliser guidelines to site-specific nutrient management.9 In the United States, results from the 1996 USDA Agricultural Resources Management Study survey of maize farmers indicate that soil tests were the most extensively used (44% of maize acreage), whilst nutrient-testing techniques were used only modestly. Numerous studies have examined the factors determining the adoption of

78 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

nutrient management systems. A survey of the literature suggests that these are both regional and practice-specific (Christensen, 2002). Adoption depends on the method of farming in the region (e.g. irrigated or not), the type of soil, and the presence of regulation. Moreover, some tests, such as manure testing, may more commonly adopted by livestock farmers.10

2.5.3.

Integrated Pest Management

Arable crop production systems suffer losses caused by diseases, weeds, insects and other pests. The goal of integrated pest management is to avoid or reduce yield losses by pests, while minimising the negative impacts of pest control through the application of the most appropriate pest control methods. Under the system of integrated pest management, the presence and density of pests and their predators and the degree of pest damage are systemically monitored.11 Pest management practices include biological controls, cultural controls (including crop rotation and strategic controls such as planting dates and location) and the use of pest-resistant plans. Integrated pest management can reduce the need for pesticides, which can also have a beneficial effect on the quality of groundwater. Unfortunately, quantitative evaluations of the uptake of integrated pest management in terms of hectares covered and reduction in pesticide use is only available for a few projects, making generalisation difficult. Integrated pest management has been introduced in many countries and for many different arable crops. According to FAO, worldwide integrated pest management applied to rice has shown significant improvements in production, in some cases simultaneously reducing costs (FAO, 2003). In the United States, farmers have used integrated pest management for more than 20 years (Hrubovcak, et al., 1999), but many of the techniques under the umbrella of integrated pest management have been used for some considerable time, the large-scale adoption of integrated pest management elements is a relatively new phenomenon. Farm structure, including human capital, is an important factor in the adoption of integrated pest management. Studies in the United States have found that human capital and farm size had a positive effect on the uptake of modern integrated pest management technologies (Caswell, et al., 2001). On the other hand, human capital had a negative impact on the use of the more traditional pest management strategy of destroying crop residues and farm size had no influence on the use of traditional pest management strategies of crop rotation and crop residue destruction. Cropping practices, especially crop choice and use of irrigation significantly affected the use of all of the pest management practices that were analysed. Moreover, natural endowment was found to be important in explaining farmers’ use of AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 79

traditional pest management technologies, but not the use of integrated pest management. Large farms are more likely to adopt integrated pest management than smaller farms. The availability of operator and unpaid family labour was found to be associated positively with integrated pest management adoption.

2.5.4.

Organic farming practices

Organic farming is a method of production comprising a range of land, crop and animal management systems. It is based on minimising the use of synthetic chemical inputs such as fertilisers, pesticides, additives and medicinal products and represents a deliberate attempt to make the best use of natural resources. Organic agriculture is circumscribed by a set of rules enforced by inspection and certification mechanisms. Organic farming generates less stress for the environment than conventional agriculture, in terms of lower pesticide residues and soil erosion, increased biodiversity and resilience to drought (OECD, 2003a; FAO/WHO, 1999). Organic farming systems also have the potential to lower nutrient run-off and reduce greenhouse gases. There is evidence to suggest that organic farming and no-till are more effective in reducing soil erosion than conventional farming practices and, therefore, in maintaining soil productivity (Loucks, 2003). However, the overall long-term effects of organic methods of food production on the sustainability of agriculture require more investigation. Although the environmental costs of organic systems are generally lower than those of conventional farming, their unit production costs are higher. Compared with conventional farms, organic yields on a given area of land are often lower and more variable (OECD, 2003a; FAO, 2003). In such cases, a significant expansion of organic farming could mean more land under cultivation, which may have an alternative value in terms of its potential use, depending on its current and historical use. From the perspective of potential environmental impacts on the arable crop sector, an expansion in crop production will have immediate impacts on land use and land-use change. The extent of the change in land use depends on the type of crop and the method of crop production introduced. However, yields might be improved if agricultural research were to place greater emphasis on organic farming. Any comprehensive assessment of the value of different farming systems needs to take account of the relative economic, social and environmental costs and benefits of these systems in terms of varying yields, soil and water depletion, pollution, landscape, wildlife habitats, and animal and human health. Organic farming systems for arable crops include practices such as organic fertilisation, manipulation of crop rotations and strip cropping,

80 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

biological pest management and composting. Soil fertility and crop nutrients are managed through tillage and cultivation practices, crop rotation, and cover crops, supplemented with manure and waste material from crops and permitted synthetic substances. Crop pests, weeds and diseases are controlled through physical, mechanical and biological control management methods. Crops produced by organic grain and oilseed farmers include traditional grains and oilseeds such as maize, soybeans, wheat, barley, oats and rice, as well as non-traditional grains, including millet, buckwheat, rye and spelt. Organic agriculture is practised in almost every country in the world, and its share of agricultural land, farms and production has accelerated in recent years. This shift has been encouraged by changes in consumer demand. Moreover, in some OECD countries, particularly in Europe, government support has been instrumental in the development of organic farming. The share of farm area accounted for by organic agriculture varies considerably in OECD countries, from under 0.2% in Japan, Korea and Mexico, to over 10% in Austria (Table 2.5). For arable crops, as depicted in Table 2.5, there is considerable variation between countries, ranging from less than 1% of area harvested under arable crops in the majority of countries, to 6% in Austria. Austria has the highest share of land under organic arable production, followed by Finland and Italy (4%). In absolute terms, the United States has both the largest organic area devoted to arable crops as well the largest number of organic farms, followed by France. In the EU, major growth of the organic farming sector has taken place in the last decade, following the implementation in 1993 of EC Regulation 2092/91, defining organic crop production. The widespread application of policies to support conversion to, and maintenance of, organic farming has been ensured by Regulation 2078/92 in the framework of the agri-environmental measures (see Chapter 4). Land area under organic arable crops production has more than tripled in the EU since the early 1990s (Foster and Lampkin, 2000). In Australia, rice is one of the most important organic crops. In Canada, organic grain production is the fastest-growing organic sector. In Korea, the market for organic products is still very small. In 2001, locally grown organic produce, comprising rice, fruits and vegetables, accounted for only 0.2% of total agricultural production. In Mexico, soybeans are amongst the most important organic crops. In the United States, organic farming has been one of the fastest-growing segments of US agriculture for nearly a decade (Dimitri and Greene, 2002). Certified organic cropland for maize, soybeans and other AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 81

major crops more than doubled from 1992-97, and doubled again between 1997-2001. Even so, less than 1% of maize, soybeans and wheat were grown under certified organic farming systems in 2001. Table 2.5. Arable crop area under organic farming, 2001 Arable Crops Sector (Cereals and oilseeds)1

Country

Number of % of total organic arable farms crop farms Australia Austria3 Belgium Canada Czech Republic Denmark Finland France Germany Greece Hungary Japan2 Korea2 Iceland Ireland Italy Luxembourg Mexico Netherlands2 New Zealand Norway Poland Portugal Slovak Republic Spain Sweden Switzerland Turkey United Kingdom United States

7804

7.0

4600 879

0.3

Organic hectares2 (1000)

6.0

19

1.0

51 78

4.1 0.7

4

0.3

3.5

12

19 16

1414

Number of % of organic arable farms crops area

77

250

576

Total Agriculture

4.3

0.2 3.0

5

2.2

57 299

1.7 0.3

1380 18292 694 3236 654 3525 4983 10364 14703 6680 1040 1237 27 997 56440 48 34862 1507 983 2099 1787 917 82 15607 3589 5441 18385 3981 6949

% of ALL farms 1.4 9.3 1.0 0.6 2.4 5.6 6.4 1.6 3.3 0.8

0.8 0.7 2.4 1.6 0.1 1.6 3.1 0.1 0.2 1.3 4.0 7.9 0.1 1.7 0.2

Organic hectares (1000)

% of total area

10500 276 22 431 218 175 148 420 632 31 105 5 1 5 30 1230 2 143 38 63 27 45 71 59 485 194 94 57 680 950

2.3 11.3 1.6 0.6 5.1 6.5 6.6 1.4 3.7 0.6 1.8 0.1 0.0 0.6 0.7 7.9 1.7 0.1 1.9 0.4 2.6 0.3 1.8 2.4 1.7 6.3 8.7 0.1 4.0 0.3

Notes: 1. For the Czech Republic: arable land; Finland: includes dried pulses; France: includes protein plants; United Kingdom: includes other crops. 2. The data for Japan refer to 1999, for Korea to 1998 and for the Netherlands to 2002. 3. Data from IACS. Sources: Foster and Lampkin (2000); Yussefi and Willer (2003); USDA/ERS; Delegations.

82 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

2.5.5.

Factors influencing adoption of environmentally benign farming practices

The environmentally friendly practices and technologies described above are interrelated and complementary, seeking to meet the dual goals of increased productivity and reduced environmental impact. Yet, experience today suggests that, despite their higher rate of returns, wide-scale adoption has not yet occurred across OECD countries. There are several reasons for the continuing dominance of conventional farming practices. Each of the environmentally benign practices is “information- and management-intensive”, because a farmer is required to have a thorough understanding of how the physical characteristics associated with farming, such as soil type, rainfall and temperature, interact with inputs such as pesticides, nutrients and soil, to affect crop production. Each practice uses inputs efficiently and may dramatically affect farm profits, the quality of the environment, and the pattern of natural resources (Hrubovcak, et al., 1999). While decisions on the amount of conventional inputs to apply are made on a seasonal or annual basis, the adoption of new technologies entails extra costs for tools and equipment, and requires complex management skills. For example, production systems that include crop rotation are more complex, they require coherent management over the longer term. The adoption of information-intensive technologies requires a certain level of educational attainment on the part of the farmer. Evidence from the United States reveals that small grain farms are generally operated by older and less educated farmers than their counterparts on larger farms. Moreover, larger grain farms are more likely to use risk management strategies, conservation or no-till systems than operators of small farms. However, larger maize farms are likely to irrigate maize and to make heavier use of chemical inputs (Foreman, 2001). The overall policy framework is also an important determinant of the type of environmentally benign practices adopted and their rate of uptake (OECD, 2001b). For example, commodity programmes that restrict base acreage to one or two crops could be an important impediment, as they encourage monoculture or the continuous planting of the same crop. In the EU, cuts in the compulsory set-aside rate brought about by the 1992 CAP reforms have encouraged some increase in the areas under cereal cultivation. In the United States, policy changes brought about by the 1996 FAIR Act, including elimination of set-aside requirements, changes in prices and loan deficiency payments (LDP) led to some farmers transferring land previously used for maize production to the production of other crops, mainly soybeans or rotations with other crops (Lin, et al., 2000). Farmers also adopted conservation tillage partly in response to incentives associated with conservation compliance provisions of the 1985 Food Security Act (FSA) (see Chapter 5). AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 83

The fact that the nexus of environmental benefits-profitability exhibits spatial variation could be another factor hindering the adoption of environmentally benign technologies and farm practices. A given technology may be appropriate in one region, but inappropriate for another. Further, there could be environmental trade-off associated with the adoption of new technologies, as controlling one type of problem might exacerbate another (for example, it is possible that conservation tillage may reduce soil erosion, but increase herbicide use). The costs and benefits of conservation tillage vary according to farm and location. Studies in the United States comparing profitability of conservation and conventional tillage systems produced mixed results. Studies at the regional level for wheat found that higher yields resulted with conservation tillage than with conventional tillage in semi-arid areas (see Hrubovcak, et al., 1999, for more discussion).

2.6.

Transgenic crops12 and the environment The main purpose of this section is to summarise the current commercial status of transgenic crops and to identify some of the main environmental issues associated with them. It is not intended to provide an exhaustive overview of the “GMO debates”.13

2.6.1.

How widespread are transgenic crops?

The first transgenic crops became commercially available in the mid-1990s. Since then, their uptake has been rising. During the period from 1996 to 2003 there was a large increase in the area grown with transgenic crops worldwide, from 1.7 million ha in 1996 to 67.7 million ha in 2003 (Figure 2.3). So far, adoption has been uneven across countries and commercialisation has involved only a few crops and traits. In 2003, two-thirds of the transgenic crop area worldwide was found in developed countries. Six countries, four crops (soybeans, cotton, maize and rapeseed) and two traits (insect resistance and herbicide tolerance) account for almost the totality of global transgenic crop area. The United States grew 63% of the global total, followed by Argentina (21%), Canada (6%), Brazil (4%), China (4%) and South Africa (1%). In addition to the producing countries, many others have approved importation of transgenic crops for domestic consumption. In the EU, for example, 18 GMOs are approved for marketing, including amongst others GM maize, GM soy and rapeseed oil.

84 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 2.3. Global area of transgenic crops, 1996-2003 B y co u n try U n ite d S ta te s

A rg e n tin a

C anada

C h in a

T o ta l

(mill. ha)

80 70 60 50 40 30 20 10 0 1996

1997

1998

1999

2000

2001

2002

2003

B y cro p S o yb e a n s

M a ize

C o tto n

R apeseeds

45 40

(mill. ha)

35 30 25 20 15 10 5 0 1996

1997

1998

1999

2000

2001

2002

2003

B y trait H e rb ic id e T o le ra n ce (H T )

In s e c t R e s is ta n c e (B t)

B t/H T

V iru s R e s is ta n c e

60

(mill. ha)

50 40 30 20 10 0 1996

1997

1998

1999

2000

2001

2002

2003

Source: James (2003).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 85

Globally, most of this area is divided among four crops: soybeans (61%), maize (23%), cotton (11%) and rapeseed (5%). Of these crops, 55% of soybean acreage, 21% of cotton, 16% of rapeseeds and 16% of maize was transgenic in 2003. The uptake has been more rapid in the United States, growing from zero in 1996 to approximately 80% of soybean, 70% of cotton, and 38% of maize acreage being planted with transgenic varieties in 2003 (USDA, 2003d). Transgenic rapeseed is planted in two countries (Canada and the United States). Currently, there are three main types of traits used in commercial cultivation: herbicide tolerance; insect resistance; and virus resistance. Insect-resistant transgenic crops are used as a way of controlling specific pests. Insect-resistant crops have been developed by integrating genes derived from various strains of a bacterium Bacillus thuringiensis (Bt), which produces toxins that kill certain insect pests, for example, the European maize borer and the Southwestern maize borer. Insect-resistance genes have been introduced in maize and cotton. For herbicide-tolerant traits, the insertion of a herbicide-tolerant gene into a plant enables farmers to spay wide-spectrum herbicides on their fields to control weeds without harming the crop. Herbicide tolerant crops include soybean, maize, rapeseed and cotton. Virus resistance genes have been introduced in tobacco, potatoes, papaya and squash. Transgenic crops have also been developed which involve two or more traits (e.g. stacked events). The most common stacked events at present are combinations of herbicide tolerance (HT) and insect resistance (e.g. Bt). During the 1996-2003 period, herbicide tolerance has consistently been the dominant trait introduced, followed by insect resistance. Seventy-four percent of all transgenic crops in 2003 were herbicide tolerant, 18% insect resistant and a further 8% contained both these traits. HT soybean was the most dominant transgenic crop grown commercially (occupying 41.4 million ha or 61% of the global total), followed by Bt maize (13%) (James, 2003). OECD’s Product Database (http://www1.oecd.org/scripts/biotech/) has information on most transgenic crops which have been approved for commercial use in OECD member countries. Despite the focus of this discussion on the relatively small number of transgenic crops which have been commercialised so far, it is important to note that there is an impressive range of crops and traits in research and development, many of which have already been in field trials. Many of these are likely to be commercialised in the near future. It takes around a decade for a new transgenic crop variety to be developed from the field-trial stage to commercialisation. Arable crops in the pipeline include soybeans with improved animal nutritional qualities through increase protein and amino acid content; crops with modified oils, fats and starches to improve

86 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

processing and digestibility, such as high stearate canola, low phytate or low phytic acid maize.

2.6.2.

What are the environmental implications?

The environmental impact of transgenic crops may be either positive or negative. They may accelerate the damaging environmental effects of agriculture or contribute to more sustainable agricultural practices and the conservation of natural resources, including biodiversity depending on how and where they are used. Releasing transgenic crops into the environment may entail risks such as gene transfer to wild relatives or conventional crops, weediness, trait effects on non-target species and other unintended effects. These risks are similar for transgenic and conventionally produced crops. Although scientists differ in their views on these risks, there appears to be an agreement on the need that environmental impacts should be assessed on a case-by-case basis and regularly monitored. Transgenic crops may also entail positive or negative indirect environmental effects through changes in agricultural practices such as pesticide and herbicide use and cropping patterns.

Main environmental benefits The increasing cultivation of transgenic crops could contribute to more sustainable agriculture. Transgenic crops have been developed in order to increase the value or reducing the costs of producing crops. In addition to market effects, there could also be positive environmental impacts, depending on the crop and trait under consideration. These benefits include use of environmentally benign methods for managing weeds and insect pests due to smaller use of chemical inputs, thereby conserving biodiversity. Table 2.6 provides a snapshot of potential environmental benefits of transgenic crops, while Box 2.1 discusses the findings of selected empirical studies. Productivity gains encompass higher returns on all factors of production or lower input requirements per unit of production. This could lead to higher crop yields (due to the presence of fewer insects or pests), lower pesticide and fertiliser applications, less demanding production techniques, higher product quality, better storage and easier processing. These gains should be assessed in comparison with conventionally produced crops, produced under the same production system. Ultimately, higher productivity may result in lower producer and consumer prices. Moreover, the reduction in production cost has the potential to raise rural incomes in developing countries in a similar way to the Green Revolution in large parts of Asia during the 1960s to 1980s (FAO, 2003).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 87

Table 2.6. Potential environmental benefits of transgenic crops

Characteristics

Rationale

Examples

Productivity enhancements

Higher output per unit of land

High-yielding rice and maize

Herbicide tolerance

More efficient herbicide use and/or safer herbicide use

Glyphosate-tolerant soybeans, canola, maize

Disease/insect tolerance

Reduction in pesticide use and/or more efficient pest control

Bt cotton, maize, potatoes; virus resistant papaya, tobacco, melon

Tolerance to biological stresses

Improved resistance to droughts, easier production in marginal areas, easier nitrogen fixation

Research on droughttolerant maize

Source: Nelson and de Pinto (1999; 2001).

Changes in pesticide use associated with the production of transgenic crops have been considered as an important possible impact (Royal Society, 1998; Ervin, et al., 2000; Fernandez-Cornejo and McBride, 2002; Wolfenbarger and Phifer, 2000). Transgenic crops could lead to a reduction in the use of environmentally harmful chemicals to control weeds and pests because certain pesticides are no longer used, the frequency of treatments is reduced, or the area treated is reduced. Due to higher yields, transgenic crops might lower pressures on land resources and diminishing the need for clearing the land or for land conversion, thereby leaving more area available for habitat protection and preservation. In the future, transgenic crops might become available that are resistant to drought (thereby saving water). Salinity-resistance of the soil could contribute towards the continuation of agriculture in regions affected by this phenomenon, which is primarily linked to irrigation.

88 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Box 2.1. What does the empirical evidence show? Several studies have attempted to assess non market benefits and impacts associated with transgenic crops (e.g. an annotated bibliography can be found at: www.isb.vt.edu). However, they are non conclusive, partly because of the novelty of such crops, because some of these crops have been grown for a short period and there are different approaches as to what should be the benchmark of comparison. Overall, available empirical evidence tends to suggest that yields are somewhat higher with transgenic crops than with their conventional counterparts, although there is significant variation by crop, location and year. The National Center for Food and Agricultural Policy, which estimated the impacts of nine transgenic crops in the EU, found that collectively the nine transgenic crops have the potential to increase yields by 8.5 million tonnes per year, increase grower net income by USD 1.6 billion per year and reduce pesticide use by 0.014 million tonnes per year. Transgenic tomato would offer the greatest yield and grower income increase, while herbicide tolerant maize would have the largest reduction in pesticide use. The largest increase in yields is estimated for transgenic sugarbeet, whereas for glyphosate tolerant maize, wheat and rice yields would be unchanged (Gianessi, Sankula and Reigner, 2003). Traxler (2003) found that yields of glyphosate tolerant soybeans are not significantly different from yields of conventional soybeans in either the United States or Argentina. A study by USDA (1999a) reports that while glyphosate tolerant soybeans appear to have low yields, in some US Midwest regions, farmers planting Bt maize had yields 26% higher than conventional, non modified crops. Brookes (2003) found that Bt insect resistant maize in Spain on yields varies depending, inter alia, on location, climatic factors, timing of planting and on whether insecticides are used or not, with a country average yield benefit 6.3%. In Australia, the yield advantage GM rapeseed offers over non GM varieties is estimated to be 12.7% (Foster, 2003), while in Canada it is estimated at 10% (Serecon, et al., 2001). The evidence also suggests that changes in pesticide use rates have been variable (van den Bergh and Holley, 2001). For example, USDA studies found that, in the aggregate, as more farmers adopted transgenic crops, insecticidal treatments have been reduced on maize, whereas, the use of glyphosate, such as Roundup ®, on maize and soybeans has increased (USDA, 1999a and 1999b). However, the use of other, more toxic, chemical decreased. The situation varies by production method and by region. Studies published so far on the effects of transgenic plants on agricultural biodiversity indicate that there is lack of consensus of the consequences of gene flow and conclude that more data and new models are needed to analyse the possible long-term unexpected effects of transgenes (Ervin and Welsh, 2005). The Farm-Scale Evaluation study initiated by the United Kingdom government compared biodiversity in fields of glyphosate-tolerant sugarbeet, maize and rapeseed with that in comparable plots of equivalent non-transgenic varieties in adjoining fields (DEFRA, 2003). The findings showed that there were differences in the abundance of wildlife between genetically modified herbicide tolerant crop fields and conventional crop fields. However, the study stressed that the differences found arose not because the crops have been genetically modified, but because the GM herbicide tolerant crops gave farmers new options for weed control. The differences depended on which and how herbicides were used.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 89

There may also be other types of beneficial environmental impacts. Transgenic crops could contribute to savings in energy and air emissions or reductions in soil erosion due to less frequent operations in the field. Herbicide-resistant crops may lead to environmental benefits by letting farmers use herbicides that do need not to be incorporated with the soil, thereby encouraging a shift to no-till and conservation tillage practices.14 In contrast to crops requiring conventional chemical applications, herbicideresistant crops may thus reduce wind and water sediment damages by allowing for reductions in plowing. These techniques also facilitate the use of winter cover crops, thereby limiting nutrients leaching (e.g. nitrates). Certain transgenic crops in the pipeline could also increase removal of toxic heavy metals from the soil, either by incorporating them in the cells or transforming them to less toxic substances (Engel, et al., 2002; Wolfenbarger and Phifer, 2000).

Main environmental concerns In certain areas, where transgenic crops are released widely into the environment, the main potential environmental risks include impacts stemming from gene flow to wild relatives. The development of resistance to pests and viruses is equally possible, as in the case of conventional crops showing similar resistance, especially in the case of monogenic resistance. An important environmental concern is the possibility that genes may be transferred by pollen or seed to populations of the same crop species or wild relatives in the surrounding area, if the gene(s) is considered to present a hazard. This is an especially important issue when considering the impact of a transgenic crop in its centre of origin and diversity, which can be considered as the geographic region where the crop has its largest diversity and where a close relationship exists with its wild relatives. Many of these issues were explored at an OECD Conference, LMOs and the Environment, which was held in the United States in 2001. A special session at the Conference considered the preliminary evidence of gene flow from transgenic maize to local varieties in Mexico, as well as issues related to the conservation of maize diversity given the possibility of gene flow from transgenic maize. Another potential environmental concern is whether the use of transgenic crops will have adverse impacts on non-target organisms or cause ecosystem damage. The Bt toxin, for example, may have adverse effects on non-target organisms like butterflies or beneficial insect populations that help control pests. There are also issues associated with the potential impacts of transgenic crops on organic agriculture due to the inadvertent presence of transgenic

90 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

crops or material in organic land. Organic farmers are not allowed to have transgenic content in seed or plants. For example, the EU Regulation for organic farming (EC No. 2092/91) forbids the use of LMOs. In July 2003, the European Commission published guidelines for the development of strategies and best practices to ensure the co-existence of LM crops with conventional and organic farming. They are intended to help EU member states to develop workable measures for co-existence in conformity with EU legislation. The guidelines set out the general principles and the technical and procedural aspects to be taken into account. Approaches to co-existence should be developed in a transparent way, based on scientific evidence and in co-operation with all concerned. Measures should be specific to different types of crop and regional and local aspects should be fully taken into account. In June 2004, a law on co-existence was adopted by the Danish Parliament, which lays down rules on the cultivation of LMOs. The key elements of the law, inter alia, is capacity building with LM farmers, information sharing between LMO- and non-LMO farmers, crop specific measures such as distances and cropping intervals, to minimise the adventitious presence of LMOs in other crops and setting up a compensation scheme. The law will be evaluated regularly, with the first evaluation planned two years after its implementation.

2.6.3.

Environmental impact assessments

All OECD countries (and many others besides) have a system of regulatory oversight in place for assessing the environmental safety of transgenic crops. In the majority of countries, these systems have been in place for a number of years; in fact, for well over a decade in many cases. As indicated above, a number of countries have approved the production and commercial use of such crops for human consumption or feed and have accumulated experience in risk/safety assessment of the large-scale use of transgenic crops in the environment. A far greater number of countries (the majority of OECD countries) have approved field trials of transgenic crop plants, which also involve a risk/safety assessment. Most countries continue to make changes and improvements to their regulatory systems in light of this experience. In parallel with this, many OECD countries have continued to sponsor large research programmes designed to address risk and safety assessment questions related to the release of transgenic organisms to the environment. The results of this research have been used to inform and improve the practice of risk/safety assessment. Similarly, a large number of countries have undertaken national studies on the implications of agro-biotechnology. In general, OECD countries have shown a practical commitment to a AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 91

proactive and scientifically-based approach to the risk/safety assessment of environmental applications of genetically engineered organisms. National approaches to biosafety have been enhanced by successful multilateral activities aimed at developing a common approach to both the principles and practice of risk/safety assessment. Much of this common understanding was developed through work at the OECD where biosafety projects, addressing, inter alia, transgenic crops, have been in place since around 1985. An authoritative description of the internationally accepted principles and practice of risk/safety assessment, as it relates to transgenic organisms, is given in a report by OECD’s Working Group for Harmonisation in Biotechnology, which was prepared for the G8 Okinawa Summit in 2000 at the request of the G8 Heads of State and Government. This report shows how environmental risk/safety assessment takes into account the biological properties of the host organism, the gene(s) introduced and their source, how the gene(s) is (are) expressed in the transgenic crop and the nature of the gene product. The characteristics of the organism are taken into account, as well as its likely performance and impact in the environment where it is to be released. For example, exposure and toxicity data are used to examine potential ecological effects to resident wildlife and biodiversity (for example, plants with pesticidal genes may impact non-target species of insects). In addition, information on the eventual use of the product is necessary to ensure a complete assessment. The kinds of information risk/safety assessors use have been developed, in part, from experience with traditional organisms. The general issues assessed for transgenic plants were developed by OECD and include the following: gene transfer, weediness, trait or non-target effects, genetic or phenotypic variability, and the use of vectors and genes from pathogens. The report of OECD’s Working Group to the G8 describes the issues addressed by risk/safety assessors in greater detail. It is important to note another significant multilateral effort, the Cartagena Protocol on Biosafety, which is a key international instrument dealing with “living modified organisms” (LMOs) in transboundary movements. The objective of this Protocol is to contribute to ensuring an adequate level of protection in the field of the safe transfer, handling and use of LMOs resulting from modern biotechnology that may have adverse effects on the conservation and sustainable use of biological diversity. It has established an advance informed agreement (AIA) procedure to ensure that countries are provided with the information necessary to make informed decisions before agreeing to the import of such organisms into their territory. The Protocol has also established a Biosafety Clearing-House

92 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

(BCH) to facilitate the exchange of information on, inter alia, LMOs used for Foods Feeds or Processing. The BCH also assists countries in the implementation of the Protocol.

2.6.4.

Current and future trends

Despite the large degree of similarity among OECD countries in terms of risk/safety assessment, there remain major differences among countries on the topic of the safety of genetically engineered crops/foods. Most of these differences appear to be focused around “risk management” issues. In other words, the measures which are taken once an application has been the subject of a risk/safety assessment and has been approved for release to the environment. These measures include, amongst other things, the monitoring and detection of transgenic material following release, labeling of products, and measures designed to avoid the development of pest resistance to insecttolerant crops.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 93

Annex 2.A. Selected Data Table 2.A1. Gross emissions of GHGs from field burning of agricultural residues, 1990 and 2001 (1 000 tonnes) 1990 CH4

N2O

NOx

2001 CO

NMVOC

CH4

N2O

NOx

CO

NMVOC

Australia Agriculture

8.8

0.3

12.6

0.4

20.8

492.1

28.7

Cereals

7.1

0.2

11.4

0.3

16.3

444.5

25.9

Wheat

4.0

0.1

6.5

0.1

Barley

0.9

0.0

1.5

0.0

Maize

0.1

0.0

0.3

0.0

Oats

0.4

0.0

0.3

0.0

Rice

1.2

0.0

2.2

0.1

2.7

0.1

2.3

56.7

0.0

0.0

0.0

2.5

0.1

2.1

52.6

0.0

Greece Agriculture Cereals

0.0

0.0

0.0

Wheat

1.6

0.0

Barley

0.2

0.0

Maize

0.6

0.0

Oats

0.1

0.0

0.5

0.0

0.5

0.0

Rice Italy Agriculture

0.6

0.0

Cereals Wheat

0.3

0.0

Barley

0.0

0.0

Maize

0.0

0.0

Oats

0.0

0.0

Rice

0.0

0.0

Japan Agriculture

8.0

0.4

0.0

149.1

0.0

6.4

0.5

0.0

123.4

0.0

Cereals

6.8

0.4

0.0

149.1

0.0

5.6

0.4

0.0

123.4

0.0

1.6

0.0

1.2

0.0

5.0

0.3

4.1

0.3

1.3

0.1

0.0

0.0

0.0

0.4

0.0

Wheat Barley Maize Oats Rice Poland Agriculture Cereals

1.5

Wheat

0.2

0.0

Barley

0.1

0.0

Maize

0.0

0.0

Oats Rice

0.0

0.0

94 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 2.A1. (continued). Gross emissions of GHGs from field burning of agricultural residues, 1990 and 2001 (1 000 tonnes) 1990

2001

CH4

N2O

NOx

CO

Agriculture

0.9

0.1

2.1

177.8

Cereals

0.1

0.0

0.2

1.5

0.1

0.0

Agriculture

2.9

1.0

Cereals

1.1

0.1

Wheat

0.3

Barley Maize

NMVOC

CH4

N2O

NOx

CO

NMVOC

0.0

0.8

0.1

1.9

16.1

0.0

0.0

0.1

0.0

0.1

1.1

0.0

0.1

0.0

2.9

1.0

36.1

60.8

8.5

1.5

0.1

3.7

31.4

4.4

0.0

0.7

0.0

0.6

0.0

0.2

0.0

0.1

0.0

0.1

0.0

Oats

0.0

0.0

0.0

0.0

Rice

0.0

0.0

0.1

0.0

Agriculture

12.7

0.3

0.0

0.0

0.0

0.0

0.0

Cereals

12.7

0.3

0.0

0.0

0.0

0.0

0.0

Portugal

Wheat Barley Maize Oats Rice Spain 35.5

61.1

8.6

United Kingdom 9.1

266.0

35.0

Wheat

11.6

0.2

0.0

0.0

0.0

0.0

0.0

Barley

0.9

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.1

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

Maize Oats Rice United States Agriculture

32.6

1.2

28.1

684.8

0.0

36.3

1.5

34.9

762.0

0.0

Cereals

24.6

0.6

13.7

516.0

0.0

24.5

0.6

13.4

514.4

0.0

Wheat

6.5

0.2

4.7

0.1

Barley

0.8

0.0

0.5

0.0

Maize

13.4

0.3

16.1

0.3

3.9

0.1

3.3

0.1

Oats Rice

Source: Greenhouse Gas Inventory Database, 2003.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 95

Notes 1.

Heavy metal contamination of soil can arise from the use of sewage sludge, fertilisers and copper-based fungicides. However, copper is not used in most arable farming systems (Boatman, et al., 1999).

2.

It has been argued that the on-farm economic costs of soil erosion, including the costs of lost soil biodiversity, are less than the off-farm costs of damage caused by sediment (Crosson, 2004). Furthermore, when markets do not function well and property rights are not well established, soil erosion and associated productivity losses are larger than would otherwise be the case (Claasen, et al., 2004a).

3.

Tobey (1991) looked at soil erosion and agrochemical use of the ten primary crops grown in the United States. In terms of soil erosion, soybean production was found to be associated with some of the highest levels of soil loss, at 17.5 metric tons per hectare, being exceeded only by tobacco.

4.

The estimates of potential production losses should be treated with care as the true value of production losses depends on how farmers change management practices to address erosion.

5.

The loss in agronomic productivity due to water-induced soil erosion in North America is estimated at 235 x 103 Mg/y for maize, 60 x 103 Mg/y for soybean, 75 x 103 Mg/y or wheat and 2 x 103 Mg/y for cotton. Globally, the value of annual production losses is estimated at USD 15 million in Africa, USD 98 million in Asia, USD 15 million in Australia, USD 15 million in Europe, USD 206 million in North America and USD 90 million in Central and South America. These losses represent an annual loss of 0.3% of the value of the global production of selected crops.

6.

In a more recent study, den Biggelaar, et al. (2003) found that absolute yield loss caused by erosion ranged between 0.5 and 1.4 kg/ha/Mg of soil erosion for grain and leguminous crops, and between 0.7 and 127.0 kg/ha/Mg for root crops. In North America, crop yields declined at the rate of 0.4%/Mg of soil erosion.

7.

Cropland includes areas used for the production of adapted crops for harvest. Two subcategories of cropland are recognised: cultivated and non-cultivated. Cultivated cropland comprises land in row crops or close-grown crops and also other cultivated cropland, for example, hayland or pastureland that is in a rotation with row or close-grown crops. Non-cultivated cropland includes permanent hayland and horticultural cropland (NRCS, 2003).

96 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

8.

See Orlick, Bauer and Jeffrey (1995) for a literature review of crop rotation and tillage literature.

9.

In Australia, the Ricecheck Programme was developed in 1986 to improve the system of rice management (AUDIT, 2001). It covers seven areas of crop management or component factors: environment (land suitability and safe pesticide use); productivity (field layout, sowing time, crop establishment, crop protection, crop nutrition, panicle initiation date and water management); and grain quality (harvest grain quality). Key checks are provided for each target, allowing for easy self-assessment.

10.

Precision farming, defined as a systems approach to optimise crop yields through systematic gathering and handling of information about the crop and the field, has the potential to contribute to nutrient management by tailoring input use and application more closely to ideal plant growth and management needs. Results from the 1996 USDA Agricultural Resource Management Study found precision agriculture adopters more likely to operate larger farms, have more maize acreage and higher yields, and have higher educational attainment than non-adopter farmers.

11.

For a detailed explanation see: www.nri.org/ipmeurope/homepage.htm.

12.

Different countries have different preferences for terms which describe products of modern biotechnology. This document uses the term “transgenic crops” or “transgenic organisms”. For the purposes of this text, the term transgenic organisms is equivalent to the terms “genetically modified organisms” (GMOs), “genetically engineered organisms ” or “living modified organisms (LMOs)”.

13.

There is a large and still increasing body of literature concerning the potential economic, social and environmental effects of transgenic crops (e.g. Ervin and Welsh, 2005; Ervin, et al., 2000; Nelson and de Pinto, 1999 and 2001, Wolfenbarger and Phifer, 2000; NRC, 2003; Alvarez-Buylla, 2004; van den Bergh and Holley, 2001).

14.

The two most common herbicides are Roundup Ready, with the effective chemical glyphosate and BASTA, with the effective chemical glufosinate (Wolfenbarger and Phifer, 2000).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 97

This page intentionally left blank

Chapter 3 AGRICULTURAL POLICIES AFFECTING THE ARABLE CROP SECTOR

3.1.

Introduction Agricultural support and environmental policies have evolved over time. Shifting government priorities, domestic budgetary pressures and the implementation of multilateral, regional and bilateral trade agreements have impacted on the agricultural policy of OECD member countries. These changes range from limited re-instrumentation, to comprehensive reform which has had particular consequences for the arable crop sector.1 In a number of OECD countries both the number and complexity of policy measures are increasing, as the centre of gravity of policy measures shifts gradually from traditional market price support and output-related measures towards sector-wide and non-commodity-specific policies, particularly those encompassing environmental and rural development concerns.

3.2.

Main policy instruments Government policy regarding arable crops in Australia is limited in scope. Support to producers is mainly provided through budget-financed general measures. Tax concessions, such as rebates on excise taxes on fuel used in off-road vehicles and machinery are available to all agricultural producers. Landholders can claim a tax reduction for expenditures relating to landcare operations and water storage. Statutory marketing arrangements are in place for wheat, barley and rice in some states.2 In Canada, there is no market price support policy. With the termination in 1997 of the various transitional programmes intended to cushion the impact of the phasing out of the Western Grain Transportation Act in 1995, support to arable crop producers was drastically reduced. Canada has not maintained support prices for arable crops since the mid1990s. Instead, risk management policies are prevalent, although these are mostly non-crop-specific in nature. These programmes take the form of Crop Insurance and the Canadian Agricultural Income Stabilization programme (disaster assistance, as well as additional programmes, being administered by provinces). A variety of other policies such as transportation regulations and advance payments in the form of interest-free AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 99

loans, are also provided. Most programmes are funded by Federal and/or Provincial governments. In the EU, market price support, provided through administered prices and trade barriers, and area payments are the main policy instruments affecting arable crop producers. The Agenda 2000 Common Agricultural Policy (CAP) reform package, which deepens and extends the 1992 CAP reforms, provides the basic legislative framework governing agricultural policy for the period 2000-06. This reform package entails, inter alia, a gradual reduction of administered prices for cereals, partially compensated by payments based on area planted. Market price support for cereals is provided through institutional prices, export subsidies tariffs and tariff rate quotas (TRQs), and is combined with set-aside land. There are no intervention prices for oilseeds and protein crops (peas, beans and sweet lupins). Area payments for cereals and oilseeds are based on historic, regional yields and are paid on condition that producers set aside a defined percentage of their arable land; small-scale producers are exempted from the sets-aside requirement. Payments are also made in respect to the land that is set aside. Following the phased implementation of the Agenda 2000 CAP reforms, the EU area payments have been harmonised across major land uses as from 2002. The CAP reforms agreed at the end of June 2003, entail, inter alia, replacement of the arable crops payment, which is based on the area planted to an arable crop, with a single farm payment (SFP) (also comprising past livestock premia), which will be independent of current production levels and prices; and a reduction of the intervention price for rice (Box 3.1). Furthermore, farmers are required to meet specified standards in production methods in order to qualify for the full amount of the payment (EC, 2003a). The crop regime in the OECD member countries joined the EU in May 2004 (the Czech Republic, Hungary, the Slovak Republic and Poland) consists primarily of market price support and supply controls. In the late 1990s, these countries began to implement CAP-type policies to align their agricultural policies with those of the EU, with a view to easing future accession. In this context, the Czech Republic amended the scheme providing area payments to producers in 2001. In Hungary, area payments have been granted to grain producers since 1999/2000, with payments inversely related to farm size. In the Slovak Republic, direct area payments for specific arable crops, including grains and oilseeds, as well as permanent pasture, were introduced in 2000: oilseed payments were reduced and converted to a production basis in 2001.

100 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Box 3.1. Key elements of the CAP reforms affecting the arable crop sector Direct payments

x The “arable crops” payments for cereals, oilseeds and protein crops, as well as payments for flax and hemp, linseed, grain legumes and setaside land, will be replaced by a SFP.

x Support for rice will be raised from EUR 52 per tonne to EUR 177 per tonne. Of this amount, EUR 102 per tonne will be paid as part of the SFP. The remaining EUR 75 per tonne will be paid as a crop-specific payment.

x Durum wheat supplements in “traditional areas” will be reduced from EUR 344.50 per hectare to EUR 285 per hectare and incorporated into the single-farm payment. Supplements elsewhere will be abolished. A special premium of EUR 40 per tonne will be introduced from 2004-05.

x Area set-aside payments for arable crops will be maintained and existing set-aside obligations will be carried over to apply to arable production under the SFP payment scheme.

x Entitlements for the single-farm payment will be based on aids claimed in the 2000-02 reference period, with adjustments taking into account the increases in premiums or the introduction of new premiums as a result of the current reforms.

x The single-farm payment is to apply from 1 January 2005. However, under certain conditions, member states have the option of delaying implementation until 2007. Options for retaining coupled support

x Where member states believe that the application of the single-farm payment will lead to the abandonment of production, they have the discretion to retain part of the pool of money available for the singlefarm payments to use at the national or regional level in order to retain the current support system.

x The share of the single-farm payment fund that can be retained for this purpose varies by type of payment: up to 25% of arable payments (for cereals, oilseeds and protein crops), or up to 40% of the supplementary aid (for durum wheat). Conditions for single-farm payment

x Farmers in receipt of the full single-farm payment will be required to meet “cross compliance” criteria on environmental practices, food safety, animal and plant health and animal welfare standards. Farmers must also maintain their land in “good agricultural and environmental condition”. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 101

x It will be compulsory for member states to apply the cross compliance provisions, with cuts in direct payments noncompliance with the relevant standards.

to

be

imposed

for

Specific voluntary environmental payments

x Member states or regions can also use an “envelope” of up to 10% of the SFP for environmental purposes, or for marketing and product quality improvement. However, these amounts have to remain within the above limits for coupled payments for each sector. Modulation of direct payments and rural development

x In order to finance extra funding for rural development, direct payments greater than EUR 5 000 per farm will be reduced by 3% in 2005, 4% in 2006 and 5% in 2007 and onwards. Intervention prices

x The cereals intervention price remains unchanged. However, the monthly increment in the intervention price, which is applied as the marketing season progresses, will be halved.

x The intervention price for rice will be cut by 50%, to EUR 150 per tonne. x Rye intervention will be abolished. In Japan, market price support, which is provided through administered prices, import barriers and supply management, is the dominant form of support. For rice, government purchase and selling prices apply to less than 5% of consumption and production. The government purchases this quantity as a national reserve from producers who follow the government’s guidelines for rice supply control. TRQs apply to rice, wheat and barley. A state trading body, the Ministry of Agriculture, Forestry and Fisheries, is responsible for importing rice under Japan’s WTO Uruguay Round Agreement on Agriculture (URAA) minimum-access commitment. Supply controls include the diversion of land from rice to other crops under the Production Adjustment Promotion Programme. Direct payments based on rice production are used to stabilise incomes under the Rice Farming Income Stabilisation Programme (RFISP). Budgetary support is also provided for irrigation and drainage, and the re-adjustment of agricultural land. Agri-environmental programmes include measures to encourage farmers to adopt sustainable agricultural practices that reduce fertiliser and pesticide use, and to improve the quality of soil with composting. Budgetary payments for farmers in hilly and mountainous areas aim to prevent the abandonment of agricultural land and to maintain environmental benefits.

102 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Over the last five years, new policies have been introduced with particular significance for rice farming, and some controls over rice marketing were abolished in the 1990s. In 2001, the government announced a Priority Plan for a Stable Food Supply and Aesthetic Land Development, which calls for programme and resources to be directed towards productive sectors, while at the same time providing safety measures to address risks associated with structural reform. Key elements of this policy include a reorientation of support to align producer incentives with market signals while providing greater stability in, and a higher level of, farm income through direct payments. Detailed policies to implement these broad guidelines are to be announced in 2005, although there already appear to be some changes to the existing agricultural policies in the direction of market orientation and direct payments. Korea’s agricultural policy relating to arable crops is dominated by market price support and barriers to imports. The Agricultural and Rural Basic Law, which came into effect in January 2000, reinforces public investment in the infrastructure and technologies of the agricultural sector. In addition, direct payments to producers based on area have been increasing, but remain small relative to total expenditures on support. Arable crop policies in Mexico are based on market price support, mainly due to import barriers, and on direct payments. The latter are now provided as deficiency payments. Other programmes provide support for the transformation of wheat and maize areas to other crops, and for very small farms. Policies relating to oilseeds tend not to lead to interventions in markets to the same extent as occurs with cereal policies, reflecting the relatively small area devoted to oilseeds. However, the deficiency payment system was extended to include safflower in 2001. In Norway, market price support policy dominates. Following the revision of the crop policy regime in 2001, guaranteed producer prices for cereals and oilseeds were replaced by target prices at the wholesale, rather than producer, level. Area payments remain in place and in the case of cereals, the payment rate increased somewhat in 2002. In Switzerland, market price support, mainly through border protection, is the principal form of support to producers. The Federal Agricultural Law (AP 2002) agricultural policy reform programme provides the basic legislative framework governing agricultural policy for the period 19982002. This programme involves the elimination of all guaranteed prices (e.g. for bread wheat and rye) and consolidation of the previous direct payment programme into a uniform area payment. Area payments are based on historical entitlements, on condition that farmers comply with a set of environmental farm-management practice requirements. Arable crop AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 103

farmers also benefit from ecological direct payments, which are granted mainly in the form of payments based on input constraints and on condition that farmers comply with a specified set of environmental standards and farm management requirements. To compensate for the price reduction following the abolition of oilseed price and marketing guarantees in 2000/01, the government introduced payments per hectare of oilseeds. In Turkey, arable crops are mainly supported by market price support (purchasing prices are fixed by co-operatives in the case of oilseeds), import barriers and state enterprises. Since 2001, administered output prices and input subsidies are in the process of being eliminated and replaced by a budgetary payment granted per hectare to all farmers. In the United States, the 2002 Farm Security and Rural Investment (FSRI) Act provides the basic legislation governing farm policy for the period 2002-07.3 The main policy instruments for the arable crop sector include support-price provisions, operating through non-recourse marketing loans, Direct Payments (DP) for crops and Counter-cyclical Payments (CCP). The DP replaced the Production Flexibility Contract Payments (PFCP) provided under the 1996 FAIR Act to programme crops (wheat, feed grains, rice and cotton – to which soybeans, other oilseeds and peanuts have since been added). The CCP replaces the ad hoc Market Loss Assistance Payments (MLAP) provided to farmers during the 1998-2001 period, with support that varies counter-cyclically with market prices for farm programme crops. While PFCP and DP are based on pre-determined rates and past production, the CCP is based on a formula that includes current market prices and past production. Input subsidies are also provided, through interest rate or fuel tax reductions and subsidies to encourage greater insurance coverage on the part of producers. Trade promotion programmes, food aid and export credit guarantees also provide some assistance to arable crop producers.

3.3.

Levels of support The Producer Support Estimate (PSE) and related indicators are the principal tools used by the OECD for measuring and evaluating policies. These indicators provide estimates of the annual level and composition of support to agriculture.4 The level of support to farmers in the OECD area as a whole has declined over the long term, although it remains unchanged in recent years (Figure 3.1). There is wide variation in the levels of support and protection across commodities for which the PSE is calculated. Support for grains and

104 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

oilseeds has exhibited relatively large annual fluctuations, while support for rice has generally remained stable since 1986.

Figure 3.1. Evolution of Producer Support Estimate, by crop, 1986-2003

% 90 80 70 60 50 40 30 20 10

19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

0

Rice

Wheat

Maize

Other grains

All commodities

Oilseeds

Source: OECD PSE/CSE database, 2004.

For the three-year period 2001-03, USD 62 million, or about a quarter of the USD 238 billion transfers from consumers and taxpayers to agricultural producers, was allocated to producers of arable crops, including wheat, coarse grains, rice and oilseeds (Table 3.1). The size of the PSE relative to gross farm receipts (%PSE) for arable crops (39%) was above the average of the whole agricultural sector (31%).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 105

Table 3.1. Support to arable crop producers in the OECD area (million USD) 1986-88

1992-94

1995-97

2001-2003

120 663

137 642

157 377

119 398

Producer Support Estimate (PSE) Market Price Support Payments based on output Payments based on area planted(1) Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income Miscellaneous payments

74 874 48 547 7 781 11 789 57 5 035 1 221 392 52

75 824 48 677 1 726 16 782 554 5 185 2 528 369 3

69 328 35 370 1 311 18 926 4 766 5 427 3 150 476 -98

62 015 22 843 4 180 19 978 7 022 4 291 2 838 845 18

Percentage PSE Producer Nominal Protection Coefficient (NPC) (Arable crops PSE/ All commodities PSE)%

51 2 31

46 2 27

36 2 27

39 2 27

Total value of production

Note: Arable crops include wheat, maize, other grains, rice and oilseeds. 1. This category provisionally includes the US counter cyclical payments. Source: OECD PSE/CSE database, 2004.

Reflecting overall trends, the average support (%PSE) levels in 2001-03 was lower than in 1986-88 for all arable crops, except rice (Figure 3.2, Annex Figures 3.1A, 3.2A, 3.3A and 3.4A). For individual arable crops, support to producers of rice, other grains (e.g. barley, oats) and wheat was higher than that observed for all commodities combined (i.e. total agriculture). With 78%, rice remained the commodity with the highest share of farm receipts derived (directly or indirectly) from government support and of all arable crops it also received the largest absolute PSE, with USD 23 billion. Oilseed producers received the lowest support, on average, with about USD 7 billion, or 24%. Although rice is produced in relatively few OECD countries, it remains the most supported and protected agricultural commodity in the OECD area, with more than four-fifths of farm receipts attributable to agricultural policy. In 200103, prices received by producers and paid by consumers were, on average, more than four times higher than the world price for rice (Figure 3.3). The OECD aggregate rice PSE is largely dominated by Japan and Korea, two of the three main rice producers in the OECD area. While these two countries have %PSEs for rice of 84% and 78%, respectively, support to Australian rice producers totals 6% of gross farm receipts. The support levels for rice in the United States (46%), the EU (37%) and Mexico (35%) are closer to the average rate of total support to all agricultural producers in the OECD (31%).

106 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

The oilseeds sector has traditionally been less directly affected by government intervention than other agricultural sectors, partly because of agreements negotiated earlier under the GATT. However, policies designed primarily to support the cereals sector have often had a significant indirect influence on the oilseeds sector. In general, the share of support in total farm oilseed receipts declined from over 25% in the late-1980s, to under 20% the mid-1990s, and then increased to over 30% as governments responded to low prices by raising levels of support. The increase in support throughout the late1990s until 2001, may have been triggered automatically by existing policies, such as deficiency payments in Japan, or gains associated with the loan programme in the United States, as well as being supplemented through discretionary policies introduced by policy makers. Figure 3.2. Producer Support Estimates by commodity, 1986-88 and 2001-03 (OECD average as % of value of gross farm receipts) Rice

Other grains

Wheat

All commodities 2001-03 1986-88

Maize

Oilseeds

0%

10%

20%

30%

40%

50%

60%

70%

80%

Notes: Products are ranked according to 2001-03 levels; All commodities = whole agricultural sector. Source: OECD, PSE/CSE database, 2004.

The OECD aggregate conceals considerable differences across member countries. There are large differences in the levels of support and protection among OECD countries, reflecting different historical uses of policy instruments and the varying pace and degree of progress in agricultural AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 107

policy reform. For grains, the %PSE ranges from more than 85% of gross farm receipts in Japan, to negative values (implicit taxation) in Hungary. For the 2001-03 period, Australia, the Czech Republic, Hungary, New Zealand and the Slovak Republic have relatively low %PSEs, with Japan, Korea, Norway and Switzerland exhibiting the highest average %PSEs (Table 3.2). Over the 2001-03 period, two countries/regions account for more than three-quarters of total OECD support for the corresponding crop: for wheat producers, the EU contributes approximately 60% and the United States 26%; for maize producers, the EU contributes just over 20% and the United States almost 60%; for rice producers, 66% of the OECD total is due to support in Japan and one-third to support in Korea; for oilseeds, the United States contributes 60% and the EU almost one-third. Figure 3.3. Producer Nominal Protection Coefficient (NPC) by commodity

Rice

All commodities

Oilseeds

Other grains

2001-03 1986-88

Wheat

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Notes: NPC = is a measure of market protection defined as the ratio between the average prices received by producers and border prices. Products are ranked according to 2001-03 levels; All commodities = whole agricultural sector. Source: OECD, PSE/CSE database, 2004.

108 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.2. Ranges of %PSE in OECD countries by crop, 2001-03

<10

10<20

20<30

30<40

40<50

Wheat

Australia, Czech Rep., N. Zealand, Slovak Rep.

Canada, Hungary, Turkey

Poland

Mexico, US

EU

Maize

Hungary, N. Zealand, Slovak Rep.

Canada, Poland

Turkey, US

EU, Mexico

Switzerland

Other grains

Australia, Czech Rep., Hungary, N. Zealand, Poland, Slovak Rep., Turkey

Canada

Rice

Australia

Oilseeds

Australia, Czech Rep., Hungary, Slovak Rep.

All commodities

Australia, N. Zealand

50<70

70<90

Switzerland

Japan, Norway

Mexico, US

EU, Switzerland

Japan, Korea, Norway

EU, Mexico

US

Japan, Korea

Mexico

Korea, Switzerland

Iceland, Japan, Korea

Norway, Switzerland

Canada, Poland

Turkey, US

EU

Canada, Poland, Turkey

Czech Rep., Hungary, Mexico, Slovak Rep., US

EU

Japan

Notes: All commodities = whole agricultural sector. In the following cases PSEs are not calculated: Wheat: Iceland, Korea. Maize: Australia, Czech Republic, Iceland, Japan, Korea, Norway. Other grains: Iceland. Rice: Canada, Czech Republic, Hungary, Iceland, New Zealand, Norway, Poland, Slovak Republic, Switzerland, Turkey. Oilseeds: Iceland, New Zealand, Norway. Source: OECD PSE/CSE database, 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 109

3.4.

Composition of support policies For the arable crop sector as a whole, market price support and payments based on output accounted for almost half of producer support to the sector (Table 3.1). For grains (wheat, maize and other grains), market price support has shown a strong long-term downward trend, declining to 9% of producer support in 2001-03 (Table 3.3). This is to be seen as a positive development, as market price support is one of the most distorting forms of farm subsidies (OECD, 2001e). In contrast, other means of support have either been increased or introduced over the last 15 years, most notably payments based on historical entitlements, which were virtually absent in the 1980s and became relevant only in the mid-1990s in Mexico, Canada and particularly in the United States, after the passage of the 1996 FAIR Act. Moreover, area payments have become the main form of support to grain producers, mainly due to the EU area-based payments to compensate farmers for income losses due to price reductions introduced with the reform of the CAP in 1992, and continued with the Agenda 2000 CAP reform.5 For rice, it is not only the magnitude of total support relative to gross farm receipts that has remained almost unchanged since 1986, but also its overall composition. The overwhelming majority of support is still given via market price support, accounting for 87% of gross farm receipts, while most of the remaining support is in the form of payments based on either output (7%), or input use (4%). All these categories are the most distorting forms of farm support. In contrast, other means of support hardly show up in the OECD average. OECD support estimates for oilseeds suggest that while the level of the support provided to oilseed producers relative to gross revenue is large, the composition is atypical as compared to other commodities. Unlike most other PSE commodities, average market price support in OECD countries, has long been quite low relative to total PSE. However, the transmission of world price signals to OECD oilseed producers is still thwarted by various government support measures, even though market price support is not the common choice for intervention. As portrayed in Table 3.3, support to producers in 2000-02 was provided largely on the basis of output or area. In either case, such payments generally provide an additional incentive for producers to plant beyond what market prices alone would justify. Payments directly linked to output are mainly used in Japan and the United States, while in the EU support is increasingly provided through payments based on area.6

110 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.3. Composition of PSE by crop, 1986-88 and 2001-03 (% share in total PSE for the commodity group concerned) 1 9 8 6 -8 8

1 9 9 2 -9 4

1 9 9 5 -9 7

2 0 0 1 -0 3

G ra in s

M a rke t P rice S u p p o rt P a ym e n ts b a se d o n o u tp u t (1 ) P a ym e n ts b a se d o n a re a p la n te d P a ym e n ts b a se d o n h isto ric a l e n title m e n ts P a ym e n ts b a se d o n in p u t u se P a ym e n ts b a se d o n in p u t co n s tra in ts P a ym e n ts b a se d o n o ve ra ll fa rm in g in c o m e

49 9 32 0 8 1 1

48 1 35 2 10 4 1

13 1 47 19 13 7 1

10 4 47 21 9 6 2

R ic e

M a rke t P rice S u p p o rt P a ym e n ts b a se d o n o u tp u t (1 ) P a ym e n ts b a se d o n a re a p la n te d P a ym e n ts b a se d o n h isto ric a l e n title m e n ts P a ym e n ts b a se d o n in p u t u se P a ym e n ts b a se d o n in p u t co n s tra in ts P a ym e n ts b a se d o n o ve ra ll fa rm in g in c o m e

88 4 2 0 3 3 0

89 4 2 0 3 2 0

90 3 1 0 4 2 0

87 7 1 0 4 0 1

O ils e e d s

M a rke t P rice S u p p o rt P a ym e n ts b a se d o n o u tp u t (1 ) P a ym e n ts b a se d o n a re a p la n te d P a ym e n ts b a se d o n h isto ric a l e n title m e n ts P a ym e n ts b a se d o n in p u t u se P a ym e n ts b a se d o n in p u t co n s tra in ts P a ym e n ts b a se d o n o ve ra ll fa rm in g in c o m e

12 63 6 0 14 3 2

9 1 68 0 12 8 1

7 1 64 2 14 10 2

4 23 39 11 11 8 3

Notes: “Grains” includes wheat, maize and other grains. 1. This category provisionally includes the US counter cyclical payments. Source: OECD PSE/CSE database, 2004.

3.5.

Developments in market price support In 2001-03, market price support remained the most important element in the PSE in five OECD member countries, comprising Japan, Korea, Norway, Switzerland, and Turkey; while in Hungary, the Czech Republic and the Slovak Republic, negative market price support (implicit taxation) was estimated for the cereal markets (Annex Tables 3.A1, 3.A2, 3.A3). In contrast, market price support did not play a major role in the main cereal-producing countries of the OECD. Both the EU and the United States, the two largest producers – together accounting for more than 74% of OECD grains market returns – have significantly reduced market price support to cereal producers over the last 15 years. The Agenda 2000 CAP reform package for grains, introduced in July 2000, entails a 15% cut in the cereals intervention prices in two equal steps over 2000-01 and 2001-02, and an increase in compensation payments. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 111

Moreover, under the CAP reform agreed in June 2003, the intervention price for rice will be reduced by 50%, to EUR 150 per tonne, although the intervention price for cereals will be maintained at EUR 101.31 per tonne. In the United States, market price support to arable crops has been zero since 1996. In the OECD aggregate, level and composition of total PSE, measured as a percentage of gross farm receipts, does not significantly differ between wheat and coarse grains. In the EU, market price support tends to play a slightly smaller role for wheat than for coarse grains due to the uniform support price level. However, considerable differences exist between certain OECD member countries. For example, while wheat markets faced a market price support of -4% in Hungary, this accounted for -26% of gross farm returns for maize markets. For rice, market price support in Japan and Korea shows that domestic prices are well above international price levels. In the case of Japan, gradual policy adjustments have allowed for some increased competition within the domestic market, as the system of government set prices has been relaxed, particularly as regards consumer prices, which are more flexible (although still higher), than world prices, with offsetting emphasis on payments based on output. In the EU and to some degree in Mexico, market price support plays an important role for rice producers, albeit to a lesser extent than in Japan and Korea. However, it is important to note that the type of support accorded to rice producers in Mexico is subject to significant fluctuations, and frequently negative support has occurred at intervals over the past 15 years. In contrast, market price support is not applied to rice in the United States. Instead, producers are supported by means of the marketing loan programme and, hence, receive payments based on output. It is only in Korea, Mexico and Turkey that a relatively high level of support for oilseeds continues to be provided through market price support (Annex Table 3.A4). On the other hand, in Hungary and the Slovak Republic, market interventions create negative market price support.

3.6.

Developments in domestic support policies 3.6.1.

Payments based on output

Support measures classified as output-based payments for current crop production are granted in Canada (mainly for wheat and barley), Japan (mainly barley), Mexico, Norway, Poland and the United States. Notwithstanding the declining trends of these measures for grains and oilseeds during the 1990s, their relative importance in arable crop support has increased during the 2001-03 period. These payments have been

112 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

eliminated in the Czech Republic (for wheat and oilseeds), the EU (oilseeds), New Zealand (wheat), Norway (wheat) and the Slovak Republic (for wheat, maize and oilseeds). On the other hand, between 1986-88 and 2000-02, output-based payments were introduced in Australia (wheat), Hungary (for wheat and oilseeds), Mexico (for wheat, maize, rice and oilseeds), Poland (wheat) and the EU (maize). In absolute terms, the United States accounts for the largest share of output payments in OECD countries. Under the 1996 FAIR Act, arable crops were supported through a non-recourse loan program providing benefits to producers through LDP, marketing loan gains, and forfeiture and interest rate subsidies. Cereals and oilseeds continued to benefit from the MLAP under the 2002 FSRI Act, which compensates farmers for the difference between the world price and the national loan rate. Loan rates have been set for the years 2002 and 2003 and then reduced slightly for the period 200407 for many commodities. For most products, loan rates are higher than in 2001 throughout the entire period. Exceptions are rice, for which the loan rate is unchanged, and soybeans, for which it is reduced. The annual payment limit on Marketing Loan Gains (MLG) and LDP is kept unchanged at USD 75 000 per person and crop year. At the same time, the optional formula to reduce a loan rate in the event of persistent price weakness is removed. Loan programme benefits for wheat and feed grains were the second-largest component of the PSE in 2001, but shrank in 2002 due to high prices. In Japan, the RFISP was introduced in 1998 to compensate rice farmers for part of the loss of revenue in the event that market prices in a crop year fall in comparison with the average price of the seven preceding years, excluding the highest and lowest years. Participation in the RFISP is voluntary, and some farmers have chosen not to participate. Furthermore, to obtain the full benefits, farmers are required to join the Production Adjustment Promotion Programme, which diverts some of their paddy fields land away from rice.

3.6.2.

Payments based on area planted

The significant increase of such payments in the second half of the 1990s is perhaps the most notable change in the average composition of support to the arable crop sector. Accounting for about USD 12 billion on average in 1986-88, payments reached USD 20 billion in 2001-03. Payments based on area planted are used in a number of countries, but they are especially important in the EU, where they represented almost three-quarters of the PSE for arable crops in 2001-03. Payments based on area planted were introduced in the EU, Switzerland and, more recently, in the Czech Republic and the Slovak Republic, to compensate for a reduction of other AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 113

forms of support, namely market price support and output-related payments. Area payments used to be most prevalent in the United States until the early 1990s, but have since largely been replaced by payments based on historical entitlements which have now become the main instrument. Production Insurance payments are the main crop support measures in Canada. The government contributes to a voluntary crop insurance scheme with coverage levels that can covers up to 90% of average yield. Federal premium contributions are expected to average about 36% of the total premiums. Producers may be able to buy insurance based on a basket of crops as opposed to a single crop and the production insurance can cover livestock producers as well. In the EU, area payments were introduced by the 1992 CAP reform to compensate cereal, oilseed and protein crops (peas, field pea beans, sweet lupins and non-textile flax seed) for reductions in price support. They were based on historic, regional yields and were granted on condition that producers set aside a defined percentage of their arable land; small-scale producers are exempted from set-aside requirements. One of the conditions for claiming area payments is that arable crop farmers producing more than 92 tonnes per year must set aside a certain percentage of their land, and must comply with strict rules for managing setaside. The arable land set aside may be used for non-food purposes; left fallow (with the possibility of rotation); afforested or used for nonagricultural purposes (e.g. conversion of arable land to grassland, the introduction of grassland buffer strips around watercourses, etc.).7 In addition, under the terms of the 2003 CAP reform, the land set aside shall be maintained in good agricultural and environmental condition; it shall not produce any crop for commercial purposes; and EU member states have the option of paying national aid of up to 50% of the cost of establishing multiannual crops intended for biomass production on it. In 2001, a total of 6.4 million ha were set aside in the EU as a whole. Half of this set-side area was in France and Spain (Annex Table 3.A5). The minimum rate of set-aside for 2003 is 10% of total area claimed. Small farmers are not required to set aside any land. The payment rate for set-aside is the same as the rate for area payments (i.e. EUR 63 per tonne of reference yield in 2002). Farmers may also set aside additional areas under the voluntary set-aside scheme, for which member states set their own maximum limits, although they must offer a rate of at least 10%. In Ireland, the rate of voluntary set-aside was increased from 20% in 2002, to 40% for the 2003 crop year. In 2003, set-aside payments are estimated at EUR 1 823 million, out of which EUR 114 million (6% of the total) were payments for voluntary set-aside.

114 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

The rice policy regime was also amended, in line with the CAP reforms launched in 1992 and the 1994 URAA, from 1997/98 to 1999/2000. The changes were based on compensatory area payments in return for a 15% reduction in intervention prices, implemented through annual cuts of 5% between 1997/98 and 1999/2000. Producers were compensated through a three-fold increase in the area payment between 1997/98 and 1999/2000, within a maximum national guaranteed area. Korea introduced a Direct Payment Scheme for Rice Income Stabilisation in 2002. The scheme covers income loss should the situation arise where market prices fall below the five-year average. In Norway cereal and oilseed producers are eligible to receive area payments under the Acreage and Cultural Landscape scheme. The scheme, which accounts for one-quarter of total budgetary support to farmers, provides for per-hectare payments to producers. Payment rates are differentiated with respect to geographical location, farm size and production, ranging from NOK 1 500 to NOK 19 000 per hectare. The payment has various cross-compliance requirements attached to it, including limitations on bringing land into production and conserving “cultural landscapes”. Approximately 94% of agricultural land receives this payment. In Switzerland per-hectare payments for extensive cereal and rapeseed farming were introduced with the AP 2002 agricultural policy reform programme, which provides the basic legislative framework governing agricultural policy for the period 2000-03. Farmers must satisfy numerous requirements to receive these payments. There are social and structural criteria (e.g. minimum farm size, age of farm manager, etc.) as well as compliance with a set of environmental farm-management practice requirements. In the United States, payments based on area planted include primarily payments for natural disasters and crop insurance. The 1996 FAIR Act programme crops were eligible for crop disaster payments and crop insurance.8

3.6.3.

Counter-cyclical payments in the United States9

Under the 2002 FSRI Act, a new programme providing CCP for wheat, feed grains, upland cotton, rice, oilseeds and peanuts was introduced to replace the ad hoc MLAP provided to farmers during the 1998-2001 period. Target prices specific to each commodity were set at an initial rate for 2002 and 2003 and at a higher rate for the period 2004-07 for most commodities. CCP is available whenever the target price of a given commodity is higher than a trigger level, which is the return per tonne (i.e. whichever is the higher – the market price or the loan rate) plus the DP for crops per tonne. The amount of the annual CCP is the payment rate (target price minus the

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 115

trigger level) of the relevant crop multiplied by the historical crop base yield and by 85% of the historical crop base area for the farm. For both DP and CCP, producers could retain their 2001 PFCP contract areas from the 1996 FAIR Act as base areas. The 1998-2001 average oilseed area could also be added to these total base areas subject to an aggregate farm area limit. Alternatively, producers had the option to update their base area to their average area planted during 1998-2001 for all eligible commodities. Payment yields for DP are those previously used for PFCP. For oilseeds, the farm’s DP yield is the 1998-2001 average yield multiplied by the ratio of the national averages for 1981-85, relative to the average for 1998-2001. The payment yield for peanuts is the 1998-2001 average yield. For the CCP, producers could use the same payment yields as for the DP. If a farmer opted to update the base area to the alternative 1998-2001 area for all eligible commodities, then the producer could also choose to update yields for the CCP under one of two mechanisms: (i) adding to the current DP yields 70% of the difference between the 1998-2001 yield average and the DP yield, or (ii) use CCP payment yields that are 93.5% of the 19982001 average yields. Planting flexibility provisions allow farmers to receive DP and CCP without obligation to plant or produce any specific commodity. There are some limitations on planting fruits, vegetables and wild rice. Participants receiving these payments must continue to abide by conservation compliance requirements and must use their base area for agricultural or conservation purposes. For each of these payments, a participant can receive a single full payment as one entity and up to a half payment from each of two additional entities. Thus, the maximum payment that an individual can receive is USD 360 000 per year for CCPs, DPs and marketing loan benefits. Producers with an average gross income of over USD 2.5 million over the three preceding tax years are not eligible for payments, unless over 75% of their gross income comes from agriculture. For the CCP, the annual payment is limited to USD 65 000 per person per crop year, with a separate USD 65 000 payment limit for peanuts.

3.6.4.

Payments based on historical entitlements

Payments based on historical entitlements depend on past support or farm receipts, and past area and yields of specific commodities. These payments are made without obligation to plant or produce any specific commodity and are not linked to current production. They are therefore potentially less production- and trade-distorting than other major forms of support. The introduction of such payments in the second half of the 1990s was an important change in the average composition of support to the crops sector in an increasing number of OECD countries. Accounting for about

116 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

USD 57 million on average in 1986-88, these payments have increased, particularly since the late 1990s, to reach over USD 7 billion in 2001-03. The relative importance of these payments has particularly increased for grains, accounting for almost one-quarter of support to OECD grain producers in 2001-03 as compared to zero in 1986-88. Historical entitlement payments tend to be more significant in coarse grains, given the higher production share of the United States in this commodity. The increase in the use of historical entitlement payments was mainly due to the implementation of various programmes since the mid-1990s in the United States. Since the mid-1990s, these payments have constituted the predominant category of US support for grains. Wheat, maize, barley, grain sorghum, oats, rice and upland cotton received government assistance through PFCP introduced by the 1996 FAIR Act. The PFCP entitled eligible producers to receive fixed but declining income payments per hectare, based on historical base. Farmers received payments for 85% of their 1996 base acreage and payment was not related to current plantings, or to production or prices. Entitlement to payments was dependent on the fulfilment of certain compliance conditions, particularly in the area of conservation. In the 2002 FSRI Act, the PFC payments were replaced by DP (see previous section). In addition, the payments were extended to soybeans, other oilseeds and peanuts. Payment rates by commodity for the 2002-07 period will be higher than those paid in 2001. Eligible farmers or landowners receive an annual DP equal to the product of the national payment rate of the applicable crop, the producer’s payment area (85% of base area) for that crop, and the producer’s payment yield for the crop. The payment limit for DP continues to be fixed at USD 40 000 per person per crop year. Payments based on historical entitlements also exist in Canada (Agricultural Policy Framework Transition Payments), Mexico (PROCAMPO) and Switzerland. In Mexico, payments based on historical entitlements have become an important means of crop support. PROCAMPO was introduced in 1994 to replace a series of agricultural support programmes, including input subsidies, price support and import protection for grains and oilseeds. The payments, which account for one-third of total support to Mexican agriculture, are based on the area planted to the main crops (maize, beans, wheat, sorghum, rice, soybeans, sunflower, cotton and barley) in a past reference period. The rate of payment is the same for all eligible producers who can devote land to any agricultural or forestry activity. Producers participating in PROCAMPO are allowed to leave the land idle only if they are registered under environmental programmes operated by the Ministry of Environment. Approximately 3 million farmers and 4.2 million farms have benefited from this programme, which covers about 90% the cultivated area of basic crops. The area benefiting from AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 117

PROCAMPO payments has been mainly cultivated with maize (60%), sorghum (16%), beans (12%) and wheat (6%). In Switzerland, this is the second-most important type of support for grains, after market price support. Launched in 1999, payments per hectare of agricultural land are granted independent of any requirements to produce particular crops. The payments are subject to the income and asset ceilings for direct payments and are not differentiated according to land use or regions. Like other direct payments, area payments are subject to crosscompliance requirements associated with environmental protection. Under the CAP reform package agreed in June 2003, a SFP will replace the various direct payments, including per-hectare payments to cereals and oilseeds (Box 3.1). In principle, farmers will receive a SFP based on a reference amount in a reference period of 2000 to 2002. Those EU member states which find it necessary to minimise the risks of land abandonment can maintain up to 25% of the current per-hectare payments in the arable sector linked to production. Alternatively, 40% of the supplementary durum wheat premia may be maintained tied to production. This single payment will be conditioned on the respect of environmental, food safety, animal and plant health and animal welfare standards, as well as the requirement to keep all farmland in good agricultural and environmental condition (“cross compliance”). For rice, the new CAP reform package entails, inter alia, an increase in the current direct aid from EUR 52 per tonne, to EUR 177 per tonne, a rate equivalent to the total cereals compensation over the 1992 and Agenda 2000 reforms. Of this, EUR 102 per tonne will become part of the SFP and will be paid on the basis of historical rights limited by the current maximum guaranteed area (MGA). The remaining EUR 75 per tonne, multiplied by the 1995 reform yield, will be paid as a crop-specific aid. The MGA will be set at the 1999-2001 average or the current MGA, whichever is lower.

3.6.5.

Payments based on input use

To the extent that these payments encourage the use of inputs, they create production and trade distortions that often result in more intensive production with heightened risks of adverse effects on the environment. The more a payment is specific to the variable inputs necessary to produce particular crops (e.g. fertilisers, animal feed, fuel, irrigation water), the greater the incentive to increase production and the greater the impact on production, trade and environment of these commodities. For the OECD area as a whole, although the level of payments to arable crops based on input use in 2001-03 was lower than the 1986-88 average, its share of support remained, on average, stable at around 6%. In particular, its

118 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

relative importance remained almost stable over time for grains and rice, but declined for oilseeds. Payments based on input use are more important for the grains and oilseed sectors (Table 3.2). They are particularly important in Australia, Mexico, Poland, Turkey and the United States. In Australia, these payments constitute the most important category of support to crop producers. However, their impact on production and trade should be low because of the low levels of support involved. In addition, a substantial part of these payments are based on on-farm services (e.g. extension, pest and disease control) rather than on variable inputs. The most widespread forms of these payments used in OECD countries include fuel rebates, plant protection and disease controls, drainage, interest concessions and capital grants. Budgetary expenditures for irrigation are used in a number of countries, including Korea, Mexico, Poland and the United States. Korea and Poland grant payments for fertilisers although, in the case of Korea, they will be phased out by July 2005. In Poland, a new fuel voucher system was introduced in 2001 which, along with existing policies to reduce financing costs, lowered producers’ input costs. In Japan, rice farmers are eligible for insurance against yield losses which are beyond farmers’ control. The insurance is part of a national system that includes local level (a municipality or insurance association), and prefectural and national levels. Rice farmers can choose coverage for individual plots or for their entire rice-farming operation. The insurance policy also covers extra precautionary expenses (e.g. application of fungicides). When, for example, the government officially alerts farmers in a region to danger from a specific disease, and advises them to apply a pesticide, the cost of the pesticide and its application can be covered by the insurance policy, in addition to the indemnity for yield loss. Moreover, various funds are made available for restructuring rice farming, including loans for co-operative ventures, irrigation and drainage infrastructure.

3.6.6.

Payments based on input constraints

These payments are often targeted to environmental goals, and will be discussed in more detail in the next chapter. They include policy measures based on constraints on the use of a specific fixed or variable input, or a specific group of inputs, by placing restrictions on the choice of production techniques. They are conditional on: the application of certain constraints such as reduction, replacement or withdrawal of the on-farm use of specific inputs; on the choice of production techniques to reduce negative externalities; or on remunerating the production of non-market goods and services. These payments are usually based on land rental costs and/or the costs of adopting and maintaining specific farming practices. Due to the constraints attached to these payments, they may actually reduce production AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 119

or fall among the categories of support that have lower impacts on the production and trade of specific marketed crops. Some countries are using payments based on input constraints to an increasing extent. For the OECD arable crop sector as a whole, although the level of payments based on input constraints was, on average, more than double that of the 1986-88 period, it still represented only 4% of the PSE for crops. Its importance increased over time for grains and oilseed producers, but declined for rice producers. Nevertheless, these payments continued to represent only a very small share of support to crop producers (Table 3.2, Annex Tables 3.A1, 3.A2, 3.A3, 3.A4). For grains, notable increases were observed in the EU and Switzerland, and for oilseeds in the Czech Republic, the EU and Switzerland. Canada does not have any measures which provide payments to crop producers based on input constraints. In Japan, such payments account for around 30% of support to oilseed producers, while for grains and rice payments are relatively insignificant and their importance has declined over time. A set-aside programme was introduced in Korea in 2002, under which farmers who set aside paddy fields receive KRW 3 million (USD 2 400) per hectare. In the United States, the relative importance of payments based on input constraints increased between the mid-1980s and mid-1990s, but it has significantly declined in the 2001-03 period.

3.6.7.

Payments based on overall farm income

These payments, which tend to be the least production- and tradedistorting, and which tend to create less pressure on the environment, are used by only a few countries. In 2001-03 they represented, on average, around 2% of the OECD support to grain producers, only 1% to rice producers and about 3% of the support to oilseed producers. Moreover, the relative importance of these payments has hardly changed since the mid1980s. These payments are particularly important in Australia and Canada, but effectively zero in most OECD countries (Annex Tables 3.A1, 3.A2 and 3.A3).

3.7.

International trade measures 3.7.1.

Import measures

For grains, trade policies remain important, particularly in countries with relatively high relative market price support. Import barriers in the form of tariffs and/or TRQS represent the most important group of measures in this respect. In the EU, import quotas replaced the former grain import regime, which was based on variable import duties. Japan maintains TRQs for wheat and barley, while imports of feed grains, which provide the basis for Japan’s intensive livestock sectors, are allowed almost duty free.10

120 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Similarly, maize and barley imports to Korea are controlled by TRQs, while wheat imports are subject to a tariff. Import tariffs allow for some market price support in Poland for its wheat and barley markets, while TRQs with relatively high in-quota tariffs protect the maize and barley markets in Mexico. While TRQs for cereals are also used in Canada, they generally do not directly affect domestic prices due to Canada’s net export position and the significant underfill of the quotas. As shown in Table 3.4, most TRQs for grains have not been binding because the above-quota tariffs that are actually used (applied tariffs) have been substantially lower than the bound above-quota tariff rates negotiated under the URAA. Also, some imports are subject to preferential tariffs that are lower than these general applied tariffs. Nevertheless, while applied tariffs are often much lower than bound tariffs, in a number of cases they are high enough to constrain or even prohibit imports. For example, Mexico applies a prohibitive general tariff of 198% to maize imports, while the rate charged under the North American Free Trade Agreement (NAFTA) for imports from Canada and the United States is only 3%. Trade measures also remain relevant in some major non-OECD countries. Following its accession to the WTO, China agreed to significantly expand its import quotas for wheat and maize: in 2005, quota levels should be 2.0 and 2.7 million tonnes above their 2000 levels. Trade policies also remain a dominant feature in OECD rice markets, with tariffs averaging about 40% globally and rising to 200% in some markets (Wailes, 2004).11 Moreover, tariff escalation is prevalent in many countries, with imports for milled rice facing much higher or even prohibited tariffs than imports for paddy rice. This pattern of protection depresses world prices for milled high-quality long-grain rice relative to brown and rough rice prices and penalises the milling sectors of high-quality long-grain exporting grains such as Thailand, Vietnam and the United States.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 121

Table 3.4. Tariff rate quotas for arable crops, 2000 Tariff rate quotas (1 000 t)

(1 000 t)

Wheat Canada 1 EU 2 Japan Mexico Poland

227 350 5 740 605 388

63 57 5 895 2 018 96

1.9 0 245 70.1 25

76.5 49 414 70.1 64

50 n.a. 235 67 15

0 0 N/A 4.5 0

NAFTA Central Europe

Maize EU Mexico 3 Korea

2 500 2 501 6 102

1 347 5 800 6 102

0 50 2.6

108 120.3 342

n.a. 198 n.a.

0 3 N/A

Central Europe NAFTA

Rice EU Japan 3 Korea

84 682 103

84 680 103

40 5 5

n.a. n.a. n.a.

n.a. N/A N/A

Central Europe

1 291

Imports

Bound tariffs In-quota Above-quota (%) (%)

Applied tariffs General Preferential (%) (%)

Regional trade agreements

NAFTA EU

Notes: n.a.: not available; N/A: not applicable. 1. The EU also provides access for 600 000 tonnes of wheat from central European countries under preferential arrangements. 2. Bound tariff rates are the ad valorem equivalent of the maximum in-quota and above-quota mark-ups that can be applied. The applied mark-up in 2000 was JPY 32.5 per kg. 3. 1999. Sources: WTO (2002); Hirad, Nelson, Andrews and Shaw (2003); AMAD database.

Japan maintains a TRQ for rice of 682 tonnes (milled rice basis), effectively restricting imports to below 0.8 million tonnes in paddy equivalent. Its above-quota tariff in 2000 was estimated at more than 1 291%. Korea’s import quota is set to increase from its current level of 171 kilo tonnes, to 205 kilo tonnes in 2004 (husked equivalent), with a revision of the import regime after 2004 under discussion. In the EU, the import regime for rice includes a conventional (full) tariff on husked rice at EUR 264 per tonne. However, ceilings for the import prices of rice are defined relative to the intervention price for paddy rice. Roughly 60% of total imports are imported under specific regimes (e.g. for ACP countries, or for basmati rice), allowing for significantly lower duties. While the EU’s "Everything but Arms" initiative gives unlimited and unrestricted access to most agricultural commodities produced in the beneficiary countries, rice was singled out, together with sugar, as one of the sensitive products subject to a transition period. Accordingly, only limited rice volumes will be allowed under the duty-free scheme until 2009. In particular, import duties for rice originating in the Least Developed Countries (LDCs) are scheduled to decline in steps from 2006 onwards,

122 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

reaching zero duty by 2009. As some LDCs have the capacity for large exports of rice, tariff reduction is expected to place serious pressure on domestic policies, leading to some speculation about the sustainability of the current regime over time. For oilseeds, the low market price support in OECD countries is manifest in the low import barriers and avoidance of export subsidies, even when allowed under the URAA. However, some OECD member countries do impose tariffs and maintain TRQs. Out of the 124 tariff quotas used by WTO for oilseeds, 41% are used by OECD countries (Canada, the Czech Republic, Hungary, Iceland, Norway, Poland, the Slovak Republic and the United States) (WTO, 2002).

3.7.2.

Export measures

The EU and the United States have traditionally been the largest users of export subsidies for grains. During the mid-1990s, the EU’s grain exports were subsidised (Table 3.5). The EU accounts for almost the totality of subsidised cereal export volumes as defined by WTO (Annex Table 3.A6). Under the URAA, the EU was required to reduce its export subsidies for wheat from 18.3 million tonnes in the base period, to 14.4 million tonnes in 2000 and beyond. For coarse grains, the required reduction was from 13.7 million tonnes, to 10.8 million tonnes. The reduction in the EU cereal intervention price under Agenda 2000 has reduced the amount of explicit export subsidy required to bridge the gap between EU and world prices. The EU was able to take advantage of a “rollover” provision of the URAA that allowed it to subsidise exports of coarse grains and rice beyond the annual WTO limits in some years, when it had exported less than its limits in earlier years. However, the EU has occasionally exported wheat and coarse grains without export subsidies since 2000. The provision of large export subsidies for grains – principally wheat and, to a lesser extent, barley – was a feature of US grain policy from the mid-1980s until the mid-1990s. Under the URAA, the United States was required to reduce the volume of subsidised wheat exports from a maximum of 20.2 million tonnes in 1995, to a maximum of 14.5 million tonnes in 2000 and beyond. The US has not granted subsidies under the Export Enhancement Programs for wheat, barley or rice since 1995, and export incentives are mostly conveyed through export credit guarantee programmes. In recent low-price periods, export policies for grains were relevant in only a few countries. Export subsidies were paid to malting barley exporters in the Czech Republic. Export credits continue to be applied by many countries, most notably the United States. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 123

For oilseeds, several OECD member countries retain the possibility to subsidise exports under the URAA, but this option has been generally avoided (Table 3.5). However, other measures which could potentially affect export competition, such as export credit programmes and food aid, have been applied to cover trade in oilseeds and oilseed products.

Table 3.5. Share of subsidised exports in total exports by crop, 1995-2001 (%)

Wheat and wheat flour EU Hungary Mexico Turkey Coarse grains EU Hungary Rice EU Oilseeds EU Hungary

1995

1996

1997

1998

1999

2000

2001

26 99 0 59

96 0 0 0

114 0 0 0

114 8 0.2 0

100 0 0 0

76 0 0 0

17 0

92 94

96 0

110 0

115 23

109 27

51 8

43 0

64

94

48

52

64

60

54

0 2

0 2

0 0

0 0

0 0

0 0

0 0

Note: The EU carried over unused subsidies from previous years for coarse grains (1999/2000, 1998/99) and for rice (1999/2000, 1997/98 and 1996/97). Source: Calculations based on country export subsidy notifications to WTO.

3.8.

Summary of agricultural policy reform in the arable crop sector Progress towards the long-term objective of policy reform entails a reduction in overall support and a shift towards less distorting policy measures. The reform process, as measured by reductions in the shares of producer support and output/input-linked support in gross farm receipts, has varied widely across countries and commodities. As portrayed in Figure 3.4, there has been some progress in reform for all arable crops, with the exception of rice, particularly in regard to reducing the share of output- and input-linked support in gross farm receipts for grains. In contrast, while support for rice slightly decreased, the importance of output- and inputlinked measures also decreased. On a country basis, the level of support for wheat and maize and the importance of the most production- and trade-distorting forms of support in gross farm receipts have decreased in all countries, except Poland (Annex Figures 3.A5 and 3.A6). In Mexico, support to wheat producers increased,

124 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

but the importance of output- and input-linked support decreased between the 1986-88 and 2001-03 periods. In Turkey, support to maize producers decreased, but the importance of output- and input-linked support increased. Figure 3.4. Policy reform in the arable crop sector by crop, 1986-88 to 2001-03 (%PSE for 2001-2003 in brackets)

% change in output and input support

100% 80%

D

A

More linked to output or input

60% 40% 20% Less support

0% -20%

More support Rice (78%)

C

B All commodities (31%)

-40% -60%

Oilseeds (24%)

Maize (24%)

-80%

Wheat (37%) Other grains (41%)

Less linked to output or input

-100% -100%

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

100%

% change in %PSE Note: all commodities = whole agricultural sector. Source: OECD PSE/CSE database, 2004.

Concerning rice, some progress was made between 1986-88 and 200103 in Australia, the EU and Korea, but the situation somewhat worsened in Japan (Annex Figure 3.A7). In the United States, while producer support for rice decreased, the share of output- and input-linked support increased between the two periods. In Japan, while support remained unchanged, the importance of output- and input-linked support increased. For oilseeds, there was no progress in policy reform in Korea, Turkey, Mexico, Poland and the United States (Annex Figure 3.A8).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 125

Annex 3.A. Selected Data Figure 3.A1. Producer Support Estimate for wheat by country, 1986-2003

% 60 40 20 0 -20 -40

Canada

EU

OECD

Poland

United States

02

03 20

00

01

20

20

99

98

Australia

20

19

96

95

97

19

19

19

19

19

94

93

92

19

90

91

19

19

88

89

19

19

87

19

19

19

86

-60

Hungary

Source: OECD PSE/CSE database, 2004.

Figure 3.A2. Producer Support Estimate for maize by country, 1986-2003

% 70 60 50 40 30 20 10

OECD

20 03

20 02

20 01

20 00

19 99

6

7 19 98

19 9

5

Mexico

19 9

4

19 9

3

EU

19 9

1

0

2

19 9

19 9

19 9

9

8

7

Canada

19 9

19 8

19 8

19 8

19 8

6

0

United States

Source: OECD PSE/CSE database, 2004.

126 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 3.A3. Producer Support Estimate for rice by country, 1986-2003

% 90 80 70 60 50 40 30 20 10

19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

0

EU

Japan

Korea

OECD

United States

Source: OECD PSE/CSE database, 2004.

Figure 3.A4. Producer Support Estimate for oilseeds by country,1986-2003

%

100 80 60 40 20

7 19 98 19 99 20 00 20 01 20 02 20 03

5

6

19 9

19 9

4

OECD

19 9

3

EU

19 9

1

2

19 9

19 9

9

0

19 9

19 9

8

7

Canada

19 8

19 8

19 8

19 8

6

0

Switzerland

United States

Source: OECD PSE/CSE database, 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 127

Figure 3.A5. Agricultural policy reform for wheat, 1986-88 to 2001-03 (In brackets %PSE for 2001-03)

Change in share of output and input-linked support in gross farm receipts

100% 80%

A

B

More output/ input linked support

60% 40% 20% 0% -20% -40% -60% -80%

Less support

More support

Japan (87%)

Mexico (31%)

D Norway (70%)

Turkey (15%)

Switzerland (60%)

Australia (4%)

United States (35%) Less Canada (19%)

New Zealand (0%) -100% -100% -80% -60%

C

OECD (37%)

output/input linked support EU (45%)

-40%

-20%

0%

20%

40%

60%

80%

100%

Change in share of PSE in gross farm receipts

Notes: For the Czech Republic, Hungary, Mexico, Poland and the Slovak Republic, 1986-88 is replaced by 1991-93. The following countries do not fit on the scale used for the graph: Quadrant B: Poland. Quadrant D: the Czech Republic, Hungary and the Slovak Republic. Source: OECD, PSE/CSE database 2004.

128 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 3.A6. Agricultural policy reform for maize, 1986-88 to 2001-03 (In brackets %PSE for 2001-03)

Change in share of output and input-linked support in gross farm receipts

100% 80%

A

B

More output/ input linked support

60% 40% 20% 0% -20%

Turkey (20%)

Less support

More support

D

C Switzerland (64%)

-40% -60%

Canada (13%) Poland (11%)

Mexico (39%) United States (21%) OECD (24%)

-80% New Zealand (0%) -100% -100% -80% -60%

EU (36%)

Less output/input linked support -40%

-20%

0%

20%

40%

60%

80%

100%

Change in share of PSE in gross farm receipts

Notes: For Hungary, Mexico, Poland and Slovak Republic, 1986-88 is replaced by 1991-93. The following countries do not fit on the scale used for the graph: Quadrant D: Hungary and the Slovak Republic. Source: OECD, PSE/CSE database 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 129

Figure 3.A7. Agricultural policy reform for rice, 1986-88 to 2001-03 (In brackets %PSE for 2001-03)

Change in share of output and input-linked support in gross farm receipts

120%

More output/ input linked support 100%

A

United States (46%)

B

80% 60% 40% 20%

Less support

More support Japan (84%)

OECD (78%)

0%

Korea (78%)

D

C

-20% -40% -60%

EU (37%)

Australia (6%)

-80% -100% -100%

Less output/input linked support

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

100%

Change in share of PSE in gross farm receipts

Notes: For Mexico, 1986-88 is replaced by 1991-93. Mexico is outside the scale used for the graph but would appear in Quadrant B. Source: OECD, PSE/CSE database 2004.

130 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 3.A8. Agricultural policy reform for oilseeds, 1986-88 to 2001-03 (In brackets %PSE for 2001-03)

Change in share of output and input-linked support in gross farm receipts

100% 80%

A

B

More output/ input linked support

Poland (13%)

60% 40% 20% 0%

Turkey (21%) Korea (89%)

Less support

More support

Australia (3%)

-20%

D

C Japan (48%)

-40% Switzerland (85%) -60%

OECD (24%)

Less output/input linked support -80% Canada (14%) -100% -100%

EU (36%)

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

100%

Change in share of PSE in gross farm receipts

Notes: For the Czech Republic, Hungary, Mexico, Poland and the Slovak Republic 1986-88 is replaced by 1991-93. The following countries do not fit on the scale used for the graph: Quadrant B: Mexico, United States: Quadrant D: the Czech Republic, Hungary and the Slovak Republic. Source: OECD, PSE/CSE database 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 131

Table 3.A1. Wheat: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03 0

Australia Market Price Support

49

24

15

Payments based on output

0

19

15

0

Payments based on area planted

0

0

0

2

Payments based on historical entitlements Payments based on input use Payments based on input constraints

0

0

0

0

40

40

49

74

0

0

0

0

11

17

21

24

Market Price Support

36

46

1

10

Payments based on output

21

2

4

2

Payments based on area planted

33

36

6

37

Payments based on overall farming income Canada

Payments based on historical entitlements

0

0

57

20

Payments based on input use

9

10

14

10

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

6

16

20 2

EU Market Price Support

89

53

7

Payments based on output

0

0

0

0

Payments based on area planted

5

36

82

85

Payments based on historical entitlements

0

1

0

0

Payments based on input use

6

5

5

5

Payments based on input constraints

0

5

6

8

Payments based on overall farming income

0

0

0

0 86

Japan Market Price Support

83

87

86

Payments based on output

0

0

0

0

Payments based on area planted

0

0

0

0

Payments based on historical entitlements

0

0

0

0

Payments based on input use

7

8

8

8

Payments based on input constraints Payments based on overall farming income

10

5

6

6

0

0

0

0

New Zealand Market Price Support Payments based on output

0

0

0

0

53

0

0

0

Payments based on area planted

0

0

0

0

Payments based on historical entitlements

0

0

0

0 0

Payments based on input use

41

21

0

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

6

79

100

0 57

Norway Market Price Support

63

70

63

Payments based on output

7

5

1

0

Payments based on area planted

0

17

23

19

Payments based on historical entitlements Payments based on input use

0

0

0

8

29

7

10

10

Payments based on input constraints

1

1

3

3

Payments based on overall farming income

0

0

0

3

132 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.A1. (continued). Wheat: composition of PSE, by country (%) 1986-88 1992-94 1995-97

2001-03

Poland Market Price Support Payments based on output Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

82 0 0 4 14 0 0

68 0 0 0 31 0 0

80 0 1 0 19 0 0

28 41 3 0 26 0 0

Market Price Support Payments based on output Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

90 0 5 0 3 0 0

79 0 11 3 3 2 0

73 0 10 7 3 4 0

44 0 3 30 7 11 0

Market Price Support Payments based on output Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

61 0 0 0 39 0 0

38 0 0 0 62 0 0

31 0 0 0 69 0 0

95 0 0 0 5 0 0

Market Price Support Payments based on output 1 Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

18 13 62 0 5 0 1

40 0 48 0 7 4 1

2 1 19 59 11 6 1

0 4 25 59 7 3 2

Market Price Support Payments based on output 1 Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

63 5 21 0 8 1 1

52 1 33 1 9 4 1

11 1 58 12 11 6 1

10 2 59 14 7 6 1

Switzerland

Turkey

United States

OECD

Note: 1. This category provisionally includes the US counter cyclical payments. Source: OECD, PSE/CSE database 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 133

Table 3.A2. Maize: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03

Canada Market Price Support

6

4

4

0

Payments based on output

44

6

38

47

Payments based on area planted

35

73

8

29

0

0

0

9

13

12

32

5

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

4

16

9 19

Payments based on historical entitlements Payments based on input use

EU Market Price Support

92

79

47

Payments based on output

0

0

0

0

Payments based on area planted

1

11

41

65

Payments based on historical entitlements

0

0

0

0

Payments based on input use

6

6

6

7

Payments based on input constraints

0

3

6

8

Payments based on overall farming income

0

0

0

0

Mexico Market Price Support

63

74

-14

40

Payments based on output

0

2

1

11

Payments based on area planted

0

0

1

0

Payments based on historical entitlements

0

15

92

44

Payments based on input use

37

9

20

5

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

0

0

0

Poland Market Price Support

64

83

82

56

Payments based on output

0

0

0

0

Payments based on area planted

0

0

1

4

Payments based on historical entitlements

8

0

0

0

Payments based on input use

28

17

16

37

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

0

0

0

60

59

53

49

0

0

0

0

36

32

31

7

Payments based on historical entitlements

0

3

7

26

Payments based on input use

3

3

3

6

Payments based on input constraints

0

1

4

9

Payments based on overall farming income

0

0

0

0

Switzerland Market Price Support Payments based on output Payments based on area planted

134 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.A2. (continued). Maize: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03

Turkey 60

68

61

98

Payments based on output

Market Price Support

0

0

0

0

Payments based on area planted

0

0

0

0

Payments based on historical entitlements

0

0

0

0

Payments based on input use

40

32

39

2

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

0

0

0

United States Market Price Support Payments based on output Payments based on area planted1

0

0

0

0

20

2

2

9 20

71

75

2

Payments based on historical entitlements

0

0

56

46

Payments based on input use

7

14

24

14

Payments based on input constraints

1

8

13

7

Payments based on overall farming income

1

1

3

5

Market Price Support

29

41

19

10

Payments based on output Payments based on area planted1

14

1

2

8

48

39

20

29

OECD

Payments based on historical entitlements

0

3

34

33

Payments based on input use

8

11

17

11

Payments based on input constraints

0

5

9

6

Payments based on overall farming income

1

1

1

3

Note: 1. This category provisionally includes the US counter cyclical payments, which fit no category well. Source: OECD PSE/CSE Database, 2004.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 135

Table 3.A3. Rice: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03

Australia Market Price Support

61

24

36

34

Payments based on output

0

0

0

0

Payments based on area planted

0

0

0

0

Payments based on historical entitlements

0

0

0

0

31

62

51

53

Payments based on input use Payments based on input constraints

0

0

0

0

Payments based on overall farming income

8

14

13

13 55

EU Market Price Support

88

93

90

Payments based on output

7

2

0

0

Payments based on area planted

0

0

5

37

Payments based on historical entitlements

0

0

0

0

Payments based on input use

4

5

5

7

Payments based on input constraints

0

0

1

1

Payments based on overall farming income

0

0

0

0

Japan Market Price Support

88

89

90

90

Payments based on output

4

5

4

5

Payments based on area planted

0

0

0

0

Payments based on historical entitlements

0

0

0

0

Payments based on input use

4

4

4

5

Payments based on input constraints

4

2

2

0

Payments based on overall farming income

0

0

0

0

Korea Market Price Support

99

96

96

94

Payments based on output

0

0

0

0

Payments based on area planted

0

0

0

2

Payments based on historical entitlements

0

0

0

0

Payments based on input use

1

2

3

2

Payments based on input constraints

0

0

0

0

Payments based on overall farming income

0

1

1

2

United States Market Price Support Payments based on output Payments based on area planted1

1

2

0

0

31

25

3

77

62

65

51

3

Payments based on historical entitlements

0

0

19

12

Payments based on input use

5

5

17

5

Payments based on input constraints

0

3

9

2

Payments based on overall farming income

1

0

2

1

Note: 1. This category provisionally includes the US counter cyclical payments. Source: OECD, PSE/CSE database 2004.

136 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.A4. Oilseeds: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

0 0 0 0 61 0 39

0 0 0 0 82 0 18

0 0 0 0 83 0 17

0 0 0 0 86 0 14

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

24 28 27 0 18 0 0

32 0 35 0 22 0 11

0 1 15 33 24 0 23

0 2 43 21 10 0 21

Market Price Support Payments based on output 1 Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

0 96 0 0 4 0 0

0 0 93 0 3 4 0

0 0 91 0 3 6 0

0 0 82 0 6 12 0

Market Price Support Payments based on output 1 Payments based on area planted Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

0 62 0 0 2 36 0

0 27 0 0 13 60 0

0 44 0 0 11 46 0

0 67 0 0 5 28 0

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

71 0 0 7 23 0 0

80 0 0 0 20 0 0

72 0 1 0 26 0 0

63 0 4 0 31 0 0

Australia

Canada

EU

Japan

Poland

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 137

Table 3.A4. (continued). Oilseeds: composition of PSE, by country (%) 1986-88

1992-94

1995-97

2001-03

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

95 0 0 0 2 0 0

88 0 0 3 4 2 0

78 0 6 7 3 4 0

43 0 36 13 2 5 0

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

45 0 0 0 55 0 0

67 0 0 0 33 0 0

76 0 0 0 24 0 0

94 0 0 0 6 0 0

Market Price Support Payments based on output Payments based on area planted1 Payments based on historical entitlements Payments based on input use Payments based on input constraints Payments based on overall farming income

0 11 28 0 50 5 8

0 2 28 0 41 25 4

0 3 0 0 57 33 7

0 35 22 17 14 7 5

Switzerland

Turkey

United States

Note: 1. This category provisionally includes the US counter cyclical payments. Source: OECD, PSE/CSE database 2004.

Table 3.A5. Set-aside uptake in the EU, 2001 Area set aside (1 000 ha) Total Non-industrial use

Industrial use

Belgium Denmark Germany Greece Spain France Ireland Italy Luxembourg Netherlands Austria Portugal Finland Sweden United Kingdom

28 218 1 156 46 1 611 1 576 36 233 2 23 104 99 198 269 848

24 196 825 46 1 562 1 212 36 211 1 22 91 91 197 247 805

3 21 332 0 49 364 0 22 1 0 13 8 1 22 43

Total

6 446

5 566

879

Set-aside payments (million EUR) Set-aside Five-year related to per- set-aside Total hectare aid (voluntary) 8 8 64 64 364 364 6 6 216 216 485 485 -11 -11 65 8 73 1 1 6 0 7 33 33 15 15 29 29 56 56 189 189 1 527

8

1 536

Source: European Commission, Directorate General for Agriculture.

138 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 3.A6. Export subsidy volume commitments, volume subsidies and utilisation, 1995-2001 (1 000 tonnes)1 1995 Wheat and wheat flour

All WTO Members Commitments

1996

1997

1998

1999

2000

2001

Volumes Utilisation (%)

59708 4349 7

56105 14410 26

52536 13038 25

48933 14393 29

45329 15894 35

41729 10470 25

32638 1650 5

Commitments Volumes Utilisation (%)

20408 2769 14

19213 14410 75

18021 13038 72

16825 14017 83

15630 15606 100

14438 10204 71

14438 1650 11

All WTO Members Commitments 2 Volumes Utilisation (%)

27924 7666 27

26690 11845 44

25458 8826 35

24225 15311 63

22993 18939 82

21761 7453 34

18008 3999 22

Commitments Volumes Utilisation (%)

13690 6596 48

13121 11845 90

12552 8770 70

11982 14775 123

11412 18379 161

10843 7080 65

10843 3922 36

All WTO Members Commitments 2 Volumes Utilisation (%)

803 1599 199

745 607 81

687 155 23

630 144 23

572 140 24

514 132 26

509 132 26

Commitments Volumes Utilisation (%)

163 89 54

157 227 144

151 155 103

145 144 99

139 140 101

133 132 99

133 132 99

All WTO Members Commitments 2 Volumes Utilisation (%)

2804 5 0

2702 4 0

2601 0 0

2498 0 0

2397 0 0

2295 27 1

2295 5 0

127 0 0

122 0 0

118 0 0

113 0 0

108 0 0

104 0 0

104 0 0

2

EU

Coarse Grains

EU

Rice

EU

Oilseeds

EU Commitments Volumes Utilisation (%)

Notes: 1. As of August 2004, Mexico has not notified for 1999, 2000 and 2001. 2. Include WTO members without export subsidy commitments. Source: Calculations based on country export subsidy notifications to WTO.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 139

Notes 1.

In the EU, for example, the arable crop sector features prominently in the reform of the CAP resulting from Agenda 2000 as the hectare-based payments, which also includes “set-aside” measures for withdrawing land from cultivation, constitute the largest category of budgetary expenditure in the EU’s budget (around 45% of the EAGGF, Section Guarantee, expenditures).

2.

Statutory marketing arrangements are also in place for sugar.

3.

The 2002 FSRI Act was signed on 13 May 2002 and will be in effect for the period 2002-07. It includes a wide range of programmes for commodities, conservation, trade, nutrition, credit, rural development, research, forestry initiatives and energy, and replaces the Federal Agricultural Improvement and Reform Act of 1996 (1996 FAIR Act), which provided the basic legislation governing farm policy during the period 1996-2002. For more details see OECD, 2003d.

4.

The PSE is an indicator of the annual monetary value of gross transfers from consumers and taxpayers to agricultural producers, measured at the farm-gate level, arising from policy measures that support agriculture, regardless of their nature, objectives or impacts on farm production or income (OECD, 2003d). The total PSE is dependent on the size and structure of a country’s agricultural sector, as well as on the monetary unit used. The PSE expressed in relation to the number of farmers or area of farmland is influenced by differences among countries in factor endowment and the number, type, and size of farms. By contrast, when the PSE is expressed as a percentage of gross farm receipts (%PSE) it shows the amount of support to farmers, irrespective of the sectoral structure of a given country. For this reason, the %PSE is the most widely used indicator for comparisons of support across countries, commodities and time.

5.

It is worth noting that the EU compensatory payments have been exempted from the URAA domestic support reduction commitments. The US PFCP of the 1996 FAIR Act were also exempted on the basis of “decoupling” arrangements that are part of the “green box”.

6.

In Japan, for example, payments to soybean producers, which had decreased significantly up to 1994, have risen in recent years, primarily due to higher deficiency payments currently representing over half of the total support.

7.

Originally, all set-aside was to be rotational land, which meant that land could only be set aside one year out of six. For the 1994 harvest year, however, non-rotation was introduced and farmers opting for non-rotation expected to leave the same land in set-aside for five years. For the 1995 harvest year, both options

140 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

were allowed, permitting set-aside to remain in the same place or be relocated as required. As from 1997, a single obligatory option exists, whereas set-aside may be left in the same place or moved from year to year. Different fields or parcels of land within fields can be treated differently, as long as the basic set-aside percentage is met. Land set aside must cover an area of at least 0.3 ha and have a width of 20 metres. Farmers are allowed to transfer set-aside requirements between farms, providing that farms are within 20 km of each other (unless the farm is in an environmental target area). 8.

Contract crops under the 1996 FAIR Act were wheat, feed grains, cotton and rice, while the 2002 FSRI Act also covers oilseeds.

9.

The classification of CCP in the PSEs is still pending as they do not fit well either with the category of payments based on area planted, or with payments based on historical entitlements (see OECD, 2003d).

10.

The in-quota tariff for is a government markup that the Food Agency charges millers for imported grain. For wheat, the maximum markup is bound at JPY 46.5 per kg, while the maximum above-quota markup is JPY 55 per kg.

11.

World trade-weighted average tariffs in 2000 are estimated at 43% on all rice types, 217% for japonica rice and 21% for indica rice (Wailes, 2004).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 141

This page intentionally left blank

Chapter 4 POLICY MEASURES ADDRESSING ENVIRONMENTAL ISSUES IN THE ARABLE CROP SECTOR 4.1.

Introduction Concerns about the effects of agriculture on the environment have intensified in OECD countries over the past two decades, particularly in the areas of water pollution, soil degradation, loss of biodiversity and landscape features. Reducing the harmful and enhancing the beneficial effects of agriculture on the environment has become a widely accepted policy objective. A wide range of agri-environmental measures has been adopted in order to address the environmental performance of agriculture, although many have been put in place to offset the environmental impacts of traditional agricultural support policies. This chapter discusses the agri-environmental and environmental policy measures designed to address environmental issues associated with arable crop farming. The classification of the various types of policy instruments used in the OECD Inventory of Agri-environmental Measures is also adopted here. At the outset it should be pointed out that, although the discussion covers a wide range of countries and issues, the list of measures is not comprehensive and significant gaps still exist. The measures and regulations most likely to affect arable crop farmers are, however, included.

4.2.

Economic instruments Many OECD countries offer payments to farmers and other landholders to address environmental problems and/or to promote the provision of environmental amenities. The use of these measures originated in the mid1980s with a significant expansion taking place throughout the 1990s. By virtue of the fact that many agri-environmental payments relating to arable crops are interwoven with agricultural support measures, it is difficult to calculate the actual level of these programmes. However, the available evidence tends to suggest that the amount of agri-environmental payments provided to arable crop producers is increasing over time. In particular, the EU, Switzerland, Norway and the United States have substantially increased their use of agri-environmental payments. It should be stressed, however, that while agri-environmental payments per se still only represent AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 143

a modest proportion of total agricultural support to the arable crop sector across the OECD area, several other policy measures have environmental components (such as cross compliance and set aside).

4.2.1.

Payments based on farm fixed assets (excluding land retirement)

Payments based on farm fixed assets are policy measures granting farmers monetary transfers, including implicit transfers such as tax and interest concessions, to offset the investment cost of adjusting farm structure or equipment to adopt more environmentally friendly farming practices. A wide range of such payments has been implemented in OECD countries over the past fifteen years. In the EU, structural programme payments are foreseen under the Rural Development Regulation (No. 1257/99) and EU member states have implemented various measures affecting the arable crop sector. France grants a range of agri-environmental payments designed to introduce, maintain and restore specific landscape features, such as hedges and trees, as part of its plan de développement rural national (PDRN) 2000-06. In Italy, France, Germany and the United Kingdom, set-aside land may be planted under the Energy Crops Scheme and still continue to receive set-aside payments under the Arable Area Payments Scheme (AAPS). Spain offers payments for improving the husbandry of irrigated water. In the United States the Environmental Quality Incentives Program (EQIP) was established by the 1996 FAIR Act. Unlike the Conservation Reserve Program (CRP) and Wetland Reserve Program (WRP), EQIP is a programme with individually negotiated contracts intended to minimise environmental damage on farmed land. EQIP is primarily a cost-share programme that pays farmers to adopt more environmentally sound farming practices. Its basic premise is to reduce the negative externalities from agriculture by providing incentives for the adoption of conservation technologies.1 EQIP provides assistance to farmers of up to 75% of the investment cost of installing or implementing structural changes to promote environmental objectives, with the contracts running for 5 up to 10 years. The objectives of EQIP were modified in the 2002 FSRI Act to be more inclusive than the previous EQIP legislation. In particular, modifications were made concerning eligibility requirements; the formula for allocating funds to regions and states; payment limitations, selection criteria; contract length and technical assistance (Klonsky and Jacquet, 2003). EQIP was initially funded at USD 200 million per year from 1997 to 2001. The 2002 FSRI Act extends funding to working land and the protection of productive farmland from non-farming development. The budget for EQIP increased from USD 948 million under the 1996 FAIR Act for the 1997-

144 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

2001 period, to USD 5.8 billion under the 2002 FSRI Act for 2002-07. Overall, water and soil conservation accounted for over half of EQIP spending between 1997-2000. Expenditures related to water use ranged from irrigation practices to providing water for livestock. Soil erosion included tillage practices and installations to reduce the movement of soil from fields. Non-livestock producers spent more on soil, land, and water conservation and water quality than livestock producers. Expenditures on crop nutrients represented 7% of total funding (USDA, 2003a). By facilitating the adoption of more environmentally benign techniques, EQIP can affect relative costs and production levels. Funding for the Farmland Protection Program increased from USD 50 million over 1996-2001, to USD 597 million for the 2002-07 period. In 2000, Agriculture Management Assistance was also made available in fifteen states to provide cost-share payments to farmers to carry out activities to address environmental issues, including the construction or improvement of water management structures, irrigation structures, and the planting of trees as windbreaks or to improve water quality. Non-crop-specific tax and credit concessions are sometimes used to offset the investment cost of adjusting farm structure or equipment to promote environmental improvements. For example, under the Agricultural Improvement Fund, which was introduced in 1999, Japan provides concessionary loans and tax relief to farmers for capital expenditure to promote more environmentally sustainable farming. Supported projects are administered by prefecture authorities and include the purchase of agricultural machinery, such as compost spreaders, and infrastructure improvements. Commonwealth tax concessions were introduced in Australia in the 1980s in order to promote a range of environmental objectives, including the prevention of land degradation and water conservation (ABARE, 2001). In Belgium, the region of Walloon grants interest subsidies of up to 5% of the cost for investments that help protect the environment through the reduction of water pollution by fertilisers and pesticides. In Canada, under the National Water Supply Expansion programme, financial contributions are made for up to one-third of the cost of projects to encourage environmentally sustainable agricultural practices in the use of water resources.

4.2.2.

Payments based on resource retirement

Programmes under this category provide incentive payments to remove land or other factors of production from crop production for environmental purposes. These measures are particularly pertinent to the arable crop sector. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 145

Land retirement programmes to promote environmental objectives have been most widely adopted in the EU, Japan and the United States. The degree of influence they exert on land use and farming practices varies considerably, depending on the length of the set-aside period, the type of land to be taken out of production, the rules governing the treatment of idled land and the possibilities for alternative land use. The payment rates of several of these land diversion schemes are intended to compensate farmers for the cost increases and/or revenue losses associated with abandoning conventional production on part of their land. In practice, however, there is often little or no differentiation in payment rates by type of farm, agrienvironmental measure, or region. In the EU, in addition to the compulsory set-aside introduced by the 1992 CAP reform as a condition for receiving area-based compensatory payments for cereals, oilseeds, protein plants and linseed, two long-term land diversion schemes were introduced as part of the “accompanying measures” of the 1992 CAP reforms [(Reg. 2080/92 and Reg. 2078/92), later encompassed by the Rural Development Regulation (Reg. 1257/1999)]. These two schemes are specifically aimed at achieving environmental objectives. The first scheme is aimed at protecting land taken out of production and the second is aimed at supporting the development of farm forestry. Under the former regulation, farmers who undertake to set aside farmland for at least 20 years with a view to using it for environmental purposes - particularly for the establishment of biotope reserves or natural parks, or for the protection of hydrological systems - are eligible for financial support (“environmental set aside”). In Denmark, the uptake of the 20-year set-aside arable land for the 1994-99 period was 5 900 ha. Moreover, as part of Denmark’s Action Plan for the Aquatic Environment (1998-2003), farmers are offered compensatory payments to take former wetlands out of agricultural production and re-establish them. In Greece, the scheme for long-term set-aside of farmland includes two distinct measures. The first aims to create biotopes and ecoparks on areas of ecological importance, and the second aims to protect water systems from agricultural pollution. Priority has been given to the implementation of the former. For the 2000-06 period, EUR 4 million have been allocated to the environmental set aside, and the projected area for implementation is 19 075 ha. In the United Kingdom, the environmental set-aside has been available since 1996. Arable land entered into certain environmental schemes such as the Habitat Improvement Scheme and the Woodland Grant Schemes can be considered part of the set-aside requirements under the AAPS. As part of the PDRN programme (2000-06), France offers a range of land retirement payments targeting a variety of environmental objectives, including the conversion of arable land to grassland and the introduction of grassland

146 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

buffer strips around watercourses. Payments for buffer strips and cover crops are also available under several agri-environmental schemes in other countries, including Germany and Finland. Under the forestry scheme, support is granted for planting costs for the afforestation of agricultural land. Its objectives are to improve forest resources, reduce the shortage of wood in the EU, encourage forms of countryside management more compatible with the environment, and combat the greenhouse effect. Co-financed by the EU, payments may also cover forestry management costs over a period not exceeding 5 years, and income compensation up to a period of 20 years. In 2003, EUR 350 million was accorded to afforestation measures, which is equivalent to 8% of the payments for rural development. Japan has implemented programmes to divert land from rice production to other crops and activities since 1971 (Wailes, et al., 1991). Having been conceived primarily as supply control measures, the programmes have increasingly come to be regarded as supporting environmental objectives as well. Environmental provisions have gradually been incorporated into successive programmes aiming at avoiding degradation by paying farmers to manage diverted paddy fields in environmentally sound ways, through appropriate cropping alternatives and/or maintenance of idle paddy fields. The diversion programme, which is currently under review, is funded jointly by grower and government contributions. The main source of funds for the diversion payments is the national budget. Rice producers are required to provide financial assistance of JPY 40 000 per hectare for the land kept under rice. Per-hectare payments from the government vary according to the use made of diverted land. Payments for various crop alternatives have also varied over time. Farmers participating in the diversion programme receive the full benefits of the RFISP. There are several alternative uses for the diverted land. These include diversion to general crops such as soybeans, wheat, barley, feed grains and forage; diversion to permanent crops, such as fruits trees; diversion to other purposes such as crops for landscape conservation, preservation of paddy fields without cropping, diversion to specific crops such as vegetables; land improvements during the production period and conservation management. The diversion programme alternative of keeping the paddy fields planted in rice, but not harvesting the rice grain for human consumption can take various forms: rice can be planted and fed to animals as fodder (the whole plant is used as feed); rice can be cut and the whole plant fermented for use as cattle feed (fermented fodder); rice can be cut and used as straw; rice grain can be AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 147

harvested and fed to animals (feed rice); rice (and other) plants can be ploughed into the soil as green manure (Fukuda, Dyck and Stout, 2003). The total area diverted has varied considerably over time, ranging from around a quarter of a million hectares in 1975, to almost million hectares in 2002, an area equivalent to over a third of Japan’s total rice paddy area. Between 1990 and 2002 the set-aside area increased by almost 15%, while total payments for set-aside increased by 5% (Annex Table 4.A1).2 Land diversion programmes have dominated agricultural environmental expenditures in the United States since the mid-1980s. The largest longterm cropland retirement programme is the CRP, which was introduced under the 1985 FSA. Farmers who agree to enrol in 10-15 year contracts to retire land from production receive an annual rental payment. Participation is tied to environmental goals by taking marginal land out of production and requiring land management, such as the production of cover crops. Farmers enter bids for the rental rate they are willing to accept for taking land out of production. The bids are selected by the Farm Services Agency, based on a formula for environmental impact. Initially, the main purpose of the CRP was to combat soil erosion, but, as the programme evolved, other objectives were added, including habitat and water quality improvements, carbon sequestration and air quality improvements. In 2000, 8.8% of cropland in the US was idled under the CRP (Vasavada, Warmerdam and Nimon, 2001). Under the 2002 FSRI Act, the maximum set-aside area eligible for CRP payments increased to 15.9 million ha, up from 14.7 million ha under the 1996 Act. The expansion of the CRP under the new Act will reduce the area of land available for crop production (Westcott, Young and Price, 2002). Although the CRP aims to retire environmentally marginal cropland, it may also generate significant output effects if land that is environmentally marginal is not marginal from the economic viewpoint. Since 1996, CRP rental payments have averaged more than USD 1.5 billion a year, or around 96% of the total spent on land retirement by the USDA. The 2002 FSRI Act maintains and extends the programmes that retire environmentally sensitive land from crop production, with the emphasis placed on programmes that support conservation on land in production and environmentally friendly farming practices on livestock operations, and the establishment of a new Conservation Security Program (CSP), which pays producers to adopt or maintain environmentally friendly practices (see next section). Under the Act, land retirement programmes have been extended, particularly relative to wetlands and funding has been increased for farmland protection. A new Grassland Reserve Program (GRP) has been created to assist landowners in restoring and conserving grassland, and new provision aims at ensuring regional equity in conservation funding.

148 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

4.2.3.

Payments based on farming practices

Payments based on farming practices are policy measures granting annual monetary transfers (including implicit transfers, such as tax and credit concessions) to encourage farmers to implement more environmentally friendly farming practices. Such payments are widely used to support arable crop producers in several OECD countries. Prominent among these measures are payments to support farmers adopting lowintensity farming systems, including organic production systems and other less input-intensive forms of production or more intensive management practices (OECD, 2003e). The EU co-finances, with EU member states, a wide range of agrienvironmental payment programmes based on farming practices under a policy framework first established in 1992 under the Agri-environmental Regulation (No. 2078/92), and later encompassed in the Rural Development Regulation (No. 1257/99) of the Agenda 2000. These programmes are often established at different administrative levels (national, sub-national, and regional). EU member states are required to implement agri-environmental payment programmes to achieve environmental benefits that go beyond those obtained through the application of “good farming practices” (which are defined as levels of environmental quality that should be achieved at the farmer’s own expense). Often farmers may select particular activities from a complementary “menu” of programmes. Farmers are reimbursed for the costs incurred or income foregone as a consequence of entering into these activities, sometimes with the addition of an incentive element. In general, the programmes are for a minimum duration of 5 years, except for long-term set-aside, which is for a period of at least 20 years. The EU co-funds up to 85% of the cost of programmes in Objective 1 areas (defined as lessdeveloped regions), and up to 60% in other regions. Under the Rural Development Regulation, a host of payment programmes to arable crop producers have been implemented in all EU member states. Payments to support the adoption of less input-intensive farming practices are the most widely used. In particular, by the mid-1990s most EU member states had introduced a variety of national or regional programmes to support organic arable crop production. These schemes generally provide transitional area-based support to farmers in relation to the area and the type of crop concerned in the undertaking for a minimum of five years, to encourage the conversion from conventional to organic farming. With the new Regulation, the upper limits of premiums, which are granted on an annual basis, vary from EUR 600 per hectare for annual crops, to EUR 900 per hectare for specialised perennial crops, and to EUR 450 per hectare for other land uses - significantly higher than under Regulation 2078/92. Member states of the EU are allowed to exceed these AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 149

amounts as state aids. Finance for training, education and consultancy services are also available in some countries. Most European countries, with a few exceptions (e.g. France), also provide government support for the continuation of organic farming beyond the initial conversion period. A uniform national policy is applied by most EU member states, but several (Finland, France, Germany, Italy, Spain, Sweden and the United Kingdom) have significant regional variations in rates of payments and requirements. Average rates of support for organic arable crops in 2001 are presented in Table 4.1. Payment rates for arable crops varied widely between and within countries, where regional variations existed. Some evidence suggests that specialist cropping farms (arable and horticulture), as well as intensive pig and poultry producers, seemed to be less attracted by the available payment rates. To address this problem, Denmark introduced in 1997 a supplement of EUR 230-266/ha/year for three years for arable farms without milk quotas and for pig farms. Requirements and eligibility conditions also vary between countries. In a few countries (Greece, Portugal, Spain and parts of Italy), the payments were restricted to specific crops and, more commonly, permanent grassland and/or set-aside was excluded from the schemes. Some countries (Austria, Denmark, Finland, Germany, Ireland and Italy) introduced additional environmental requirements. In Ireland and Finland, participation in the main agri-environmental programme was compulsory, while in the United Kingdom, additional environmental restrictions were incorporated into national organic production standards. In the United Kingdom, the Organic Aid Scheme provided financial assistance to farmers in conversion, starting in 1994. This scheme was replaced by the Organic Farming Scheme in 1999. As under the first programme, support is offered for five years for organic conversion. Annual payment levels average GBP 90 per hectare for land eligible for AAPS or under permanent crops; GBP 70 per hectare for land not eligible for AAPS; and GBP 10 for unimproved land. Additional payments of GBP 300 per organic farm in the first year, GBP 200 in the second year and GBP 100 in the third year are available to help cover costs associated with such items as training and organic certification.

150 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 4.1. Payment rates for organic arable crop farming, 2001

C O UN TRY

Typical paym ent rates for m aintenance

Typical paym ent rates for conversion (1st tw o years)

(EU R/ha)

(EU R/ha)

C rops eligible for arable area paym ents 1

Australia

C rops eligible for arable area paym ents

O ther arable crops

O ther arable crops

0

0

0

0

Austria

327.0

327.0

327

327

Belgium

111.6

223.1

180

299

C anada

0

0

0

0

56

56

C zech R epublic D enm ark Finland France 7 G erm any G reece

0 102-230 129.1-301.7

0 102-230 301.2-301.7

H ungary 2 Japan Korea

140

87

280-498

498-600

151

212

102-281

102-281

129.1-301.7

301.2-301.7

85-140

85-140

2

Iceland Ireland

242

(3 )

242

(3)

332

(3 )

332

(3)

Italy

185

309

Luxem bourg

173

173 226

M exico N etherlands

279

279

226

0

0

0

0

374

374

49 (4 ) 217

49 (5) 362

N ew Zealand N orway Poland

37

37

Portugal Slovak R epublic Spain Sweden 6 Switzerland

92.32

(4 )

149

162.27

(5)

92.32

(4 )

162.27

(5)

149

104-185

104-185

50

50

290

290

0

0

0

0

513-769

Turkey U nited Kingdom U nited States

Notes: 1. For the Czech Republic: arable land; Finland: includes dried pulses; France: includes protein plants; Portugal: cereals; United Kingdom: includes other crops. 2. The data for Japan refer to 1999, for Korea to 1998 and for the Netherlands to 2002. 3. Includes payment for REPS. 4. Dryland arable. 5. Irrigated area. 6. Organic support is for the whole farm and not for specific sectors. 7. Differentiation of payments according to the conversion years was introduced subsequently. Sources: Foster and Lampkin (2000); Yussefi and Willer (2003); USDA/ERS; Delegations. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 151

The Rural Environment Protection Scheme (REPS), which was introduced in 1994, is the primary source of direct funding available to organic farmers in Ireland. The REPS is a voluntary scheme available to all farmers. They receive an annual basic payment, which is area-related, and they may also undertake any Supplementary Measures. One of these is the Organic Supplementary Measure (OSM). Financial support under the OSM is provided to farmers who are converting to, or continuing with, organic farming production systems. Support varies according to farm size. For example, for farms of at least 3 ha and up to 40 ha the rate is EUR 181 per hectare for inconversion farms, and for farms with full organic status EUR 91 per hectare. For farms smaller than 3 ha the corresponding rates are EUR 242 per hectare and EUR 121 per hectare. The in-conversion rate of payment is available for up to two years. Norway, which introduced support to organic farming in 1989, offers an organic conversion payment, paid on a per-hectare basis, together with ongoing area and headage payments for organic farmers. Spending on area payments used for organic farming rose dramatically in 2001. This subsidy is supplemented by additional government support to help producers establish themselves in this market. A number of countries, including Australia, Canada and New Zealand do not provide support for either conversion or continuation of organic agriculture. In the United States, a few states (Iowa, Minnesota and New Jersey) have made funds available to support organic producers through EQIP. In most EU member states arable crop farmers have been eligible for environmental payments based on farm practices under the Regulation targeting biodiversity and landscape objectives (Reg. 2078/92 and Reg. 1257/99). For example, in the United Kingdom, under the Environmentally Sensitive Areas Scheme (ESAS), incentive payments per hectare are offered under 10-year contracts to farmers who adopt agricultural practices to safeguard and enhance biodiversity in areas of particularly high landscape, wildlife or historic value – there are now 22 ESAS in England, covering some 10% of agricultural land. A variety of payment programmes also exists in Belgium, Finland, France, Germany, Greece, Ireland, the Netherlands, Portugal, Sweden and Spain to encourage farm practices to preserve specified cultivated areas, protect water quality and combat erosion. These programmes include, inter alia, per-hectare payments to encourage farmers to reduce the density of fertilisers and pesticides in the production of cereals. Many EU member states, including Austria, Belgium, Denmark, Finland, France, Germany, Greece, the Netherlands and Spain, have also implemented a variety of other payment programmes to encourage less input-intensive farming practices. These include programmes to promote the extensification of crop production and the adoption of integrated crop

152 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

production. For example, in the Walloon Region in Belgium, annual payments of EUR 150 per hectare are paid to reduce and contain the use of herbicides in maize, and of EUR 90 per hectare to reduce the use of industrial fertilisers and chemicals in grains. Moreover, farmers are paid EUR 100 per-hectare per-year for cultivation of older varieties of cereals, or buckwheat or spelt wheat in less- favoured areas. In the Netherlands, the Crop Protection Policy 2001-10 aims to reduce the environmental burden through a 95% reduction in the use of chemical pesticides by 2010, compared to 1998. The reduction will be achieved by encouraging farmers to move towards integrated crop production. Payments based on farming practices have also been implemented in a number of other OECD countries. Switzerland’s AP 2002 offers a range of payments based on different standards of agricultural practices, including per-hectare payments for crop preservation strips. The use of nitrogen fertilisers, insecticides and chemical or mechanical weed control in these strips is prohibited. Suitable seeds are grain (except for maize), rape, sunflower, protein peas, field beans and soya. Only plain and hill regions are eligible. In 2002, farmers received an average of CHF 1 000 per hectare. In addition, area payments to preserve or improve water quality were made. Certain restrictions on fertiliser and pesticide use and specified farming practices can be negotiated between farmer representatives, the cantons and federal authorities. In 1999, Korea introduced direct payments to farmers restricting the use of fertilisers and pesticides in drinking water conservation areas, which was revised in 2002. In the new system, the payments are nationwide and targeted to certified environmentally friendly farmers. In the United States, the EQIP provides financial and technical assistance to farmers to promote the adoption of conservation practices in environmentally sensitive areas. EQIP provides assistance of up to 75% of the costs of certain conservation practices, such as nutrient management, manure management, integrated pest management, irrigation water management, and wildlife habitat management. Farmer contracts are for up to 10 years (see also Section 2.1). In addition, the CSP provides several tiers of payments to farmers based on different achievement levels expected by conservation practices, and is expected to achieve environmental enhancements considerably beyond established programmes such as EQIP. The CSP focuses on land-based practices and specifically excludes livestock waste-handling facilities. Under the CSP, producers develop conservation plans and enter into conservation security contracts that provide annual payments for implementing or maintaining the practices designated in the conservation plans. All agricultural land (cropland and grazing land) is AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 153

eligible. Cropland must have been cropped in four of the six years prior to 2002. Land enrolled in the CRP, the WRP and the GRP is not eligible. It is estimated that around USD 2 billion will be spent on this programme over the next ten years.

4.2.4.

Environmental taxes

Environmental taxes are per-unit charges for actions contributing to environmental degradation. Charges may be associated with emissions or farm input use such as pesticides and fertiliser. Environmental taxes are consistent with the “polluter pays” principle, and discourage the expansion of environmentally damaging activities. As in other agricultural sectors, the implementation of taxes and charges appears to be rare in arable crop farming. This may at least partly reflect practical problems of measurement, particularly the “non-point” nature of pollution from farming. That is, unlike other economic sectors, where pollution can usually be monitored at “point”, the pollution from farming is much dispersed, as it tends to originate from many different farms and in varying intensities. In a number of OECD countries, agricultural producers, including arable crop producers, are subject to taxes and charges on the sale of inputs identified as having a potentially adverse impact on the environment. For example, various taxes were introduced on pesticides in Belgium, Denmark, Norway, Sweden, the United States and in one province in Canada. Taxes on commercial fertilisers are also applied in a few OECD countries, including Denmark and Sweden. In Denmark, excessive application of nitrogen is subject to a levy at a rate of DKK 10 per kg of N, and DKK 20 per kg of N when the excess is over 30 kg of N per hectare. In Norway, excise taxes on nitrogen and phosphorus fertilisers were abolished with effect from January 2000. In the United States, taxes on fertiliser and pesticides exist in only a few states, although their rates are too low to have a significant impact (Claassen, et al., 2001). In most OECD countries where a tax on pesticides exists, the tax is relatively small. At the same time, the revenue is used to compensate farmers who invest in environmentally friendly equipment. The tax is high in Denmark, amounting to around one-third of the retail price. It was introduced in 1998 and is imposed on retail prices, ranging from 54% on insecticides and soil disinfectants, to 33% on fungicides, herbicides and growth regulators to 3% on other pesticides. Around 55% of the tax revenue is channelled back to farmers and landowners, 10% is allocated to payments supporting organic farming and the remaining 35% finances research and

154 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

monitoring of pesticides in the environment and the administration costs of the pesticide approval system. In Sweden, the environmental tax on fertilisers, which was introduced in 1994, is compulsory and applicable to the whole country. Its rate depends on nitrogen and phosphorus content: a tax of SEK 1.8 per kg of N is charged if the content of N exceeds 2% of total inorganic fertilisers; and of SEK 30 per gram of Cadmium (Cd) in phosphorus fertilisers, if Cd contents exceed 5 grams per tonne of phosphorus. In 2001, the tax revenue was SEK 364 million. Water-use charges are also now commonly applied in many OECD countries. However, water charging in agriculture tends to be less comprehensive than in other sectors, and in many cases is limited only to the cost of administration and delivery (e.g. issuing permits, maintaining infrastructure), rather than the full opportunity cost of water derived from other potential uses (Brouwer, et al., 2002). Some OECD countries have, however, begun to encompass the principle of more comprehensive cost recovery for water in policy. For example, in the EU, the Water Framework Directive (No. 60/00), which was adopted in 2000, requires member states to take account of the principle of cost recovery relating to water services, including both environmental and resource costs. In Australia, a package of rural water reforms, including changes to water prices and the water rights, buying and selling of water and water infrastructure was agreed in 1994. By 2005, water pricing will reflect the full cost of providing water services and water rights, instead of being tied to land, will become tradeable business assets.

4.2.5.

Tradeable rights/quotas

Tradeable rights/quotas are measures that establish environmental quotas, permits, restrictions and bans, maximum rights or minimum obligations to economic agents which are transferable or tradeable. They are aimed at addressing a lack of clearly defined property rights. While there are a number of advantages associated with tradeable rights relative to other economic instruments for improving environmental outcomes, their application in agriculture is fairly rare (OECD, 2003e; ABARE, 2001). This could be partly attributable to the often high transaction costs involved in setting up and monitoring workable systems, due to the predominance of non-point source pollution associated with agricultural farming. Tradeable rights have been established to assist in the management of natural resources. Tradeable water extraction rights for irrigation exist in some regions/states of Australia and the United States. For example, water AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 155

entitlements were issued to farmers in the 1980s in the Murrumbidgee Irrigation Area in New South Wales, Australia. Farmers may sell excess water on a temporary or permanent basis, either inside or outside the region. In 2002, Australia announced its intention to further develop a marketbased system for water use by 2005, including the introduction of the trading of water across state boundaries. In the United States, under the Clean Water Act, landowners wishing to drain wetlands on their property are required to offer compensation to the authorities in order to receive a permit. Options for compensation include “mitigation banking”, for which federal guidelines were established in 1995. Essentially, this scheme allows developers to purchase credits in larger, centralised wetland mitigation projects in order to compensate for the effects of their own projects on wetlands. By the late 1990s, 160 operating mitigation banks were identified, with 80 established for the sale of credits. Many are owned and operated by government entities.

4.3.

Regulatory measures Measures classified under this category involve a compulsory restriction on the choice of farmers. Enforcement mechanisms such as penalties, reduction or withdrawal of support are used when producers are found to be in breach of specific rules.

4.3.1.

Regulations

Regulatory requirements are compulsory measures imposing requirements on producers to achieve specific levels of environmental quality, including environmental restrictions, bans, permit requirements, maximum rights or minimum obligations. Failure to respect these regulations is liable to financial penalty and custodial sentences, although this also depends on enforcement systems. Some of these requirements are specific only to arable crops, while most are part of broader national environmental legislation affecting many sectors of the economy. Regulatory requirements are generally less flexible than economic instruments, as they tend not to allow producers the freedom to determine for themselves the most appropriate way of meeting environmental objectives. However, they tend to minimise risk and uncertainty, and therefore constitute a vital element of environmental policy in most OECD countries, particularly with respect to acute environmental problems. Regulatory requirements have long been applied in the agricultural sector to deal with problems relating to the pollution of soil, air and water, and the protection of environmentally sensitive areas. Regulatory measures

156 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

to help to achieve agri-environmental objectives impose varying degrees of restrictiveness on landowners in a variety of different ways, ranging from broad prohibitions or requirements, to very prescriptive details about farm management practices in specific regions. Four main categories of regulations affecting arable crop producers can be distinguished: water quality and quantity; soil; air quality; and landscape and biodiversity.

4.3.1.1. Water quality and quantity Since the 1980s there has been a general expansion in regulatory measures to protect waterways and ground water. These regulations are aimed at protecting water from contamination from nutrients, sediments or pesticides and at restricting water allocation. Arable crop producers are mainly affected by regulations to protect ground water. Policies to control the effects of nutrient enrichment by nitrates and phosphates and the associated pollution from fertilisers require arable crop farmers to respect a wide range of on-farm constraints. Many of these are multi-purpose. In Canada, the federal government has set standards for nutrients, bacteria and pesticides, while the primary responsibility for the environmental regulation of agriculture rests with the provincial and municipal levels of government. Nutrient enrichment by nitrates and phosphates is a high-priority issue in the EU, as contamination of both ground and surface waters and soils is a serious problem in parts of the EU. The main causes of these problems are inappropriate use of chemical fertilisers and manure, especially in areas with intensive livestock production (mainly pigs and poultry), or specialised crop farms (particularly intensive horticulture). An increasing number of regulatory requirements imposed in OECD countries derive from state, provincial, regional or local measures, often under the framework of overarching legislation. The Nitrates Directive (No. 91/6761) provides the framework in the EU for controlling water pollution. A wide range of on-farm constraints is applied to farming in order to meet the requirements of policies to control nutrient enrichment by nitrates and associated pollution from fertilisers and livestock wastes. Implementation of the Nitrates Directive by many EU member states is currently not complete. Each EU member is responsible for meeting the targets set by the Nitrate Directive, so differences emerge at country level. In Denmark, for example, the Act on Levies on Nitrogen sets a maximum application rate of effective nitrogen for each particular farm according to its crop structure, type of soil and climatic zone.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 157

In the United States, permits and mandatory requirements by crop producers for nutrient discharges can be regulated by EPA’s Confined Animal Feeding Operation (CAFO) regulations. If a landowner has a CAFO, they are required to keep records and manage nutrients on their operation. These operations must keep records for the past 5 years that document that they have and carry out an acceptable nutrient management plan. In addition, further state level regulations may apply. Regulations are mainly used as support to other voluntary programmes to control pollution (Carpentier and Ervin, 2002). In some states (e.g. Arizona, Nebraska), general permits requiring best management practices are needed for the application of nitrogen fertiliser, for example. The permit requires the use of practices including timing, precise application, irrigation water management, testing, and tillage practices that maximise nitrogen uptake. All OECD countries have a wide range of restrictions regarding pesticides use. They range from applying good farming practices to strict control and prohibition of certain pesticides in specific zones. These restrictions have typically been amended over time in such a way that many countries now approve new pesticides for a limited period only (commonly five to ten years). Some requirements relating to inputs have been implemented in response to international pressures – for example, the phasing out of the marketing and use of methyl bromide pesticides under the 1987 Montreal Protocol on Substances that Deplete the Ozone Layer. In the EU, the Drinking Water Directive (98/83) sets strict standards for maximum concentrations of pesticide. The great majority of restrictions on the use of pesticides in the EU apply at national or sub-national levels. Several member states put restrictions on farmers, limiting the use of pesticides in environmentally sensitive areas. Aerial spraying of pesticides is one of the most restricted pesticide application practices. It is prohibited in some parts of the EU (Austria, Greece and Sweden) and Australia, although it is still permitted in Canada, New Zealand and the United Sates. It is heavily controlled in many other regions and countries, with licences or permits commonly being required. In certain areas such as the EU, the United States, Canada and Australia, the use of pesticides is restricted within a certain distance of watercourses, through mandatory measures, while New Zealand applies voluntary measures (Brouwer, et al., 2002). Absolute quantitative restrictions to limit the extraction of water for irrigation purposes are becoming increasingly common in regions where water is scarce. For example, in order to address problems associated with nutrient loads and eutrophication, increasing water salinity and declining wetlands, caps on water extractions in many irrigation zones in Australia were set in the 1990s, and in some cases embargoes still exist on further irrigation licences to extract ground water. These caps have sometimes also

158 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

been combined with the creation of tradeable rights. Such measures primarily affect rice farming, as grain farming in Australia is mainly rainfed. In New Zealand, irrigators are required to apply for permission to use water and comply with any conditions imposed, including reductions in usage, to protect minimum flows in rivers. In the United States, irrigation return flows are regulated at the state level. Few states or areas have imposed water-quantity restrictions, and adoption of more efficient irrigation systems is voluntary and usually costshared (Carpentier and Ervin, 2002). Both EQIP and CSP provide incentives for efficient irrigation systems. Low-interest loans are also provided to improve the efficiency of the irrigation system.

4.3.1.2. Soil quality and soil erosion Regulatory requirements regarding land use have become increasingly common in relation to soil quality, either at the national or state/regional level. For example, Switzerland’s Act on Soil Damages, introduced in 1998, requires that farming practices prevent long term soil compaction and soil erosion, in order to maintain the long-term fertility of soils. In most regions of Sweden, only 50% of farmland at most is allowed to be left fallow. In Queensland, Australia, the Soil Conservation Act 1986 requires landowners to apply for approval of “property plans”, which must specify soil conservation measures and can also be related to land-clearing practices and other aspects of land management. In 1992, Switzerland introduced legislation imposing stricter limitations on farmland use, including bans or limitations on the use of agrichemicals in specific regions such as marshes and wetlands. In Germany, the Federal Soil Protection Act, which incorporated the EU directive on soil protection into national law, was introduced in 1998. It defines the good farming practice a farmer has to adopt in order to prevent soil erosion on compaction, loss of soil organic matter, or the reduction of soil biodiversity. If a farmer does not apply the good farming practice and environmental damage occurs, the authorities can take enforceable action to prevent further damage occurring in the future and can impose an administrative fine. The Federal Soil Protection Ordinance, which is based on the Federal Soil Protection Act, sets out certain limits for heavy metals and other contaminants in agricultural soils. In the United States, erosion control has received more assistance than any other US agricultural conservation practice (Carpentier and Ervin, 2002). Most regulations are implemented at the local level and exhibit great diversity. For example, some states (e.g. Virginia, Texas) take enforceable action after pollution has occurred, while others prohibit states from taking AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 159

action against a farm that has implemented best management practices even if the farm is causing pollution (e.g. Iowa). Some governments have established regulations to control nutrient leaching by requiring farmers to maintain a minimum amount of green cover during certain times of the year (catch crops). Requirements for catch crops are most stringent in Denmark, Sweden and Switzerland. In Denmark, for example, the proportion of winter crops on all farms must not fall under 65%, while in Sweden, in the southern and central part of the country, at least half of the cultivated land must be under green cover during the winter period. Buffer strips around water courses and ground water sources have also become a common requirement in many OECD countries, including Australia, Canada, New Zealand, Switzerland and some northern EU member states.

4.3.1.3. Air quality The burning in the field of crop residues such as straw is the main air quality issue associated with arable crop farming. In the EU, several member states apply restrictions at local level to ensure that the process is conducted safely. Germany, for example, applies a ban on the burning of cereal straw and oilseed rape residue. In the United States problems of air quality stemming from crop burning were related to grass straw in Oregon and rice straw in California. Regulations enacted in 1997 in California prohibited burning on 62% of rice fields.3 A fund was created to find alternatives to burning the rice straw – a practice common to approximately 99% of rice growers at the time. The fund is cost-share among the state rice board, the rice growers, and other public funds. Rice straw tax credits are also given to off-farm users of rice straw to develop incentives for its use.

4.3.1.4. Biodiversity and landscape Most OECD governments at federal and provincial/state level have implemented legislation focusing on the impact of agriculture on landscape and biodiversity (for example in the area of the protection of valuable farm and non-farm habitats and species). Such legislation includes the control of protected and invasive alien species as well as the protection of wetlands from drainage – or the protection of bush or forest from clearance – for farming. These measures have been shaped by international as well as domestic considerations, including the obligations of OECD member countries to stem the loss of biodiversity under the International Convention on Biological Diversity, which was agreed at the UN Conference on the Environment and Development in 1992.

160 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

In the EU, the Birds Directive (No. 409/79) and the Habitat Directive (No. 43/92), oblige all member states to identify and maintain at favourable conservation status a large number of important habitats and species, many of which are found on farmland. However, most EU member states are still in the process of devising measures to address this obligation fully (Brouwer, Dwyer and Baldock, 2002). Similarly, in the United States, the Endangered Species Act (1973) protects endangered species and their habitats, and requires federal permits for certain practices such as filling wetlands for the purpose of agricultural production. Many OECD countries have also legislated to protect valuable non-farm habitats often found adjacent to farmland, such as wetlands, hedgerows, bush and forests. In the United Kingdom, legislation was passed in 1997 to protect the most important hedgerows on farms from deliberate removal. Legislation is also in place to protect margins alongside farms in Denmark and other EU member states. In France, as part of the landscape protection measures, a compulsory environmental impact assessment was introduced to the process of land consolidation programmes in 1984. In Germany, the Federal Nature Protection Act contains requirements for farmers designed to protect biodiversity such as safeguarding of landscape elements, prohibition to convert grassland to cropland in sensitive areas, field specific documentation of the use of fertiliser and plant protection products.4 Moreover, regulatory measures to protect agriculture from invasive species are common in countries where farm production and ecosystems are most vulnerable, such as Australia and New Zealand.

4.3.2.

Cross-compliance mechanisms

Eligibility for agricultural support payments to farmers is often coupled to various environmental performance standards that oblige farmers to comply with such support, or face its reduction or complete withdrawal.5 As shown in Annex Table 4.A2, although implementation of cross-compliance measures has been variable in OECD countries, it is particularly important for arable crop producers. Cross-compliance measures were first introduced in the United States as part of the 1985 FSA, subsequently amended by the Farm Acts of 1990 and 1996. They have been used principally in an effort to control soil erosion by encouraging farmers to adopt appropriate management practices for vulnerable arable land; by reducing the incentives for converting grassland on highly erodible soils to arable; and by discouraging farmers from converting wetlands into arable lands.6 The 1985 FSA introduced three new measures that imposed crosscompliance conditions on eligibility for regular farm support programmes. These measures all had explicit environmental and conservation objectives.7 AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 161

The conservation compliance provision required farmers with highly erodible land (HEL) to present a conservation plan for approval by 1990 and to implement it fully by 1995. Similarly, the sodbuster provision, which aimed to reduce the rate of soil erosion, denied programme participation to anyone who, after December 1985, ploughed up HEL in contravention of, or without having, a conservation plan. The swampbuster provision imposed a cross-compliance condition on participation in support programmes in order to prevent the conversion of wetlands to arable use after December 1985. Because of the broad participation in farm support programs, large areas of farmland - some 44 million ha of highly erodible cropland and 31 million ha of wetlands - are subject to these compliance requirements. Producers who violate these conservation provisions may be denied federal commodity support programme payments, including price support, CRP payments, storage facility loans, disaster payments, PFCP (for 1996-2002), WRP and LDP payments. Compliance requirements are essentially retained in the 2002 FSRI Act. Farms participating in EQIP must implement structural and land management practices, or develop a comprehensive nutrient management plan in order to be eligible for payments. In the first 6 years of the sodbuster provision, 1 185 cases of noncompliance were recorded, resulting in a loss of USD 6.4 million of programme benefits. The environmental impact of the sodbuster provision appears to have been difficult to assess (the decline in the ploughing-up of HEL observed in the years following the legislation may have been driven more by market conditions) (Baldock and Mitchell, 1995). The swampbuster measure is judged to have been rather more successful in cutting the conversion rate of wetlands to less than 10% of the pre-1985 rate (OECD, 1997). The swampbuster provisions are estimated to have discouraged conversion of 0.6-1.34 million wetland hectares to agricultural uses. Rates of compliance are very high. In assessing the conservation compliance provision, Horan, et al. (2001) reported a non-compliance rate of only 3%, with an additional 3.8% of farmers switching to an adapted plan due to unforeseen circumstances. According to these authors, a few evaluations of this provision have reported win-win outcomes (higher income and reduced soil loss), whilst others have found significant erosion reduction, but moderate cost increases. However, although the scheme has been very successful in reducing erosion, it has applied only to eligible (programme crop) farmers. Horan, et al. (2001) also reported the finding of a national survey of producers that 73% did not expect conservation compliance to decrease their earnings. This raises the intriguing question of how much improvement, if any, might have been obtained with a “softer” approach (such as programmes to increase environmental awareness and motivation).

162 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Claassen, et al. (2004b) found that about a quarter of the 40% decline in soil erosion on US cropland can be directly attributed to compliance. However, a large share of cropland erosion reduction occurred on land that was not subject to compliance requirements, suggesting that other factors, such as technology, information and market conditions, played an important role. In the EU, environmental rules emerged during the 1990s, reflecting the growing efforts to integrate environmental consideration into agricultural policy. In the 1992 CAP reform, environmental measures were attached to the management of compulsory set-aside in arable cropping, and EU member states were given the possibility of attaching environmental conditions to direct payments in the beef and sheep sectors. The Agenda 2000 CAP reform created opportunities for environmental conditions to be attached to all sectors covered by CAP payments, and cross compliance has become a policy concept. Since 2000, under the First Pillar of the CAP, compulsory cross compliance was introduced for set-aside (Article 19.4, Regulation 2316/1999). In 2001, farmers receiving payments under the Small Farmers’ Scheme (Regulation 1244/2001) were required to keep their entire farm in “good agricultural condition”, as defined by their member states. Moreover, cross compliance under the Second Pillar of the CAP has become mandatory in regard to payments for environmentally sensitive areas and less-favoured areas (LFAs), with farmers being required to comply with good farming practices (GFP) in order to be able to benefit from Compensatory Allowances in the LFAs throughout the entire farm.8 Each EU member state is obliged to define GFP in its rural development plans, and the new member states have also been required to define GFP for their pilot agri-environmental programmes under SAPARD. The recent CAP reform agreed in June 2003 makes cross compliance compulsory and the SFP to farmers will be linked, inter alia, to the respect of environmental, food safety, animal welfare and plant health standards, as well as the requirement to keep all farmland in good agricultural and environmental condition. Out of the nineteen pieces of legislation, five are environmental. The SFP entered into force in 2005, although EU member states may delay implementation up to 2007, while cross compliance shall be applied as of 2005. In the EU, implementation of past voluntary cross compliance has been patchy and has tended to focus on relatively specific farm management activities (Baldock, et al., 2002; Dwyer, Baldock and Einschütz, 2000). Cross compliance on support payments for arable crops has been introduced by a number of EU member states: Denmark, Greece, Finland, France, Italy, the Netherlands and the United Kingdom. In France, cross compliance aimed at reducing over-abstraction of water in the irrigated maize sector was introduced in 2000. Farmers have been required to obtain appropriate permits from the authorities in order to qualify for direct AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 163

payment. The compliance measure was strengthened in 2001, from which date farmers are also required to register the abstracted water quantity (Petersen and Shaw, 2000). In Denmark, the environmental conditions attached to arable area payments and livestock premia in the beef sector include a compulsory fertiliser plan, the maintenance of green cover throughout winter on arable land and a prohibition on the cultivation of twometre strips adjacent to rivers and streams. Penalties for non-compliance can equal up to 6% of the payments. In Finland, cross-compliance measures related to arable crops impose environmental requirements such as riparian buffer strips of 60 cm in width, green cover on arable land, requirements for land to be set aside in certain cases and some regulations concerning the use of fertilisers and inorganic waste. Penalties for failure to comply with the environmental conditions entail a reduction of payments up to 10%. In Greece, compliance with environmental conditions has been compulsory for the arable crop sector since 2001. In Italy and Spain the environmental requirements focused primarily on soil management, to minimise surface water run-off. In the United Kingdom, arable area payments, including setaside payments, are made if farmers respect certain environmental conditions regarding the management of set-aside land, over-grazing and the unsuitable supplementary feeding of animals. These are designed mainly to protect habitats and species in cropped landscapes, partly through minimising damage to ground-nesting birds and other species which may breed or feed in set-aside fields. Environmental conditions include the retention of traditional field boundaries adjoining set-aside land; restrictions on the timing of certain operations on the land (including ploughing, spraying and the sowing of new crops); the establishment of green cover by natural regeneration, sowing grass to provide shelter for wild birds, etc.; the avoidance of pesticide and herbicide applications without prior approval from DEFRA; and restrictions on fertiliser application (organic and inorganic). Breach of these national set-aside management rules entails penalties of a flat rate of EUR 140 per hectare. Korea introduced a scheme of direct payments per hectare for paddy crop field farmers in 2001, conditional on the promotion of environmental conservation – including reduced use of fertilisers and pesticides – and the submission of farming records to the authorities. The area-based payments to cereal and oilseed producers in Norway under the Acreage and Cultural Landscape Programme are granted on the condition that farmers meet the “cultural landscape” requirements, which were introduced in 1991. Specifically, to be eligible for area payment farmers are not allowed to: close or canalise open streams and ditches; cultivate areas such as border zones or forest edges; remove stone walls; level fields; close walking paths; use pesticide on border zones; or farm

164 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

within two metres of a watercourse. Additional requirements can be placed on land at risk from soil erosion. The payments are divided into two parts: an “acreage” part (differentiated into four field-size classes and four regional zones) and a smaller, “cultural landscape” part (differentiated into two regional zones); an extra area-based payment is given per hectare of root crops. Cross-compliance requirements are also placed on the receipt of headage payments, which account for around 10% of budgetary support. Since 1999, all direct payments offered to arable crop farmers in Switzerland have been made on condition that farmers comply with environmental standards and farm-management practice requirements. In addition to the structural and social criteria that are decisive in qualifying for direct payments, fulfillment of specific ecological preconditions is indispensable. These environmental standards include the need for all farmers to comply with baseline environmental legislation expressed through four federal laws concerning environment, nature conservation, water protection and animal welfare. The required environmental services serve as minimum standards to which the farmer has to prove compliance through a certification process. Criteria for the proof of ecological performance include measures to ensure minimum nutrient loss, annual crop rotation to maintain soil fertility, compliance with crop-specific soil protection indices to prevent erosion, restricted use of plant protection products and an appropriate share in ecological compensation areas (Hofer, 2000). Violation or infringement of the relevant requirements may lead to a reduction or even a refusal of the payment. A more detailed summary of the compliance requirements is outlined in Box 4.1. The most important measures affecting Swiss arable crop farmers are area payments, payments for organic farming and payments for extensive cereal and rapeseed farming. Under the Organic Farming Programme, which came in operation in 1993, an annual payment per hectare of crop production is granted on condition that it is undertaken according to specific organic farming standards. Under the Extensive Grain and Rape Production Programme, an annual payment per hectare of bread cereal, fodder cereal or rapeseed produced is granted, on condition that the whole production is undertaken without plant growth regulators, fungicides or insecticides. The programme was introduced in 1992 to reduce soil and water pollution by chemicals. The area payment per hectare of agricultural land, introduced in 1999, is granted independent of any requirement to produce particular crops. The payment is subject to an income and asset ceiling. In 2002, area payments accounted for over half of the total direct payments, while payments for organic farming and for extensive cereal and rapeseed farming represented just one per cent each. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 165

Box 4.1. Cross-compliance requirements relating to crops in Switzerland Balanced use of fertiliser In order to reduce the loss of nutrients from the environment and to keep the cycle as closed as possible, nitrogen and phosphate inputs must be calculated in terms of the plants’ requirements and individual farm’s potential for production. The use of fertilisers has to be balanced, but surplus inputs of up to 10% are tolerated. At least every ten years soil analysis must be conducted for each plot of land to establish the soil’s nutrient reserves and adjust the applications of fertiliser needed to maintain soil fertility. Plots using no added fertiliser, such as extensive grassland meadows, are excluded.

Regular crop rotation In order to avoid monoculture, maintain the fertility of the soil and ensure plant health, an annual crop rotation plan must be devised to include at least four different crops. On farms with more than 3 ha of open land, the main crops must occupy the majority of land under rotation; pauses between crops may also be stipulated.

Appropriate soil protection Soil protection indices are defined for each crop. In order to reduce soil erosion and the loss of nutrients or plant health products, farms with more than 3 ha of open land are required to achieve a certain number of points as an average protection index for field crops.

Environmental Compensation Areas for semi-natural habitats At least 7% of the holding must be devoted to “environmental compensation” (3.5% for special crops such as vineyards and orchards) in order to promote diversity of flora and fauna. Farmers can choose between 15 different habitat types/feature, e.g. extensive meadows and pastures, and crop strips free of fertilisers and pesticides. Grassy strips of at least 0.5m in width must be maintained alongside paths and alongside watercourses, stretches of water, hedges, wooded riverbanks and forest edges, this increases to at least 3m wide. Source: Hofer (2000).

4.4.

Advisory and institutional measures Advisory and institutional measures include collective projects to address environmental issues, and measures to improve information flows to promote environmental objectives. This information can be provided to producers in the form of technical assistance and extension and to consumers, via labelling.

4.4.1.

Research and development

Continuing innovation in the agricultural sector is generally recognised as crucial to attaining environmental goals. Across OECD countries, funds for research cover a broad range of scientific enquiry, including engineering, farm management practices and farmer behaviour. Increasing attention is being given to environmental public goods or public interest, as opposed to traditional production enhancing research.

166 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Across OECD countries, various measures and initiatives exist focusing on the development, availability and application of technologies for conservation and for nutrient plans. In the United States, a nationwide programme has been put in place to conduct research on new alternative crops and on agricultural technology, with the aim of reducing agriculture’s adverse impact on soil and water resources. In the EU, research focuses on detecting and preventing plant diseases, finding ways to make organic farming more competitive, and on management practices for landscape ecosystems that encourage and support biodiversity. Food quality and safety is a key thematic priority area under the European Commission’s Sixth Framework Programme for research, which has a budget of EUR 685 million for the 2002-04 period. In New Zealand, funding is available for research on sustainable land management and sustainable farming methods (including organic farming and methods of reducing chemical usage, such as minimising herbicide, fungicide and pesticide applications and integrated pest management control). Further, funding also exists for the provision of advice on science policy relevant to agriculture and forestry: for conducting operational research focused on identifying, monitoring and reducing the adverse effects of agricultural production and processing on the quality of land, water and air. Another area of research aims at improving scientific understanding of the agricultural sector’s contribution to greenhouse gas emissions. In Canada, crop nutrient management is one of the priority areas to be addressed by such research under the Climate Change Funding Initiative for Agriculture programme. In some countries, research funding is increasingly being channelled through joint agreements with industry. For example, Australia, Canada and New Zealand have created special research institutes jointly funded by government and industry, with a specialised agricultural research focus. Similar co-operative research programmes are in place in the Denmark, the Netherlands and the United Kingdom.

4.4.2.

Technical assistance and extension

These measures aim at promoting sustainable practices in farming by providing technical advice and information on the development of best-management practices and by monitoring environmental impacts. Arable crop producers benefit directly from such measures. These programmes have traditionally focussed on improving on-farm productivity, but over the past two decades much greater emphasis has been placed on

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 167

increasing farmers’ understanding of resource and environmental issues (OECD, 2003e). In Canada, crop farmers are provided with assistance on greenhouse gas issues and agricultural best-management practices, such as crop rotation, improved methods of fertiliser application and reduced tillage, through the Climate Change Skills and Knowledge Transfer Program (CCSKTP), operated by the Soil Conservation Council of Canada. The CCSKTP, which came into force in 1999, is a nationwide programme aimed at raising farmers’ awareness of the impact of climate change on the agricultural sector. A wide range of technical assistance programmes is also offered in the EU. For example, training and demonstration projects have been introduced by member states under the Agri-environmental Regulation (No. 2078/92) and the Rural Development Regulation (No. 1257/99). In Belgium (Flanders region), crop protection is one of the Five Codes of Good Agricultural Practices. Denmark provides training and education concerning organic production. Under the Pesticide Action Plan, it also provides technical assistance and advice on the use of fertilisers through information groups, advisory services, training and demonstration. Technical assistance is also used to assist farmers in meeting the requirements of environmental standards such as those stipulated by the Nitrate Directive. In Japan, under the Law for Promoting the Introduction of Sustainable Agricultural Production Practices introduced in 1999, emphasis is laid on the provision of technical assistance, and information on the reduction and use of chemical fertilisers and pesticides. In the United States, the Conservation Technical Assistance programme provides assistance to farmers with the planning and implementation of soil conservation and water-quality practices. Technical assistance and extension are also provided as part of the major environmental cost-share assistance and conservation programmes, such as EQIP and the CRP. Under the 2002 FSRI Act, producers may obtain technical assistance from providers other than USDA’s Natural Resources Conservation Service for the preparation of conservation compliance plans. Over time, the provision of information has also tended to encompass an increasingly comprehensive range of information, for example, Environmental Farm Plans in Canada, which focus on developing riskmanagement strategies for farmers, and Australia's Environmental Management Systems, which integrate individual environmental farm objectives with regional targets. A common feature of technical assistance in recent years in OECD countries is the increasing use of the Internet as a tool to distribute

168 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

information and best-management practices to farmers. For example, in the United States, the federal government maintains databases on a wide range of subjects, from insect infestations to soil types and natural resource inventories, and provides network access to them over the internet. In the United Kingdom, farmers can now access comprehensive Codes of Good Agricultural Practice for soil, water and air in the same way. In particular, these Codes provide practical guidance to farmers and growers engaged in commercial crop production, particularly in regard to the regulations controlling the use of pesticides. They also provide guidance for those involved commercially in the sale, supply and storage for sale of “pesticides provided for agricultural use”. The Code provides guidance on meeting the obligations imposed on individuals and companies involved in these activities under UK and EC legislation. Technical assistance has also been provided to assist farmers to maximise economic returns from fertiliser use whilst minimising the risk of pollution. Some countries, such as Australia, Canada, New Zealand and the United States, make widespread use of community-based approaches to resource management in rural regions, with the purpose of mobilising and motivating citizens to take on greater responsibility for addressing environmental issues. Much emphasis is placed on the exchange and transfer of information. For example, Australia’s National Landcare Programme, which originated in the mid-1980s, aims to encourage community groups to develop a self-help attitude and capacity in planning, promoting and using sustainable land, water and vegetation management practices. Around onethird of farming families now participate in Landcare groups. A number of more recent community-based environmental programmes is also being applied which has implications for farming regions to address regional water-resource problems, for example, the National Action Plan for Salinity and Water Quality and, the Murray-Darling 2001 Programme and the National Rivercare Programme. In the United States, support is provided for a range of local resource conservation projects affecting agriculture. For example, under the Small Watershed Rehabilitation Program, up to USD 35 million is provided annually in support to projects initiated by local groups to protect watersheds of less than 100 000 ha, including cost-share support for structural and non-structural improvements to reduce erosion, sedimentation and run-off. Funding under this programme is budgeted to increase significantly under the 2002 FSRI Act. Community-based support for environmental purposes in rural regions is also available in some EU member states under the Rural Development Regulation (No. 1257/99). For example, in England, the Rural Enterprise Scheme, introduced in 2000, provides support to local projects involving AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 169

farmers, rural businesses and rural community groups, with the aim to protect the environment in the sphere of agriculture, forestry, landscape and animal welfare. In the Netherlands, a number of farmer-led environmental initiatives at the local level was established in the 1990s (OECD, 1998c). The Rural Development Regulation measures of the 2003 EU CAP reform (Reg. 1782/2003) include a new Farm Advisory System. Member states will be free to establish such system on a voluntary basis for until 2006; from 2007 onwards, however, they will be obliged to offer advisory systems to farmers, although farmers’ participation will be voluntary. Support will be provided to farmers to help with the costs of using farm advisory services, up to a maximum of 80% of the cost of such services, subject to a ceiling of EUR 1 500. In 2010, the Council shall decide whether the advisory system should become compulsory.

4.4.3.

Product information

A number of OECD countries have established specific eco-labelling standards to enable consumers to distinguish products grown without chemical fertilisers or pesticides from conventionally produced agricultural products. Arable crop producers are affected by these measures, particularly those concerning organic crop production. In the United States, there are at least 25 major labelling schemes for goods produced using environmentally friendly practices. Certification schemes can be private or public, such as the first national standards for the labelling and processing of organic food, adopted in 2000 (OECD, 2001c). In the EU, one of the most important initiatives has been the introduction of EU-wide legislation covering organic crop production and products (EC Reg. 2092/91). Some of these labelling schemes are entirely market-based, while others are government-based. Many countries in the OECD area, including Australia, the EU, Japan, Norway and Switzerland, introduced government-enforced national organic labelling standards over the past decade. Canada’s National Standard for Organic Agriculture, introduced in 1999, was implemented by the government, but the labelling criteria were developed by industry, while, in New Zealand, industry groups are steering labelling activities for organic products. The development of the marketing structure and establishment of new retail outlets has received considerable attention in many countries. Policy support for marketing and processing in organic farming varies considerably. For example, Austria, Germany, Italy and Denmark have national programmes that specifically target organic farming. In the EU measures exist to provide information on, or promote, agricultural products and food on the EU internal market, including participation

170 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

at events and fairs, information campaigns on the EU system of protected designations of origin, protected geographical indications, and guaranteed traditional specialities, and information on EU quality and labelling systems. The EU co-finances these measures up to an amount not exceeding 50%, the remainder being paid by the professional organisations that proposed them and by the EU member states concerned. The annual budget appropriations are approximately EUR 40 million. Under the Rural Development Regulation measures of the 2003 EU CAP reform (Reg. 1782/2003), incentive payments will be made available to farmers who participate in recognised schemes designed to improve the quality of agricultural products and the production process used, and give assurances to consumers on these issues. Such support will be payable annually for a maximum 5-year period, and up to a maximum of EUR 3 000 per holding in a given year. Support will be provided to producer groups for activities intended to inform consumers about and to promote the products produced under quality schemes supported under the above measure. Public support will be permitted up to a maximum of 70% of eligible project costs.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 171

Annex 4.A. Selected Data Table 4.A1. Japan’s Rice Diversion Programme R ice pa ddy fie ld diversion program m es Year 1971 1980 1985 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

A rea dive rted

Total subsidy paym ents

(1000 ha) 541 585 594 849 852 751 713 588 663 787 798 955 960 969 973 978

(billion yen) 171.2 357.8 222.5 159.8 156.7 130.0 90.2 63.3 77.4 131.2 130.5 109.5 107.8 129.0 150.2 167.7

Land u se on diverted rice pad dy fields in 2000 (1000 ha) a) Types of land use elig ible for governm en t paym ents P la nting alternative crops in paddy fields M aintaining idled paddy fields C onverting pad dy fie lds to other uses S ub-total b) Types of land use not eligible for governm ent paym ents

55 0 13 0 7 68 7 27 7

M ain alte rnative paddy field crops, 2000 (1000 ha) S oya bean

86

P asture W heat S oil-im proving crop s B uckwheat S orghu m R ed be an Flowe rs C orn

79 75 73 27 12 12 12 10

Source: MAFF.

172 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 4.A2. Use of cross-compliance requirements in OECD countries Country Australia Austria Belgium Canada Czech Republic Denmark

Year introduced NO

Set aside land

2000-2003

2000

2000 France Germany

2000 NO

Greece

2001

Hungary Iceland Ireland

NO NO 1998

Italy

2001

2001

Compulsory fertiliser plans, the maintenance of green cover over winter on arable land, and a prohibition on cultivating land equivalent to a 2m strip along rivers and streams. Beef Legal requirements for the use of fertilisers and manure. Riparian buffer strips of 1m width, green cover on arable land, Arable crops, hemp, flax, requirements to be set aside in some cases, and some regulations concerning the careful use of fertilisers and inorganic potato and seeds waste. Maximum stocking density, preventing overgrazing, and requirements for farmers to maintain sufficient stock to prevent Livestock undergrazing. Extra measures: careful use of inorganic and organic fertilisers. Arable crops Compulsory authorisation to irrigate from the water authorities. Arable crops, cotton, sheep and goats

Area payments in Nitrate Vulnerable Zones; sheep and goat headage premia.

Sheep Arable crops, set aside land, grain legumes, flax, hemp, tobacco, seeds, rice, olives

Limited number of sheep in areas vulnerable to overgrazing.

Sheep and beef

NO 2001 NO NO

Netherlands

2000

Silage maize

2000-2003

Starch potatoes

Paddy field farmers

Maintenance of the outlet rill; permanent draining ditch and creation of temporary water gullies perpendicular to the maximum slope (the latter does not apply for olives). Conditions on the storage of slurry from in-house livestock in specific facilities must be respected. Area payments

Integrated weed control and maximum limit on the application of herbicides and pesticides of 1kg/hectare. Use of mechanical means for removing potato haulm and no use of chemicals for killing off potato leaves and stems on 70% of the crop area.

NO

Norway

1991

Poland Portugal Slovak Republic Spain Sweden

1991 NO NO NO NO NA

Switzerland

1999

United Kingdom

United States

No use of waste water, sludge, compost, pesticides

Arable crops

Japan Korea Luxembourg Mexico

New Zealand

Method

NO NO NO

2000-2003 Finland

Commodity coverage

1985

Arable crops, oilseeds, fruits and vegetable All livestock

Area payments Headage payments

All farmers receiving payments Environmental conditions for the management of set-aside land such as restrictions on the timing of certain operations on the land, including ploughing, spraying and sowing of new crops; Arable crops, set aside establishment of green cover by natural regeneration, sowing land wild bird grass etc., avoidance of pesticides and herbicides without prior approval from DEFRA and restrictions on fertliser application. Sheep and beef Limited number of cattle and sheep to prevent overgrazing. All sectors Sodbuster and swampbuster provisions

Source: Petersen and Shaw (2000); OECD Secretariat.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 173

Notes 1.

The initial EQIP legislation required that at least 50% of funds be targeted for natural resource concerns related to livestock (USDA, 2003a). In practice, during the1997-2000 period, 62% of available funds went to contracts with livestock producers.

2.

Rice diversion measures have been classified as an environmental measure within the URAA, and are thus not subject to support reduction commitments.

3.

Curey, Sumner and Howitt (2000) found that the phasing out of rice straw burning in Sacramento Valley cost the rice industry around USD 28.8 million between 1992 and 1998 and trading burn credits saved the industry about USD 5 million. The authors argue that an extension of the trading allowance would realise important gains (e.g. burning will be reduced to around 18% of planted rice acreage at the same cost to the industry as a 25% burning allowance with no trading).

4.

The Act is a framework law which the States (Länder) are required to implement through their own laws.

5.

The term was developed to help integrate environmental concerns into agriculture, and in particular to combat the detrimental impacts of agricultural intensification. Broadly speaking, the concept of cross-compliance as a policy term refers to the linkages between environmental and agricultural polices (Baldock and Mitchell, 1995).

6.

In 1977, a cross-compliance strategy was introduced to improve the operation of the Acreage Reduction Program (ARP) as a supply control measure for wheat, feed grains, cotton and rice, and had no environmental objectives. Participants claiming a payment for one of these commodities had to comply with set-aside provisions relating to other commodities for which they had base acreage, even if they were not claiming payments under other programmes that year.

7.

According to Allen (1990), the 1985 FSA was able to impose tough conservation measures because at that time the farm organisations and the environmentalists formed a united front – the former wanting more supply control, the latter more soil conservation.

8.

Although programmes to support the continuation of farming in areas considered economically marginal (because of difficult growing conditions, or because of the danger of becoming depopulated) have often been established with rural development objectives in mind, they are increasingly seen as also contributing to preserving landscape values and preventing the abandonment of extensive farming systems.

174 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Chapter 5 ENVIRONMENTAL EFFECTS OF AGRICULTURAL SUPPORT POLICIES FOR ARABLE CROPS 5.1.

Introduction The environmental effects of agricultural production in developed countries have already been extensively documented in the literature.1 In many areas, the environmental consequences discussed in the previous chapter are the result of several decades of accumulation and environmental loadings may have been high for some time before public concern was raised.2 In other cases, degradation of marginal land and of landscape, retreat of indigenous species from particular areas, or serious groundwater pollution have occurred relatively recently and with alarming rapidity. In the last 15-20 years, awareness of the full extent of these problems has increased throughout the OECD area. Governments are now responding to public pressure – and in some cases taking the lead – in seeking ways of containing or reversing these trends. Traditionally, in capitalist democracies, land ownership has conferred on landowners a full set of exclusive de jure property rights regarding the use, transfer and maintenance (or neglect) of their land. Moreover, land ownership conferred extensive de facto rights over the use of other natural resources, such as the ambient atmosphere or the local stream, but to which property rights were not explicitly assigned. The degradation of these natural resources represented an input into farming activities against which no payment was required to be made. The negative externalities discussed in the previous chapter are instances of market failure, where an agent’s actions (the farmer’s), legally pursued, given the current assignment of property rights, reduce the economic benefits or utility of third parties (people who also use or value the same natural resources in their undegraded state), who are unable to seek redress through normal market interaction because they lack any legally enforceable claim. Bromley and Hodge (1990) observed that, with industrial development, property rights concerning land became translated through the political process into “presumptive entitlements in the policy arena” that have persisted over time, so that “any change in the status quo production domain of the farmer must inevitably be purchased by the state with bribes” (i.e. “inducements”, “subsidies or concessions”). An alternative, more benign theory is that agriculture’s “favourable treatment” has stemmed from widespread social acceptance of agrarian fundamentalism, according to AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 175

which rural life embodies certain intrinsic values and beliefs that have public good characteristics. Regardless of whether social historians can reach consensus on explaining the development of property rights claims and policy-based entitlements in agriculture, it is interesting to bear in mind the important political economy dimension that underlies the “greening” of agriculture, and the policy options which will be discussed in this chapter. The contemporary policy agenda that aims to control and internalise the externalities produced by farming involves nothing less than establishing and extending the rights of society at large to determine how rural resources can be used by farmers. This implies a shift in farmers’ de facto and de jure property rights. The OECD’s annual monitoring of transfers from consumers and taxpayers to agriculture shows the extent to which society in most member countries is still prepared to honour the long-standing policy entitlements of the farm sector (OECD, 2003d). Some environmentalists, economists and others contend that the stimulus these entitlements have provided to farmers in developed countries, extending over many decades, and the fact that they have largely taken the form of price support, have exacerbated the negative environmental effects of modern farming.

5.2.

Environmental effects of agricultural support policies 5.2.1.

Links between high arable support and negative environmental effects

Arable production has been a particular target of the observation that high levels of support, principally price support, have led to environmental deterioration. The question is: what would the likely environmental consequences be if that support were reduced or delivered in a different way? In most OECD countries, time series of agricultural support (as measured by the PSE), production indicators (e.g. cropland area and crop yields, agrochemical use and machinery power per hectare) and environmental indicators (such as nitrates (mg/l) in drinking water, sedimentation in navigational waterways, species loss in intensively farmed areas and so on) all show long-term increasing trends over the second half of the last century. Casual inspection reveals strong correlations between these variables over time. To quantify a direct link between support and environmental indicators, Lewandrowski, et al. (1997) used panel data for 22 countries, including Australia, Canada, the EU, Japan, New Zealand, Korea and the United

176 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

States, to examine the relationship between producer support (measured by the PSE) and fertiliser use per hectare farmed and per total arable area, which served as environmental indicators. They found a strong positive link between support and fertiliser use per hectare for both high- and low-income countries. By contrast, support had no impact on total arable area in lowincome areas and was negatively related to arable area in high-income countries. The authors interpreted this last result as showing the effect of higher support on intensification, and hence as potentially negative for the environment. This kind of study provides statistical evidence of an association over time between support and environmental damage, but it does not demonstrate causality. A causal link requires two mechanisms to operate: first, support must stimulate farmers to change their management practices and rates of input use, and second, these farm management changes must have environmental consequences. A causal link is a necessary but not sufficient condition for supposing that reducing support will reduce environmental damage. Production economics provide strong theoretical support for the first of these mechanisms. According to theory, the profit-maximising producer responds to an increase in output price by increasing supply of that output, as well as use of productive inputs. Such a producer will react to a reduction in the price of an input via a subsidy by increasing the use of that input and other inputs used together with that input, while possibly reducing the use of any substitute inputs, and certainly by increasing output. Thus, price support and input subsidies both provide incentives for output expansion and intensification of input use. In a multi-output setting, theory predicts that support to a number of crop commodities, even if delivered at a uniform rate, will alter the mix of crops produced because of different crop responses (some crops use inputs more productively, or require higher levels of input use for efficient yield performance). If higher levels of support are given to high-performance crops that are more input-intensive, then the impacts on input use and crop mix will be even greater. Thus, support will also alter the mix of crops grown, which may not be neutral for the environment. It is important to identify a third longer-term effect of agricultural support. When high levels of support are maintained over time, this stimulates the development of new yield-enhancing and cost-reducing technologies, which will be biased in favour of those crops receiving the highest support. Indeed, it can be argued that the greatest impetus for longterm intensification and environmental deterioration has come not from producers’ short-term reallocations of existing resources with existing AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 177

technology, but rather from the rapid adoption of new technologies, which have been largely triggered by sustained high support levels. Support may slow down structural change in the sector by allowing smaller farms to survive, thereby also slowing down the adoption of labour-saving, scaleincreasing technology. At the same time, the arrival of these new technologies and the capitalisation of their benefits into land prices enhance the underlying pressures for farm consolidation and production intensification. Regarding the second link, which relates to production changes and environmental outcomes, theoretical generalisations are more problematic. The environmental impact of farm management changes is site-specific and in some cases also weather-dependent. Therefore, in general terms, a change in input or management practices can only be described as having a potential positive or negative environmental impact. Nonetheless, at a conceptual level this second link is clear. Many attempts have been made to model the supportoenvironment nexus by coupling a short-run profitmaximisation production model and a model for deriving expected or average environmental effects based on farm management decisions. There is a great deal of empirical work based on this framework of a two-stage linkage between support and the environment. For example, Tobey and Reinert (1991) used a general equilibrium model with a highly aggregated depiction of the US agriculture sector to compare the changes in environmental damage arising from (1) an increase of 14 million acres in CRP enrolment and (2) a 40% reduction in output-enhancing support with a 20% reduction in land under the ARP. The model was enriched by distinguishing highly erodible and non-erodible land and by incorporating an environmental damage function, which depends on fertiliser use and on the area of each type of land used for production. This model showed that the greatest reduction in environmental damage occurs when price support alone is reduced. Even when the ARP area is decreased at the same time, the environmental improvements exceed those obtained by increasing CRP enrolment alone. Using farm-level mathematical programming models, Painter and Young (1994) compared the environmental impacts of the US 1990 Farm Bill with different alternative policy packages (including increased planting flexibility and termination of all commodity programmes). This was done for two contrasting regions in the United States (a diversified cropping region in North Carolina and a Washington-Idaho dryland grains region). They found that, with no support, nitrogen leaching fell significantly in North Carolina, whereas the erosion rate, and on- and off-site erosion damage, were not affected in Washington-Idaho unless incentives were also offered for farmers to adopt cropping patterns based on resource-conserving

178 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

alternative crops like low-input corn. LaFrance (1992) used a theoretical model to study the impact of different types of support measure on soil degradation and showed that, as far as output price support is concerned, the erosion impact could be in either direction, depending on the relative impacts of conservation and cultivation on long-run soil stock. He also concluded that input taxes or a conservation subsidy would always reduce the rate of soil degradation. Vatn, et al. (1997) described a model in which the second link of the causal chain is depicted in unusual detail. This model was the result of a multidisciplinary effort. A production model in which producers on different types of farm and in different regions make production decisions based on relative prices and their resource and technological possibilities, is linked to a suite of models relating these decisions to hydrological processes, nitrogen turnover and soil erosion. A watershed model then aggregates the environmental effects at the level of the watershed. This model was simulated for three cropping regions in Norway in order to compare the environmental impacts of a 100% tax on the N-component of fertilisers, a rotational constraint and a subsidy for a specific soil-saving practice. The results show that “no simple solution exists for reducing non-point-source pollution from agriculture”. Policy impacts depend on: input levels; the way in which inputs are combined; and climate and soil characteristics. Although Vatn, et al. (1997) did not report an experiment in which output prices were reduced, this model could be used for such a purpose. Wu and Segersen (1995) emphasised the importance of allowing for differences in soil characteristics when simulating the environmental effects of different policy changes. They defined “polluting acreage” as land that is susceptible, because of its physical characteristics, to causing groundwater pollution and is used to grow a chemical-intensive crop. In their central Wisconsin study area, 62% of the land fitted this description. Using a set of econometrically estimated acreage equations, they compared a reduction in maize price (2.3%), an increase in maize ARP enrolment (25%) and a tax on agrochemicals (1.1%) as instruments for achieving a 1% reduction in polluting acreage. Their results show that the ARP increase leads to the greatest shifts in area between crops, and that it specifically targets maize for grain (a highpolluting crop). Thus, this instrument achieves the 1% reduction in polluting acreage with the lowest reduction (0.25%) in total cropped area. With the other two instruments, the reduction in total acreage is approximately 1%. These studies all illustrate how price support and related policies can be linked to environmental effects by explicitly passing via a production model that determines the management changes triggered by the changes in support which, in turn, impact on the environment. However, most of these models reflect the limitation that they are derived from a theoretical set-up where AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 179

production changes represent short-run management reactions to relative price changes, while the underlying technology is assumed to remain constant. It was argued above that an important contributor to the environmental effects of high support has been technological change over time. It is likely that a sustained and sufficiently large fall in support prices would not only stimulate immediate on-farm changes in input use within the range of current management options, but would also precipitate a “technology shift”, which is not represented in the simulation models based on feasible technological options during years of high support. In the absence of an empirical study of this phenomenon for the cereals sector, the results of a study on milk producers in the Netherlands are reported (Komen and Peerlings, 2001). This study showed that milk quota abolition, accompanied by restrictions on nitrogen use, caused dairy producers to shift to available but previously unused low-N technologies. In a model based on recent actual production data these options would not be reflected, and hence such a model would not predict the technological shift. A similar on-farm technology shift in the case of cereal producers could also be expected to occur. Most importantly, however, the prospect of lower future prices is also likely to encourage new directions for agricultural research and technology development (despite the lower expected profitability of the industry), so that available technological options will expand. Moreover, how respectful of the environment such new technology will be is not easy to predict, although it can be assumed that its development would conform with current policy constraints and incentives. None of the models discussed above makes allowance for such developments, which are very difficult to predict and quantify ex ante.

5.2.2.

Assessing the environmental effects of lower support

After decades of support at high levels, predictions of the environmental effects of lowering support cannot be based on recent experience and observation.3 The question has to be addressed using “what-if” model-based simulations. The studies reported above provide ex ante analyses of how producers are expected to react to future policy changes, including removal of support, and what their reactions could mean for the environment. These simulation studies are valuable in establishing a benchmark that summarises the most likely developments, given the best current information about available technology, reasonable assumptions about producer behaviour and expectations about how production changes will impact on the environment. The results are, however, highly model-dependent. In fact, there are many reasons why real events may not coincide with the simulation results from current models. These include technical and methodological limitations of the models, as well as characteristics of producer behaviour, environmental processes and the complexities of the

180 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

policies themselves. These reasons are discussed below, together with some empirical evidence and possible implications for model-based studies.

Uncertainty about producers’ reactions to large changes and time lags Many models of producer behaviour are designed to capture short-run reactions to relatively small changes. When policy changes involve substantial changes that take the model outside the range of variation that has occurred over a relevant recent period, then uncertainty about how producers will actually react increases greatly. This may be because large price changes trigger new, unforeseen technological choices, as already described, or because they provoke structural changes that alter the farmsize distribution of the sector or region and thereby change “typical” behaviour. Moreover, the time-frame over which adjustments to large policy changes will occur is also uncertain, and there will be little empirical evidence on which to form reasonable assumptions.

Asymmetry of reactions both by farmers and the environment Producers are unlikely to adjust to lower support by retreating back along the same pathway they took when adjusting to higher support. In the meantime, considerable changes have taken place in farming styles and in the supporting technology. For example, in the absence of specific incentives to do so, it is unlikely that intensive production will move very far in the direction of former levels of extensiveness (e.g. yield reductions, smaller fields, etc.). There are also asymmetric, or irreversible, processes involved in the second link (farming practicesoenvironment). It may take years of reduced environmental stress to produce any discernible improvement in environmental conditions. Whether loss of species-richness and visual amenity can be reversed will depend on local circumstances. If an endogenous process of decline has been set up by high environmental loadings, reduction of these stresses alone may not be sufficient to halt the process.

Endogenous input price changes When simulating the effects of policy changes, many variables are assumed to remain constant at “base run” levels. This may not be realistic. It is generally assumed that the prices of purchased inputs are exogenous to policy and production decisions. If inputs such as fertiliser and pesticides are purchased on free competitive markets, it is reasonable to assume that changes in the policies of one country will not affect their price. But this may not be the case. In particular, when these inputs are supplied by imperfectly competitive industries, a fall in policy-supported output prices AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 181

may be accompanied by falls in input prices if suppliers act to maintain market shares or market size. These sympathetic input price changes may absorb any incentive to reduce the use of these inputs.

Producer psychology The assumption of rational profit maximisation drives most model simulations that analyse the effects of lowering support. Whether or not farmers are totally rational in their production decisions has been the subject of debate by economists for many years. Certainly, it is reasonable to suppose that when faced by a choice involving the trade-off between production and environmental objectives, farmer behaviour will display greater psychological complexity than is assumed by the theory of the “rational producer”. In a survey of 144 of the largest (>300 ha) farms in the south of England, Walford (2002) found that the same individual farmers had responded to various elements of the 1992 CAP reforms in different ways – as “productivist” farmers (in using compulsory setaside as a management tool rather than as an opportunity to achieve environmental objectives)4 and at the same time as enthusiastic environmentalists with respect to the voluntary schemes in which they had enrolled. Saunders (1996) reported that farmers who participated in land conservation covenants in New Zealand exhibited a high degree of altruism, being willing “not only to provide conservation at a low cost, but also to surrender property rights”. The sensitivity of outcomes to detailed, but not always predictable, management decisions was underlined in the study by Ackrill, et al. (2001). Using farm-level programming models, they looked at the effects on cropping farms in eastern England of the Agenda 2000 decision to bring the oilseed rape price down into line with cereal prices. Whether oilseed acreage expanded or contracted was found to depend on whether producers followed standard agronomic advice and included just one crop break in the (predominantly winter wheat) rotation, or responded to economic incentives by including two crop breaks. Not only are net margins higher for the two-crop-break rotation, which involves an increase in oilseed area despite the price fall, but there are also implicit reductions in pesticide use through reduced disease risk and fertiliser applications.

Risk and producers’ attitudes to risk An important element of producer psychology concerns the way producers adapt their behaviour to accommodate perceived risk. Apparent farmer irrationality can turn out to be the by-product of simplistic thinking by economists. The old debate about whether farmers are behaving irrationally when they are perceived to be using inputs at “below-optimum” levels, given current prices, was in many cases resolved when economists

182 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

allowed for output price risk and risk aversion in their formal thinking about farmer behaviour. Similarly, perceived “over-use” of inputs may in fact be an insurance strategy of risk-averse producers facing yield risk. Yadav, et al. (1997) found that on the three farms they studied in south-eastern Minnesota, United States, fertiliser use over a 3-year period was consistently well above the recommended level, which was, in turn, above the economically optimum rate calculated for each farm. Residual nitrogen in the soil on these farms at the start of the growing season was found to be significant, penetrating to a depth of 3 feet (where it is most readily available to the maize crop), and on two sites going down as far as 7-8 feet (where it can never be accessed by the plant and is a potential source of groundwater pollution). Gren (1994) documented similar over-use of pesticides by Swedish producers. Risk-averse producers may consider this to be a cost-effective form of insurance.5 However, this may not be society’s point of view, as producers do not take into account the negative externalities of input over-use. Most important, such behaviour represents another source of potential inaccuracy in the results of models that predict rational reactions to price signals based on assumed risk neutrality.

Policy reform as a package deal Policy reforms are often introduced as a complex package whose separate elements may interact with each other in unforeseen ways. Simply reducing more than one price simultaneously can produce results that differ from expectations formed on the basis of simpler, ceteris paribus, reasoning. When additional measures, such as set-aside, direct payments or planting restrictions, are involved in the package, the potential for unforeseen outcomes increases. Therefore, models that ignore some aspects of a package can give misleading signals. Moro and Sckokai (1999) studied the impact of the Agenda 2000 package on cereal producers in northern Italy. They found that the direction of the change in area allocated to each crop can differ depending on whether or not direct payments are given in conjunction with the price changes, and that the sign of the change may also depend on whether the payment rate remains different for maize than for other cereals, as under the 1992 CAP reform. Whether or not input usage falls was also found to depend on whether the area payment differential for maize was maintained. Thus, unless all the various components of a policy reform package are analysed together and at their precise payment rates, outcomes can differ in ways that are not neutral for the environment.

Implementation details The findings of Moro and Sckokai are an example of the more general point that the impacts of policy reform can depend crucially on the details of implementation. Mims, et al. (1989) analysed two of the key changes AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 183

introduced by the US 1985 FSA, namely replacing the 2-year average used for calculating the programme base with a 5-year average, and introducing some limited (non-environmental) cross compliance, whereby participation in a government programme for one crop ruled out increases in base acreage of any other programme crop, even if the producer did not participate in the programme for the other crops. The move from 2 to 5 years affected only marginally the crop mix and output levels on two representative Alabama cotton farms, but when the limited cross compliance was added, this implementation detail was of major significance in determining crop shares and the incentive to participate in the programme for cotton. Mims, et al. (1989) did not translate the differences they found into environmental consequences. By contrast, Reichelderfer and Boggess (1988) showed explicitly how the details of programme design can strongly affect environmental outcomes, as well as programme costs. They examined the performance of the CRP in its first year and showed that the choice of bid selection criteria for that programme had a significant effect on both the environmental and supply control outcomes of the programme. The three criteria used implicitly gave the highest priority to maximising the number of acres enrolled, whereas, with more flexibility of the control variables, both erosion reduction and supply control targets could have been better achieved and at lower cost (see also Section 5.4). Winter and Gaskell (1998) pointed out that the precise definition of land eligible for direct payments in the 1992 CAP reform (“land under eligible crops or temporary grass” on 31 December 1991) had unforeseen negative environmental consequences because temporary grass was not designated as an eligible crop and did not attract a direct payment. Therefore, producers wishing to maximise their entitlement to payments in any given year were encouraged not to include temporary grass in their rotation. This had a negative landscape impact and increased the use of agro-chemicals. Moreover, although the total eligible area was fixed regionally (and, if exceeded, payments were to be reduced proportionately), there was no restriction on producers planting non-eligible land with eligible crops without claiming an area payment when prices were sufficiently high. At the same time, producers wishing to take advantage of high prices for noneligible crops (whose area was unrestricted) faced a disincentive to plant them on eligible area, since direct payments would be foregone. In their survey, Winter and Gaskell found that in response to such market forces, farmers had been ploughing up permanent pasture.

Unanticipated exogenous changes Model predictions depend on many assumptions about variables thought to be exogenous to the simulated changes. These variables may change in

184 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

unanticipated ways, or may in fact turn out to be endogenous. For example, in assessing ex post the environmental impact of the 1992 CAP reform, Winter and Gaskell (1998) documented that in the early years, because of high world prices for cereals that were transmitted to United Kingdom cereal prices, the direct payments fixed in advance over-compensated producers for the cereal price falls. Understandably, the environmental impact of the reform was small. Abler and Shortle (1992) examined the interaction between EU and United States policy reforms ex ante. Their scenarios involved unilateral and bilateral policy reforms, with and without nitrogen quotas. When both the EU and the United States abandon commodity support and no N restrictions are in place, agro-chemical use in the EU falls, but it increases in the United States. This simple model illustrates the point that the expected results of domestic policy changes are always conditional on assumptions made about trade competitors’ policies and world market price movements. Changes in behaviour that was assumed to be exogenous could mean that policy reform leads to unexpected changes in production decisions and environmental impacts.

Environmental and management heterogeneity The fact that the environmental effects of policies can be extremely localised is well understood. This is aptly illustrated by Vatn, et al. (1997). It means that as long as much of the analysis and design of the environmental effects of policy changes is carried out at sector level, or aggregated across environmentally heterogeneous regions, then apparently appropriate policies can fail to produce the desired environmental results. Burrell (1989) drew attention to the discrepancy in estimates of the elasticity of demand for fertiliser in the United Kingdom, which were much lower when derived from farm-level studies of specialist arable farms than for the agricultural sector as a whole. These differences were fuelling the debate over the effectiveness of a fertiliser tax for reducing nitrogen pollution. An explanation was offered in terms of specialist arable farmers having fewer options for moving to low nitrogen production than exist in the sector as a whole, and therefore showing only a small response to changes in nitrogen price. Hertel, et al. (1996) revisited this question. Even when restricting the analysis to crop specialists, and after accounting for differences in land quality, slope, crop rotation, etc., they found unexplained differences that had to be due to heterogeneous managerial ability, and which would cause the industry elasticity to be greater than the farm-level elasticity. Clearly, if the objective is to reduce agriculture’s aggregate use of fertiliser, then a uniform rate of tax calibrated on an industry-level elasticity would be appropriate. However, if the objective is to reduce N-use in areas with the highest level of N-contamination, and if these are areas where specialised AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 185

cereals farms predominate, then such a tax would fail to reach environmental targets. This example is pertinent to the issue of farmer/farm heterogeneity and to the broader discussion of the effectiveness of using price-based instruments that cannot discriminate between types of producers or regions in order to achieve environmental targets. What conclusions can be drawn so far about the impact on the environment of reducing support for arable production? The following points emerge from the above discussion:

x

Expected environmental impacts of reducing support cannot be directly inferred from simple correlations of support and environmental damage, or by assuming a simple retreat down the pathways that led to the current situation. Expected effects have to be based on sophisticated models that attempt to depict the relevant current and future choices and trade-offs involved in farmer decision-making, and the way in which these decisions impinge on environmental indicators.

x

Available models are useful in providing benchmarks conditional on their particular assumptions. However, they have a number of limitations, and predictions of likely outcomes will always be inexact and may be misleading.

x

In particular, the difficulty of predicting new technological options and discrete behavioural jumps – let alone incorporating them into models – adds to the imprecision of model-based predictions.

x

Quite independently of the ability or otherwise of models to provide precise answers, it is clear that environmental heterogeneity and site-specificity make policy changes such as price support reductions or input taxes rather blunt instruments for achieving the desired environmental outcomes.

x

In particular, the complexity of producer reactions and the heterogeneity of environmental responses make it quite possible that the expected environmental effects of reducing support do not materialise, unless they are accompanied by extra constraints and more targeted incentives.

186 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

5.2.3.

Environmental effects of shifting from price support to direct payments

This section will review evidence on the environmental effects of shifting support away from price support to direct payments. It is useful to distinguish decoupled payments at fixed rates (i.e. based on historical references such as output or area in a fixed time period); partially coupled payments at fixed rates (i.e. linked to current resource usage) such as the area or headage payments implemented in the EU after 1992; and decoupled payments varying inversely with market price, such as the CCP introduced in the United States with the 2002 FSRI Act. Production theory predicts that if producers are risk-neutral, the reduction in price support will lead to a reduction in output and variable input use. When payments are decoupled, lower price support is also likely to encourage a reduction of cropped area, depending on the extent of other land-use opportunities on the farm. When payments are coupled to current cropped area, producers are more likely to maintain or, if possible, increase their cropped area in order to qualify for payment entitlements, whilst still reducing input intensity. Hennessy (1998) demonstrated that when producers are risk averse, direct payments have two additional risk-related effects.6 If producers normally become more risk averse as their wealth decreases, and if the shift of support from price to direct payments decreases their wealth, this will have a negative impact on production and input use (the wealth effect). If the policy shift increases the variability of their income (more risk), this will have an additional negative impact on production and input use (the insurance effect). The size and direction of these risk-related effects depends on the characteristics of the policies (whether price support was given in order to reduce risk, whether direct payments are delivered in order to reduce risk and the net effect on risk of shifting support from one delivery mechanism to another) and on producers’ degree of risk aversion. A study by the OECD (2004b) analysed the extent to which various forms of support reduced the variability of producer returns for certain cereals and countries, and for the policies that were in operation in the cereal sector during the period 1986-2001. Different instruments affect price risk in different ways. Market price support always led to a large reduction in domestic price variability relative to the world market. Clearly, when more insulating border measures are used to deliver price support (variable import levies rather than tariffs), the risk reduction offered by market price support is greater. A market intervention system supports prices by truncating the range over which prices can vary at the intervention price. When the intervention price is reduced, price variability increases. Support via direct AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 187

payments is more or less risk-reducing, depending on how the payment is determined. If direct payments are counter-cyclical with respect to market price movements, then they can play a significant risk-reducing role. When direct payments are paid at fixed rates based on historic references, they have no risk-reducing potential and an insurance effect will be absent from producers’ behaviour. For example, the calculations reported in OECD (2004b) show that for coarse grains in the United States and wheat in the EU, despite much lower market price support in the United States than in the EU, the net risk reduction in total revenue from all the United States support measures for these crops over the period was not negligible (37%, compared with 71% for the EU). Of particular interest here, these calculations indicate that the move from market price support for wheat to direct payments was risk-increasing for EU wheat producers, since the EU’s area payments offered only about 13% as much protection from world market price movements as did market price support. An early study of producers’ reactions to changes in revenue risk due to policy changes is that by Meilke and Weersink (1990). The authors developed a model to separate the effects of the risk-reducing component from the income-support component of the income stabilisation programmes operating in western and eastern Canada during the 1980s. If these policies had been removed in the late 1980s, the total area of grain and oilseed crops would have fallen by 5% in both parts of the country. In the east (where each crop price was stabilised individually), 10% of this fall would have been due to the increase in risk, whereas in the west (where the return to a basket of high-value crops was the target for stabilisation) the contribution of increased risk to producers’ reactions would have been negligible. Analysis of the risk-related effects of direct payments began primarily as an investigation of the circumstances under which direct payments are totally production-neutral. Therefore, this approach usually focuses on the effects of direct payments on production, rather than on input use or area (which might serve as environmental indicators). However, this framework can also generate some hypotheses about how the shift from price support to direct payments will affect input intensity and cultivated area when producers are risk averse. Table 5.1 summarises the expected ceteris paribus effects of shifting support from price to direct payments. It distinguishes three kinds of direct payment: decoupled payments at fixed rates, partially coupled payments at fixed rates (i.e. rates that are independent of market price movements) and decoupled payments whose rates are determined so as to be negatively correlated with market prices. It is assumed that when support shifts from

188 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

prices to fixed-rate payments, the variability of market prices increases, as does the net variability of total revenue. In the case of decoupled payments at counter-cyclical rates, it is assumed in the table that the payments are calibrated so as to fully offset any additional price risk, and therefore total revenue risk remains constant. This is a special case – the additional price risk could be more or less than fully offset by the counter-cyclical direct payments. If the direct payments decrease the net variability of revenue, an increase in input use would be encouraged, thereby partially offsetting the effect of the price reduction. Table 5.1 does not include a wealth effect, since all empirical estimates suggest that it is small. The changes summarised in the table show that replacing market price support with decoupled payments at fixed rates is expected to decrease input use and cropped area, thus producing potentially beneficial environmental effects. When payments are linked to current area, there would be an incentive to maintain or increase cropped area whilst reducing input use. When decoupled counter-cyclical payments are used, and in the special case that they are calibrated so that revenue risk remains unchanged, there would be no insurance effect to reinforce the static price effect. Table 5.1. Expected effects of shifting support from price to direct payments Effects on

Inputs

Total cropped area

p

p

p

p

p

=/n

p

=/p

p

p

=

=

Decoupled payments at fixed rates

Pricep + paymentn (static effect) Net riskn Payments linked to current area at fixed rates

Pricep + paymentn (static effect) Net riskn Decoupled payments at counter-cyclical rates

Pricep + paymentn (static effect) Net risk=

Available empirical studies of the shift from price support to direct payments relate mainly to the 1992 CAP reform in the EU. The first wave of studies was based on models estimated or calibrated on pre-reform data, and the ceteris paribus effects of the reform were simulated on the assumption that producers maximise profit subject to the new policy incentives. Oude Lansink and Peerlings (1996) simulated the 1992 CAP changes in the arable sector in the Netherlands. They looked at the effects of a 33% price fall, AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 189

plus area payments on 12 different farm types differentiated by soil quality and size, with and without set-aside. For all farm types, and with set-aside in operation, they found that the support shift led to a reduction in cropped area, N-fertiliser and pesticides, with average reductions of -10%, -6.7% and -2.8%, respectively. When set-aside was not applied, reductions were smaller for larger farms.7 The greatest percentage decreases in the use of fertiliser (-8%) and pesticides (-5%), and in area growing cereals and oilseeds (from -10% to -57%, depending on farm size) occurred on “lowproductivity” farms, growing more than 50% cereals, all of which are situated on more fragile land. Therefore, the environmental benefit of input reductions would be relatively greater on these farms. Moreover, since the land is less suitable for high-value, input-intensive crops, like sugar beet and potatoes, it is likely that the land released from cereals was allocated to less intensive forms of production. Guyomard, et al. (1996) analysed the effects of the 1992 reform on individual crops over the first 4 post-reform years in France, using a simulation model based on pre-reform data. For all eligible crops, there were reductions in both area and yield. Although the authors did not report the effects on variable input use, the area and yield reductions together suggest significant reductions in aggregate variable inputs. Whether or not, in reality, the policy changes brought about the expected net environmental benefits depends on a number of factors. An important consideration is how the released cropland was used. Brouwer and van Berkum (1996) reported that in France, where the price falls for eligible crops were not fully compensated by the direct payments (as was the case in the United Kingdom) producers responded to the policy change by diversifying into non-eligible crops such as fruit and vegetables, which use higher applications of pesticides and fertiliser. Moreover, the ceteris paribus effects predicted in these ex ante simulations may not have been realised because producer behaviour is more complex than just expected profit maximisation, or because unforeseen developments are overlooked in the model. Indeed, Rainelli and Vermersch (1997) found that in the early years after the 1992 CAP reform, yields and nitrogen applications per hectare for French arable farms remained at their pre-reform levels. In addition, the level of aggregation of simulation models can be too high to capture relevant effects and may preclude the signalling of effects concentrated in particular locations or on particular inputs. Blogowski and Pingault (2002) found ex post that French arable producers adjusted individual variable inputs differently following the 1992 CAP reform: although fertiliser expenses per hectare were lower in 1999, relative to 1990, phytosanitary and seed expenses had risen so that, allowing for price changes over the period, the combined load of environment-compromising variable inputs per cropped hectare for these farms had slightly increased.

190 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Oude Lansink and Peerlings (1996) and Guyomard, et al. (1996) each used a production framework estimated using pre-reform data, and assumed risk neutrality. OECD (2002) reports a study based on post-reform data from a panel of specialist cereal farms in Italy that allowed for risk effects. For three of the four individual crops modelled, strong insurance effects on production were found, and the total effect (price effect + payment, insurance and wealth effects) was negative. Once again, effects on input use were not reported. However, the estimated elasticities imply a strong negative total impact on the use of seeds and chemicals, and an increase in total area planted. These results can be interpreted as indicating more extensive cereal production on an area expanded at the expense of other (non-cereal) uses. The evidence to date suggests that the switch from price support to direct payments is potentially favourable for the environment in that the static price and payment effects can involve lower input use, or at least may halt rising secular trends in input use. This switch may also lead to a reduction in area devoted to production of the supported commodities, if the full price fall materialises (although this will bring environmental benefits only if the land released from production is used in less environmentally challenging or more environmentally enhancing ways than previously). The net risk-related effects of direct payments reinforce the static effects of the reformed CAP. It is, however, possible, that should counter-cyclical payments over-compensate for additional price risk, the insurance effect would work against the static effects and input reductions would be smaller than expected with risk neutrality.

5.3.

Cross compliance 5.3.1.

Background

In North American terminology, cross compliance related to environmental or conservation objectives is generally termed “conservation compliance” or “resource compliance”. In Europe and elsewhere, the term “environmental cross compliance” is often used when the conditions placed on eligibility relate to environmental objectives. The new CAP legislation (Reg. 1782/2003) sets cross-compliance conditions for the receipt of the SFP and other direct payments (Box 3.1). These conditions address several objectives, of which the environment is only one. Hence, the more general term “cross compliance” is used in that legislation. In this chapter, the focus is primarily on environmental cross compliance. However, much of the discussion is relevant to cross compliance with a wider focus, and would hold for objectives as diverse as animal welfare, public access to farmland or biosecurity measures. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 191

Therefore, unless it is important to distinguish the specific objectives being addressed, the term “cross compliance” rather than “environmental cross compliance” will be used from here on. The term “cross-compliance objectives” will be used to denote whatever objectives are being addressed by the cross compliance, and “cross-compliance conditions” or “crosscompliance requirements” will mean the conditions that have to be met in order to satisfy the cross-compliance provision. Three different forms of cross-compliance provision have been distinguished in the literature (Batie and Sappington, 1986; Baldock and Mitchell, 1995):

x

Red ticket approach: Eligibility for agricultural support payments is determined according to criteria not related to the cross-compliance objectives. Receipt of payments is conditional on the cross-compliance requirements being met. Farmers who do not comply with cross-compliance requirements face partial or complete withdrawal of agricultural support.

x

Green ticket approach: Eligibility for agricultural support and receipt of support are determined according to criteria not related to the cross-compliance objectives. Farmers who meet the cross-compliance conditions receive an extra payment.

x

Orange ticket approach: Eligibility for, and receipt of, support payments depends on meeting the cross-compliance conditions. Eligibility for agricultural support is dependent on farmer enrolment in an otherwise voluntary agri-environmental scheme.

These three approaches could be labelled “staying in”, “topping up” and “opting in” respectively. The conservation-compliance provision of the US 1985 FSA can be described as red ticket cross compliance for those producers with HEL. The 2003 CAP cross-compliance provision is also a red ticket approach, applying to producers who are eligible for the SFP and other payments. Many authors are of the opinion that the green ticket approach does not qualify as cross compliance, and is better described in terms of two separate programmes – one that delivers standard income support payments, and another that offers the option of taking part in a separate, additional, voluntary agri-environmental programme. In practice, voluntary agri-environmental measures that deliver additional “topping up” payments are not considered to be cross-compliance measures, or to be necessarily associated with cross-compliance measures.8 This classification is used in the following discussion of cross compliance and other types of measures.

192 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

The term “green payment” has been used to refer to payments made in line with the green ticket approach (i.e. no cross-compliance element). However, this terminology seems to lack precision and will not be used in this chapter.9 The generic term “agri-environmental payment” will be used to cover payments made to farmers to stimulate or reward a specific environment-enhancing practice or action.

5.3.2.

Advantages and disadvantages of red ticket environmental cross compliance

The main potential advantages and disadvantages of red ticket cross compliance can be summarised as follows:

Potential benefits: x

better harmonisation of agricultural and environmental policies

x

increased conformity with existing legislation and codes of practice, in situations where codes of practice form part of the cross-compliance conditions

x

involvement of intensive producers who would otherwise not enrol on a voluntary basis10

x

heightened awareness of farmers of the consequences of their actions on the environment, in particular if cross compliance is made legally binding

x

increased public acceptance of income support to farmers

x

progress towards application of the Polluter-Pays-Principle in agriculture

x

lower producer payments for certain environmental improvements than occurs under voluntary schemes11

x

lower administrative costs than voluntary schemes12

x

increased awareness and uptake of more demanding voluntary agrienvironmental programmes13

x

in cases of non-compliance, the option of withholding payments, rather than extracting a fine, may prove to be administratively less burdensome.

Potential disadvantages x

limited environmental impacts, if all producers are not eligible for support payments

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 193

x

imbalance in environmental obligations, if some sectors receive support and others do not

x

a weakening of the cross-compliance incentive if support payments are reduced

x

in particular, if payments are counter-cyclical, their influence will be weakest when economic incentives for environmentally damaging intensive production are strongest14

x

where compliance conditions are not part of statutory requirements, financial penalties for non-compliance may provide a weaker incentive than the threat of legal sanctions; but

x

where compliance conditions are part of standard statutory requirements, legal redress will be more appropriate for enforcing compliance than agricultural policy mechanisms 15

x

if cross-compliance measures are closely targeted, this will entail a heavy administrative framework and high monitoring costs; but

x

if cross-compliance measures are more general and less targeted, the outcome for the environment will be uncertain, although administrative and monitoring costs are likely to be lower

x

in financial terms, there is no perception of compensation for producing environmental benefits,16 unless farmers are able to perceive a link between compliance and receipt of payments

x

imposing homogeneous requirements across all farmers fails to address the fact that individual farmers have different compliance costs17

x

if cross-compiance conditions do take heterogeneous compliance costs into account, administrative and monitoring costs will be higher.18

These last two points highlight the trade-off between allowing for farmer- and site-specificity, and keeping programme costs low. This tradeoff characterises both cross-compliance provisions and specific agrienvironmental measures and will be discussed in Section 5.4. Not all of the potential benefits and disadvantages listed above will affect all cross-compliance schemes, and some could be modified by appropriate programme design. Others, however, are inherent in what is essentially a second-best situation. In an ideal world, agricultural and environmental objectives would be maximised simultaneously, so that the trade-offs between them could be fully incorporated in policy solutions

194 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

(Christensen and Rygnestad, 2000). Likewise, in an ideal world, features of the interaction between agriculture and the environment (such as multiple contributors to a common environmental problem; the difficulty of observing and measuring the effects of individuals’ actions; heterogeneity of underlying conditions and so forth) would not generate transaction costs. In such a world there would be no justification for cross-compliance measures. In reality, cross-compliance provisions are usually seen as achieving a degree of harmonisation between two different objectives, subject to extensive constraints imposed by existing policies and underlying political economy considerations. Thus, cross compliance is rather a second-best policy approach and may be less cost-effective than agri-environmental measures targeted to specific agri-environmental objectives. Section 5.4 examines in more detail what it has to offer and its inherent limitations as a measure for reducing environmental damage.

5.3.3.

Design of cross-compliance provisions

In order to highlight some of the issues involved in the design of cross compliance schemes, it is instructive to compare the approaches of the EU (as embodied in Reg. 1782/2003), Switzerland and the United States (the three provisions inaugurated in the 1985 FSA). Table 5.2 summarises their key characteristics. There are various strategic differences among the United States, Switzerland and EU approaches. The US provisions are more sharply targeted, and apply to a narrower range of objectives. In principle they also make greater allowance for producer heterogeneity than does the EU approach. However, to what extent this is true is not easy to determine, as the EU regulation on cross compliance leaves many details of design and implementation up to individual member states. For example, under the Nitrates Directive (one of the 19 directives included in the compliance conditions), individual member states are responsible for designing both the action plans targeting the use of inorganic and organic fertilisers within their Nitrate Vulnerable Zones, and also the codes of good nutrient practice outside these zones in such a way that the common standards set out in the Directive can be met under local conditions. The requirement that land corresponding to SFP entitlements must be maintained in good agricultural and environmental condition also leaves member states some discretion in prioritising potential environmental threats to farmland at national or local level, within the various environmental characteristics for which standards must be set, as laid down by Annex 4 of the Regulation. Whether member states will go as far as requiring individual farm plans or contracts remains to be seen. In Switzerland, farmers have to comply with a set of minimum environmental standards in order to be eligible for any direct payment. These requirements are identical on all the territory. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 195

Table 5.2. Comparison of EU, Switzerland and US cross-compliance provisions EU

Switzerland

United States

Scope

Environmental (agrochemical pollution, groundwater protection, wildlife, habitat, soil protection by sludge in case of use of sewage), food safety, animal welfare and health objectives

Environmental and biodiversity (conservation of natural resources, animal welfare)

Soil erosion, wetland preservation (habitat)

Specificity

Horizontal, most CAP production activities

All direct payments are subject to cross-compliance conditions

Cropping activities

“Ticket”

Red

Red

Red

Coverage

Whole surface of farmland with SFP entitlement

All utilised agricultural land

HEL and wetlands for farmers participating in support programmes

Relationship with property rights

Cross-compliance conditions are in line with statutory requirements for farmers, thus coinciding with existing property rights boundary

Cross-compliance conditions go beyond what is required from all farmers by common statue

Cross-compliance conditions go beyond what is required from all farmers by common statute

Flexibility

Limited – 19 EU-wide directives (with scope for national options) + national/regional variations in standards for land maintenance

National programme

Individual contracts for farmers under conservation compliance and sodbuster

Penalties

Reduction in payments proportional to the severity, extent, permanence and repetition of the infringement

Reduction in payments proportionally to the extent of the infringement and the damage caused

Non-compliance can result in loss of many programme benefits, on all farms operated

196 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

The comparison made in Table 5.2 suggests two other possible strategic differences among the EU, Switzerland and the US approaches. First, in the US there seems to be a greater emphasis on deterring farmers from bringing environmentally sensitive land into production. In the EU and Switzerland, rural areas are more intensively settled and farmed than in the US, and in general soil conditions and soil maintenance are also quite different in the EU from the US, following several millennia of agricultural production. These differences are reflected in the cross-compliance conditions attached to the SFP, which aim at re-orienting working farms towards more environmentally respectful farming practices, regardless of land use, rather than keeping land out of crop production. Having said this, the US conservation-compliance provision, which involves conservation planning for HEL that remains in production, is a reminder that if this difference in approach is real rather than illusory, it is a matter of degree only. A second difference concerns the relationship between the compliance conditions and what is already expected from farmers because of existing legislation or agreed codes of practice. In the EU and Switzerland, the compliance conditions are more or less aligned with the current boundary that sets limits to farmers’ property rights.19 By contrast, the compliance conditions for the US provisions impose constraints over and above standard statutory requirements. This suggests that, in theory, opting out of programmes with cross compliance attached might be a more realistic proposition for a US farmer than for an EU farmer.20 At the same time, the financial consequences of noncompliance would appear to be more severe for farmers in the United States. The concept of a code of good farming practice is an attractive one that is being pursued by various countries. For a code of practice to have a significant effect on the behaviour of farmers as a whole, leverage of some kind is necessary. This leverage could be simply by exhortation, or it could be financial (compliance being a condition for receiving a support payment), or legal (compliance being a statutory obligation enforceable for all farmers, independent of any agricultural policy instrument). An important question is what kind of leverage is most appropriate for promoting codes of good agricultural practice, and whether there is a case for including certain environmentally beneficial management practices among the set of legal requirements (e.g. farm safety, hygiene, etc.) that are already imposed on farmers. As an illustration, it might be considered whether, given the seriousness and extensiveness of nitrate pollution problems, soil N testing should become a legal requirement for all arable farmers and grazing livestock producers with stocking rates above a minimum level, with penalties related to increases in residual nitrogen with respect to a 5-year site average.21 Fuglie and Bosch (1995) illustrated the potential gains to soil N testing for a group of maize-producing farms in Nebraska, whilst also highlighting the AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 197

issue of heterogeneity across farms.22 Farm heterogeneity means that some farms may suffer a loss in profit from particular environmentally friendly farm management practices, and without securing an increase in net social benefit. This raises the question of which – if any – management practices should be mandatory for all farmers, whether absolutely by law, or conditionally, as part of cross compliance. For decisions that do not involve potential negative externalities, the choice of management practices can generally be left to the farmer, although in some cases it will be desirable to promote more environment qualityenhancing management practices even if a current practice is environmentally neutral. However, in situations where there are likely to be harmful externalities, it may be efficient to require by law that farmers observe precautionary management practices, even if it cannot be demonstrated that it will turn out in every case to be of private benefit to the farmer. A positive consequence would be that the management practice concerned would become obligatory for all farmers, regardless of programme eligibility or participation (including, in the US, crop producers not claiming support under any programme or, in the EU, vegetable and fruit growers. Both these groups are by-passed by current cross-compliance provisions). A likely negative consequence is the lower efficiency and costeffectiveness of one-size-fits-all policy measures when applied to heterogeneous producers. The question of targeting producers and sites under cross compliance will be returned to in Section 5.4. The question arises as to how far measures that promote environmental enhancement in a proactive way can be accommodated within a cross-compliance system. The integrated farming system (IFS) approach described by Morris and Winter (1999) is an example of a set of sustainable farming practices based on a strong post-productivist philosophy. IFS is characterised as an holistic approach to sustainable farming, and a third way coming between conventional productivist farming and organic agriculture, the latter requiring a special kind of commitment that is too rigorous for most farmers. The main principles of IFS are crop rotation (to promote soil structure and fertility and reduce agrochemical use); a minimum of four crops in the rotation; minimum soil cultivation (to reduce erosion); disease-resistant cultivars (to reduce pesticides, etc.); modification of sowing times (to reduce risk of pest and disease outbreaks); targeted application of nutrients; no prophylactic spraying; management of field margins to create habitats; and so forth. Many of these practices have already been incorporated into conservationinspired farm plans. What is new in IFS is the whole-farm approach and the idea that conservation practices can fit well in the high-tech environment of the modern farm, exploiting modern technology to implement them. Röling and Jiggins (1998) stressed how IFS makes intensive use of scientific knowledge,

198 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

learning from observation and technology (as a tool for managing water, soil, pests, etc., without using intensive agrochemicals). Such an approach signposts a pathway from modern input-intensive agriculture that exploits the technological gains of the last century, rather than retreating from modern technology. However, it probably goes beyond what could currently be made mandatory for all farmers in any farming population or what could be imposed by cross compliance, and would have to be fostered by voluntary remunerated schemes.

5.3.4.

Various options for linking income transfers and environmental objectives

The following paragraphs present a schematic overview of various policy options involving transfers of income to farmers, associated with different approaches to agri-environmental constraints and incentives. Four “options” are distinguished, which could be viewed as stages in a chronological sequence of policy developments or as a portfolio of policy components, some of which might be adopted simultaneously. In comparing the different options, it is assumed, for simplicity, that the total income transfer received by farmers remains the same, but it is delivered by government in the form of direct payments that are decoupled from all current production decisions rather than as transfers from consumers via market intervention. Note that Options 2, 3 and 4 correspond broadly to the red, green and orange ticket cross-compliance approaches, respectively. Payments under Option 4, however, will be seen as subject to cross compliance only if they are still primarily considered to be “support” payments. If they are perceived simply as payments to farmers in remuneration for environmental goods and services, then the idea of cross compliance becomes irrelevant. Option 1 describes a situation where market price support has been replaced by payments that are decoupled from current production (for example, payments based on historic production levels), but no conditionality is associated with these payments. This shift results in an increase in social welfare due to the reduction in distortions to food and other markets (as summarised by the deadweight loss of these policies), which is offset to an unknown extent by the deadweight loss of raising tax revenue to fund the direct payments. The reduction in market price support may itself have positive environmental effects even before any additional benefits have been secured by the cross-compliance provision on direct payments (see Section 5.2). There may also be positive distributional consequences: since market price support acts as a regressive tax on food consumption, the welfare gains due to lower food prices will favour poorer households. With Option 2, environmental cross-compliance conditions are imposed on the receipt of direct payments. Even when these conditions simply reflect preexisting statutory requirements, the cross-compliance link may provide an AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 199

additional incentive to producers, and income support is explicitly linked to good environmental practice. This is a desirable and overt step towards the harmonisation of agricultural and environmental policies. When cross compliance involves more than statutory obligations, then additional environmental benefits can be secured by imposing obligations that presumably were considered politically unacceptable without some explicit recognition or compensation for farmers. This may facilitate the political acceptance of adjustments in the boundary that defines the property rights of landowners. The shift in the terms of conditions for receiving payments need not stop at a situation characterised by direct payments plus cross compliance. Another option describes a hybrid situation where income transfers are partly provided in the form of direct payments based on historic data and partly via targeted agri-environmental measures,23 where the payment depends on current environmental practices or outcomes. A fourth option represents a situation where the criteria for receiving government payments depend solely on environmental services performed, or benefits obtained. These options can be viewed as forming a policy continuum along which environmental objectives become increasingly dominant at the expense of other objectives for transferring income to farmers. A move along this continuum involves closer targeting of environmental outcomes and greater environmental cost effectiveness, but entails a possible loss of efficiency with respect to other social and political aims of farm support. If the payments received for providing environmental benefits under Options 3 or 4 are more equally distributed among farmers than direct payments under Options 1 and 2, then there will be a redistribution of income among producers. This could occur if direct payments are distributed on the pattern of past production levels or resource ownership, whilst smaller and less commercial farms would have disproportionately greater scope for providing environmental benefits. Furthermore, if some larger producers do not opt into agri-environmental schemes or if total payments per farm for environmental schemes are capped, these factors will also promote a more equal distribution of transferred income among farmers with Options 3 and 4 relative to Options 1 and 2. It should be stressed that a pure Option 4 strategy is unlikely to be adopted in practice, even in the long term. As long as farm income remains a political issue, then who receives income transfers and how they are distributed among recipients will remain important policy considerations. Since not all producers are well placed to deliver environmental benefits, governments will wish to retain some direct payments that are determined according to non-environmental criteria. Moreover, the administrative cost of making the desired total income transfer to farmers via voluntary targeted agri-environmental schemes may render a pure Option 4 approach less efficient than other alternatives. These stylised options are summarised in Table 5.3.

200 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Note that US policies for the arable sector can be described as a mixture of Options 2 and 3. The EU’s medium-term strategy of shifting CAP payments out of the first pillar towards second pillar measures roughly compares with a move from Option 1 towards Option 3. With the 1992 CAP and Agenda 2000 CAP reforms, the EU began moving away from Option 1 towards Option 3, but with Regulation 1782/2003, it has now added the main Option 2 instrument – cross compliance linked to direct income support – into its policy framework.24 Although these stylised options are presented as alternatives, in practice each country can be expected to seek its own optimal balance between direct payments for income support, crosscompliance mechanisms to enforce “minimum environmental standards” and voluntary agri-environmental payments. Table 5.3. Various options involving direct payments and agri-environmental measures Policy change Market price support (MPS) converted to direct payments

Result Removal of distortions due to MPS

Comment Distortions due to MPS are replaced by distortions due to raising additional tax revenue

Option 2 (red ticket)

Cross compliance imposed on receipt of direct payments

Additional environmental benefits if cross compliance requires more than statutory obligations

Better integration of agricultural and environmental policies

Option 3 (green ticket)

Direct payment entitlements reduced and voluntary agrienvironmental schemes increased

Further increase in provision of environmental benefits

Distributional changes among farmers

Option 4 (orange ticket)

Direct payment entitlements independent of environmental criteria disappear and all payments are for voluntary environmental schemes

Financial limit for securing environmental improvements is reached

Transaction costs of delivering payments to farmers may be very high Some producers may not receive any payments

Option 1

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 201

5.4.

Efficiency and cost effectiveness of cross compliance and alternatives This section examines the efficiency and cost effectiveness of five stylised programmes in which payments are made by government to farmers and which involve an environmental objective. The main results are summarised in Table 5.4.

5.4.1.

Efficiency and cost effectiveness of various programmes

A policy mechanism achieves maximum environmental efficiency when it produces a situation in which the marginal social benefit of environmental improvement is equal to its marginal social cost. For simplicity, it is assumed that the marginal utility of income for payment recipients and taxpayers is identical, and there is no deadweight loss of taxation, so that a direct payment from taxpayers to farmers can be netted out as a pure transfer without net welfare implications. Then, an efficient programme for obtaining environmental benefits from farmers by using the leverage of direct payments is one that secures environmental benefits up to the level where society’s valuation of the marginal unit of environmental benefit is equal to the cost (in terms of extra costs or profit foregone) to the farmer who produced it. At this point, net environmental benefit is maximised. A programme can be described as more or less efficient according to how close it comes to achieving maximum environmental efficiency. The concept of cost effectiveness relates net benefits obtained to the programme cost. The most cost-effective policy mechanism maximises the net benefits that can be obtained for that cost. Cost effectiveness can be a more useful concept than efficiency when budget funds devoted to a specific objective are subject to a fixed limit.

202 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 5.4. Summary of cost effectiveness of five stylised programmes Trade-off between objectives (1) Direct income payments with compulsory cross compliance (2) Direct income payments with voluntary cross compliance

(3) Direct income payments with voluntary cross compliance, + crude environmental targeting

Income support objective dominates Income support still dominates

Comparison

Full cost

Nd + A(N)

Cost effectiveness (incremental cost only) 4

¦ (BC i CC i )

i 1

Compared with (1): x opters-out (N1 + N2)d achieve higher + A(N1 + N2) income x gross environmental = D + A(N1 + N2) benefit n x total costs p Income Compared with support to (2): some would-be x some Nsd + A(Ns) participants “volunteers” + C(s) traded-off excluded against some x net environmental environmental targeting benefit n (likely) x total costs p (likely)

Cost effectiveness (full cost) 4

¦ (BC i CC i )

i 1

A(N)

Nd A(N)

¦ (BC i CC i )

¦ (BC i CC i )

2

i 1

A(N1 N 2 )

2

i 1

D A(N1 N 2 )

Examples SFP + cross compliance (EU) and conservation compliance (US) could be in either category, depending on whether viewed as voluntary or compulsory

BC s CC s N s d A(N s ) C(s)

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 203

(Table 5.4 continued) Trade-off between objectives

Comparison

(4) Agri-environmental programme with bidding for contracts

Environmental Compared with objectives (2): dominate; x coverage n support reduced x environmental to the incentive benefit n component of x total the contract bid payments: no change (by assumption

(5) Agri-environmental programme with bidding for contracts and environmental targeting

Environmental Compared with objectives (4): dominate; x coverage np support reduced x environmental to the incentive benefit n component x total of the contract payments: no bid change (by assumption)

Full cost

Cost effectiveness (incremental cost only)

D + Ac(Nc)

Cost effectiveness (full cost)

Jc

¦ (bc j cc j )H j

Examples

CRP until 1995

j 1

D A c (N c )

D + Ae(Ne)

204 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

CRP since 1995

Je

¦ (bc j cc j )H j

j 1

D A e (N e )

Which benefits are to be related to cost depends on the objective of the policy. As already emphasised, direct income support plus environmental cross compliance has two objectives. To examine the cost effectiveness of this measure with respect to both objectives simultaneously is not a straightforward task. Therefore, in what follows, the income support objective is ignored, and focus placed on the cost effectiveness of the payments in securing net environmental benefits – that is, environmental improvements net of profit foregone by farmers. Profit foregone, or compliance cost, is the reduction in marketed net value added of farm production. A programme’s “environmental cost effectiveness” is therefore used to mean the extra environmental benefit net of profit foregone, per dollar of programme cost. It is not a simple task to measure programme cost with accuracy. For a programme consisting of direct income support with a cross-compliance provision, where the income support objective dominates and the support would have been given even without the cross-compliance provision, then it may be more appropriate to relate the net environmental benefit of the cross-compliance provision to the incremental cost of administering the cross compliance and not to include the direct payments themselves as part of the cost of achieving the environmental objective. By contrast, where the environmental objective is the raison d’être of the programme, the appropriate cost would certainly include the payments made to farmers. The five stylised programmes considered below include examples of programmes corresponding to each of these situations, as well as a programme where the income objective is partly traded off against the environmental objective. First, consider a situation where all farmers have the right to a uniform fixed payment (equal to (d) per hectare), which is determined according to non-environmental criteria. However, in order to receive the payment, the farmer must comply with a standard package of environmental norms and constraints applicable on the whole farm. For simplicity, it is assumed that compliance with these conditions translates in a known, deterministic way into environmental benefits. It is assumed that all hectares on a given farm are homogeneous, that farmers treat the cross-compliance provision as binding, and that no deliberate evasion occurs. Producers can be separated into two groups, according to whether or not their compliance costs per hectare (cc) are covered by the payment. They can also be divided according to whether the social value of the benefit per hectare that they generate by complying with the package (bc) exceeds their compliance cost per hectare.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 205

Thus, producers fall into four groups, as shown in Table 5.5. The groups are identified by the index i=1,2,3,4. The classification in Table 5.5 is used below when analysing various policy options. Table 5.5. Classification of farmers according to compliance benefits and costs Positive net environmental bc – cc > 0

bc – cc < 0

D – cc > 0

Group 1

Group 2

D – cc < 0

Group 3

Group 4

Payment covers compliance cost?

Compulsory cross compliance With mandatory cross compliance and a payment that is uniform for all producers, all groups i=1,2,3,4 participate in the scheme. Groups 3 and 4 suffer an income loss within the programme, as the direct payment does not cover their compliance costs.251 Moreover, the total net social benefit of the scheme is below the maximum efficiency level, since the participation of Groups 2 and 4 actually reduces this total. The total programme costs are Nd, where N is the total number of hectares operated by the four groups of farmers. Although producers in Group 3 contribute positively to the net social benefit of the provision, the fact that they are not fully compensated for this could create a source of grievance and a motive for deliberate evasion on compliance. As well as not being compensated for their compliance costs, the participation of producers in Group 4 actually reduces the total net social benefit of the programme (although these producers do contribute towards the gross environmental benefit of the provision). The existence of farmers in Groups 3 and 4 lies behind the argument advanced by some researchers in Europe that the compliance conditions for mandatory cross-compliance provisions should consist mainly of requirements already imposed on farmers by other regulations, and that any additional conditions should involve low costs for the farmer (Baldock and Mitchell, 1995; EC, 1998). When net benefits are related to full programme costs, the cost 4

effectiveness of this programme is ¦ (BC i CC i )/[Nd A(N)] , where i 1

BCi and CCi are the total benefits and costs of compliance of the i-th group and A(N) is the administrative cost of implementing the cross-compliance provision for N hectares. When environmental cost effectiveness is calculated on an incremental basis (i.e. considering only the costs of

206 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

implementing the cross-compliance provision), the cost effectiveness of the 4

cross-compliance provision is ¦ (BC i CC i )/ A(N) , which will be much i 1

greater, relatively.

Voluntary cross compliance If the cross-compliance provision is voluntary, producers in Groups 1 and 2 will sign up for the payments, whilst those in Groups 3 and 4 will not. As long as the cross-compliance provision is attached to the payment, these last two groups are better off if they opt out of the income support programme. It is not possible theoretically to say whether the net social benefit will be higher or lower than with mandatory compliance, since with voluntary participation the abstainers include both producers whose compliance would have increased the net social benefit, and some whose participation would have reduced it. Moreover, it is not possible to say a priori whether the cost effectiveness, on either a full-cost or incremental-cost basis, is lower or higher than in the compulsory case, although in both cases the budget cost of the programme is lower, since only a sub-set of producers will volunteer to participate.262 In the case where there is a strong negative correlation between the benefits and costs of compliance across producers, those whose compliance would generate large environmental benefits have a small increase in costs, and vice versa. In this situation, many more producers would fall in Groups 1 and 4 than in Groups 2 and 3, and therefore with voluntary participation and an appropriate level of (d), the net social gain would be higher than with compulsory participation.273 The joint distribution of compliance benefits and costs is an empirical question, and depends on the nature of the compliance conditions as well as on the production and site characteristics of the farms involved. For many types of environmental damage, Groups 2 and 3 could well contain relatively large proportions of farmers and hectares. For example, high-input/high-output cereal farms, responding to strong market prices, may result in considerable environmental damage. However, they would also face significant compliance costs in terms of income foregone if compliance involved a large yield reduction. These producers would fall under Group 3. Organic growers would have relatively little scope for generating additional environmental benefits by, for example, reducing agrochemical use, and their compliance costs could well be very small.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 207

Thus, they would probably come under Group 2 and would opt in, although their net contribution would be negative.

Voluntary cross compliance with restrictions based on environmental targeting As demonstrated above, a voluntary provision where farmers receive a fixed payment as long as they comply with a standard package of environmental conditions would attract all producers for whom (d) – (cc) > 0 (i.e. producers in Groups 1 and 2). Participation of producers in Group 2 would reduce both the efficiency of the scheme and its cost effectiveness, relative to a scheme that contains producers from only Group 1. Various screening mechanisms could be used to exclude Group 2 producers from the scheme. Suppose, for example, that producers with compliance costs below the level of the payment, but with low potential to generate environmental improvements, can be identified by farm type (e.g. organic farmers or farmers already adopting good conservation practices); by soil type (e.g. on soil with low erosion potential); or by region (e.g. in areas without nitrate contamination problems). If this can be done, then the payment and cross-compliance provision can be offered to Group 1 and Group 3 producers only by specifically excluding probable Group 2 and Group 4 producers according to appropriate criteria. Suppose first, that the selection mechanism can be set so as coincide with the boundary between Groups 1 and 2 and Groups 3 and 4 (although in reality this could be difficult to achieve). On this assumption, both the efficiency and the effectiveness per dollar spent are greater for this restricted scheme compared with the previous unrestricted scheme.284 Unlike the unrestricted voluntary scheme, where those groups that remained outside the income support programme were better off, in this case, producers in Group 2 would prefer to be inside the payment scheme. This raises several issues that have received attention in the literature. Claassen, et al. (2001) distinguished schemes that reward producers for improved performance from those that reward good performance. In the example used here, Group 2 producers, who by definition have relatively low compliance costs (since d > cc), may be in this position because they are already performing at levels satisfying, or nearly satisfying, the compliance conditions. For these producers, the necessary additional adjustments would involve small, or no, compliance costs. This could be because they have already made investments to improve their environmental performance (e.g. organic farmers or environmentally responsible farmers), or because they farm in regions that can support modern farming methods without

208 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

incurring environmental stress. Whatever the reason, their environmental performance is already good, given the characteristics of their farm site.295 A major issue here is one of fairness. Excluding these farmers from the payment scheme because they are already performing well penalises them for past decisions or exogenous factors. A second issue relates to the selection mechanism itself. If broad criteria (as illustrated above) are used, this will in practice result in some farms being wrongly classified, and hence efficiency losses will be added to the equity problem. Incorrect classification will reduce the net environmental benefit. Alternatively, a more accurate separation could be attempted by estimating the expected benefits of each producer’s compliance.306 This could involve establishing a reference level, and estimating the expected extra benefit above that level, which would require extra information and increase the programme cost. A practical question is whether the lower net benefits due to erroneous classification or the extra information cost will be sufficiently great to cancel out any increase in cost effectiveness from excluding Group 2 producers. If so, the bottom-line cost effectiveness of the restricted scheme would turn out to be inferior to that of the scheme that includes the “good performers” as well as the “potential improvers”. An amended measure of cost effectiveness for this measure, then, is BC s CC s , where BCs and CCs are the total environmental N s d A(N s ) C(s) benefits and compliance costs of the producers who are admitted by the selection mechanism: Ns is the number of hectares selected; and C(s) is any extra cost of operating the selection mechanism. (BC s CC s ) could be 2

greater or less than

¦ (BC

i

CCi ) , and N s d A(N s ) C(s) could be

i 1

greater or less that (N1 N 2 )d + A(N1 N 2 ) . Thus, any comparison of the cost effectiveness of the unrestricted and restricted voluntary programme is a priori indeterminate. The restricted scheme, which aims to enrol Group 1 producers only, is unattractive to an important group of “potential improvers”, namely those in Group 3. Clearly, a provision that maximises the participation of producers from both Groups 1 and 3, whilst deterring producers in Groups 3 and 4, would improve both environmental efficiency and cost effectiveness. Two possibilities for increasing voluntary participation and/or improving environmental targeting will now be examined. However, it should be pointed out that in these two schemes, the income support objective has become completely subordinate to the environmental objective.317 AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 209

To reiterate, Group 3 producers do not volunteer for the unrestricted voluntary payment scheme described above, because the payment (d) is below their compliance cost per hectare. Moreover, under this scheme Group 1 producers typically receive payment in excess of their compliance cost. If income support is an objective of the payment, then this excess compensation for compliance is not necessarily a defect of the policy. However, from the point of view of the environmental cost effectiveness of allocating a given budget, the voluntary flat-rate payment + CC provision is sub-optimal, because some producers receive a higher payment than is necessary in order to provide a compliance incentive. This is a drawback of all flat-rate compliance provisions offered to producers with heterogeneous compliance costs. Ideally, the compensation rate would be fixed for each producer at the precise level required to secure participation (compliance cost + incentive payment). Generally, however, it will be impossible for the regulator to determine the appropriate payment structure because of asymmetry of information on compliance costs: farmers know their compliance costs, but the regulator does not. The literature has identified two alternative approaches to dealing with information asymmetry in this context. In the “screening” model, the regulator (the uninformed party) makes the first move – for example, the authorities state the payment rate for a particular scheme, which is then progressively adjusted until a target number of participants is willing to enrol – whereas, in the signalling model, the farmer (the informed party) moves first and signals the payment rate that would induce him to comply. Fraser (1995) has shown that the screening model leads to inefficient outcomes, with the farmer extracting informational rent. The following paragraphs therefore, focus on procedures conforming to the signalling model. However, it must be pointed out that the theoretically established superiority of the auction model depends on the assumption that there is no strategic bidding in the auction setting. This does not hold when, over a sequence of auctions, producers fix their bids in successive auctions in the light of previous outcomes. In this case, and taking into account the higher administrative costs of conducting the auction, the efficiency gain of the auction option may well disappear.

Auction of contracts, bids selected according to compliance costs A bidding system, in which producers tender for contracts to provide environmental services, is a means of inducing them to reveal their compliance costs. The following situation may be supposed: the contract is specified in terms of a standard set of services to be provided. Producer j bids the amount (bj) per hectare on (Hj) hectares, thereby signalling that this

210 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

is the payment that it would be required for meeting the contract; the budget limit for total payments is (D), i.e. the same amount that was spent on the voluntary flat-rate scheme described above (i.e. D=(N1 + N2)d). Here, the administrative cost is [Ac(Nc)], where this represents the cost of running the auction and all other implementation costs, and (Nc) is the number of hectares contracted after the auction. Jc

Bids will be accepted up to the point where ¦ b j H j

D , where Jc is

j 1

the last producer to have a bid accepted when producers are ranked in ascending order of bid size (i.e. compliance cost). Because the amount of over-compensation paid to producers with low compliance costs is reduced, the given budget will allow more hectares to be enrolled than would be the case with the flat-rate scheme. This means that the average payment per hectare will be lower than under the flat rate scheme. The number of hectares enrolled (Nc ) is 6jJc Hj. In terms of Table 5.5, this result is equivalent to lowering the horizontal line separating Groups 1 and 2 from Groups 3 and 4, so that some producers from these last two groups now participate in the scheme and thus accept (voluntarily) the compliance conditions. Thus, Nc > N1 + N2. These “new recruits” will have made bids in excess of the flat rate (d), and will be paid at this higher level. Producers in Groups 1 and 2 will receive payments that are lower than d. Gross environmental benefits increase, but whether or not the environmental cost effectiveness improves relative to the voluntary flatrate scheme is not clear, since the new recruits include producers making negative contributions to net social benefit, as well as those making positive contributions.

Auction of contracts, bids selected according to net environmental benefit The environmental cost effectiveness of the voluntary flat-rate scheme can certainly be improved if producers are ranked in terms of the net environmental improvement they can generate, rather than in terms of compliance cost. The situation where the regulator can estimate the size of each producer’s compliance benefit per hectare (bc) is supposed. This information can be used, along with information on compliance costs (as revealed by producers’ bids) in order to rank producers according to their estimated (bc-cc) or, equivalently, by (bc/b). When this is done, the budget limit operates as follows: bids are accepted up to the point where

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 211

Je

¦ b jH j

D , where (Je) is the last producer to have a bid accepted when

j 1

producers are ranked in descending order of compliance benefit relative to compliance cost. The number of hectares enrolled (Ne) is 6jJe Hj, which may be greater or less than (Nc). Administrative costs per hectare are likely to be greater than for the previous scheme, which does not take environmental benefits into account. Both bid selection criteria described above have been used to determine enrolment in the CRP. In the first nine sign-ups, priority was given to producers with the lowest compliance costs (i.e. bid prices). Thereafter, estimated environmental benefits were also taken into account (see Reichelderfer and Boggess, 1988; Claassen, et al., 2001).328 The superior cost effectiveness of the second approach has been documented by these authors, and was also demonstrated formally by Latacz-Lohmann and van der Hamsvoort (1997), who showed for a hypothetical case that, for the same total cost, when bids were selected according to the benefit/cost criterion, environmental cost effectiveness increased by 29% compared with the flat-rate scheme (and the same budget), with a 2% reduction in hectares enrolled and a 43% reduction in over-compensation. Even when only compliance cost (as revealed by the size of the bid) was used as the bid selection criterion, there was an improvement of 16% in environmental cost effectiveness, an increase in hectares enrolled of 11% and a reduction in over-compensation of 17%, holding the budget constant. These results indicate the success of the environmental targeting in bringing more producers into the programme from Group 3 at the expense of those in Group 2 (and possibly Group 4). When the regulator can obtain information on each farmer’s potential environmental compliance benefit (bc) using sources or methods not available to the farmer, a second source of information asymmetry is introduced, this time in favour of the regulator. The question has arisen as to whether the cost effectiveness of a bidding process might be improved if the regulator were to share his information about the size of expected environmental benefits with farmers. In an experimental setting, Cason, et al. (2003) showed that, on the contrary, if bidders (farmers) have information on the expected benefits of their compliance in addition to knowing their costs, they will pitch their bids higher (i.e. less abatement per dollar spent) and the percentage of maximum abatement possible will also be lower.

212 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Standard or customised compliance conditions? Both the compulsory cross-compliance provision attached to the SFP under the CAP, and the CRP contract, which is a voluntary agrienvironmental measure, specify a standard package of conditions that have to be met on all areas, subject to receiving payments. The question arises as to whether further gains in environmental cost effectiveness would be secured if producers could negotiate individual contracts in which incentives could be specifically tailored to those environmental outcomes for which the producers’ compliance cost is particularly low and/or the potential compliance benefit would be particularly high. In theory, individual producers should be able to generate higher environmental benefits for a given compliance cost if their compliance conditions recognise site-specific comparative advantages in achieving environmental improvements. Indeed, individually negotiated contracts are offered under certain environmental schemes in Europe, and the conservation compliance provision of the US 1985 FSA is based on conservation contracts drawn up by each individual farm and approved by the appropriate authority. Despite their superior environmental targeting, however, it is generally assumed that the transaction costs of individually specified contracts will be higher than for programmes operating with standard contracts or uniform sets of compliance conditions. Very little information is available on this question. The study by Falconer and Saunders (2002) is notable because its results contradict this widely held expectation. The authors compared the total transaction costs (incurred by the regulating authority and the farmer) of individually negotiated management agreements within Sites of Special Scientific Interest (SSSIs), with those of standard management agreements under the Wildlife Enhancement Scheme (WES) in the United Kingdom. Compensation payments under the SSSI contracts tend to be for benefits foregone, whereas under the WES they are for positive actions taken. Although the WES contracts had lower negotiating costs, their on-going implementation costs (measured in GBP/ha per year) were much higher than the two different legal forms of SSSI contract studied. As a consequence, the total transaction costs of the standard management agreements (in GBP/ha per year) were considerably higher than for the individually negotiated contracts. These results need to be explained in terms of the design features of the two programmes, in order to learn appropriate lessons. Moreover, the study should be replicated for other similar schemes before drawing any general conclusions. Nevertheless, on a question where information is scarce, these results provide valuable food for thought. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 213

5.4.2.

Participation, monitoring and non-compliance

So far, it has been assumed that producers who participate in a voluntary payment programme with cross compliance both collect the payment and comply with the conditions. However, this is an unrealistic assumption. In the second part of this chapter, figures were quoted on non-compliance rates and payments withheld for several US compliance provisions. Various studies of voluntary agri-environmental programmes in Europe have also reported that, with a 5% monitoring rate of farmer compliance, noncompliance is detected in 1-6% of the cases monitored.339 Therefore, it is worth considering the possibility and consequences of non-compliance (i.e. participation in the scheme, but deliberately evading on the compliance conditions). Allowing for this possibility means recognising that a producer faces three options: opting in and complying, opting in and not complying, and opting out. The simple framework of Latacz-Lohman and van der Hamsvoort (1999) can be adapted to describe these options. There are two conditions for a producer to opt in and comply. First, a producer will decide to opt in as long as cc < d (this is the individual rationality constraint, or the participation constraint). However, for producers who have opted in, the expected penalty involved in not complying is p(d+F), where (p) is the probability of being monitored and (F) is a fine imposed if caught not complying, and where it is assumed for simplicity that if non-compliance is detected the full payment is lost.3410 Therefore, opters-in whose morality does not rule out such a gamble and who are risk–neutral, will find it an interesting option not to comply, providing cc t p(d+F). Latacz-Lohman and van der Hamsvoort added a second constraint in order to guarantee that producers who choose non-compliance would not in fact have preferred opting out; that is, it must be true that d t cc t p(d+F), which implies d t p(d+F). Rearranging this and solving for d yields d t pF/(1–p). As long as this condition holds, then there is an incentive for a risk-neutral opter-in not to comply. As the authors point out, the choice between opting in and opting out depends entirely on the compliance costs and is therefore subject to asymmetric information in favour of the farmer. For risk-neutral opters-in, the choice between complying and not complying depends on (p), (d) and (F), all of which are under the control of the regulator. Thus, for a given d, increasing the probability of detection (i.e. increasing the monitoring rate) and/or increasing the fine sufficiently can invalidate the condition and make non-compliance unattractive. However, increasing the detection rate will add significantly to the monitoring costs of the programme, whereas

214 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

increasing the fine to prohibitive levels will make the programme very unpopular. These results relate to the special case where producers are riskneutral. Ozanne, et al. (2001) analysed the situation of information asymmetry and programme compliance in a more complex framework, allowing for: risk-averse producers; monitoring costs that are independent of – or, alternatively, proportional to – monitoring effort; and social welfare maximisation that takes into account the costs of monitoring and a positive deadweight loss of taxation. The authors show that when producers are very risk averse, the socially optimal choice of parameters for the programme achieves the same results as the first-best situation.3511 Clearly, these are questions that will need more analysis in the future, as more such schemes come on stream.

5.5.

Assessment and conclusions The fact that decades of high levels of price support are a major factor behind the intensification of arable production and the resulting environmental deterioration does not mean that reducing support can put the whole machine into reverse gear. New technological options and management practices are needed to take the arable sector in a more sustainable direction. Even with significantly lower support, the sector will not return to the farming practices and styles of fifty years ago. Yet the high-cost, inputintensive technologies currently in use will certainly be modified and adapted in various ways. In the past, the agricultural research establishment and the large commercial companies serving agriculture responded to the policy signals given to farmers by developing and promoting yieldimproving, labour-saving technology to help farmers exploit the opportunities created by the productivist policy climate. Nowadays farmers are receiving new signals and being set new targets related to sustainability and environmental balance. It is important that the inventiveness and commercial drive of the upstream industries are harnessed to help producers turn this corner successfully. Policy mechanisms such as environmental cross compliance and specific agri-environmental measures, as well as aiming at tangible changes in farming practice and performance within existing options, can also be expected to send a strong signal to the industries supporting agriculture that it is worth developing new technological options for farmers to help them follow this new policy orientation. It is suggested in Section 5.2 that farmers responding to price reductions alone may not always adjust their production in ways that secure the AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 215

expected environmental benefits. Thus, there is a strong case for supplementary policies that provide clear guidelines and additional incentives by means of environmentally oriented policies, be they crosscompliance provisions or more targeted agri-environmental policies. Section 5.3.2 lists the potential advantages and disadvantages of environmental cross compliance. Some of the advantages seem to be of only transitional importance. For example, one can imagine that within ten years the need to raise farmers’ awareness of the environmental consequences of their farming practices, or of the existence of agri-environmental measures, will have all but disappeared. On the other hand, most of the potential disadvantages seem more likely to persist over time. Moreover, the disadvantage, observed by many authors, that the cross-compliance incentive will weaken with a phasing down of direct payments could possibly act as a brake on the process of shifting away from “pure” income support and towards programmes aiming at environmental and other objectives via cross-compliance provisions. It is unlikely, however, that the need for “better harmonisation of agricultural and environmental policies” will become irrelevant in the medium term. Two important questions thus arise: to what extent does environmental cross compliance achieve a reconciliation between these two objectives and how much further can it be taken? The analysis and discussion in Section 5.4 show that when the cost effectiveness of an environmental cross-compliance provision grafted onto a direct income support programme is measured relative to the incremental cost of the cross compliance only, the cross compliance provision will score very highly. Undoubtedly, this feature makes the introduction of such a programme look very attractive. By piggy-backing on an existing policy measure, environmental improvements are secured at low additional cost. However, even with this programme, the income support and environmental objectives are in conflict. If the income support payments are high enough and the cross-compliance conditions are sufficiently modest, all producers will find that the programme improves their income. However, in this case, by definition, the cost in income support payments will be very high, or the environmental benefits will be small, or both. On the other hand, if the cross-compliance conditions were set so as to aim for a significant impact on environmental targets, some producers would either suffer an income loss (when remaining in the scheme was compulsory), or leave the programme (when participation was voluntary). Gross environmental benefits will be lower, and net environmental benefits may also be lower. As direct income policies become more effective in delivering environmental improvements, it is likely that, because of the way payments

216 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

are allocated and earmarked largely as compensation for additional costs, the whole thrust of the policy will be deflected away from the original “pure” income support objective. This can happen for several reasons. First, because of large differences between farmers in their scope for providing environmental improvements and services, the distribution of environmentally targeted payments will probably not correspond to policy makers’ preferences for the distribution of income support. Second, because the most cost-effective methods of allocating a given amount of funding involve “self-revelation methods” (such as competitive bidding), most of the payment received will go to covering additional costs, and farmers’ policy rent and information rent will be reduced to very low levels. Inevitably, improving the net environmental impact of direct income support plus cross compliance involves targeting producers and environmental objectives in such a way that the income support objective becomes subordinated. At the limit, the package “direct payments plus cross compliance” is transformed into a set of dedicated, targeted agrienvironmental policies. Once the income support objective becomes secondary, it becomes more appropriate to consider the environmental cost effectiveness of the policy in relation to the whole budget cost (including the payments). Clearly, this causes an apparent deterioration in cost effectiveness. In conclusion, cross-compliance provisions attached to direct income support programmes that are determined by, or originate from, agricultural support programmes are unlikely to reconcile the long-standing conflicts between agricultural support policies and environmental policies aimed at reducing negative externalities. In the short run, they may offer new opportunities to gain some small environmental improvements at low additional cost, but subsequent attempts to improve the environmental performance of the package will reveal the underlying unresolved opposition between the two policy objectives. In the longer term, policy is likely to move in two directions simultaneously. First, certain restrictions that are urgently needed to address threatening environmental problems and that are considered too expensive to secure through additional voluntary, compensated measures, could become legally binding on producers, independent of any support payments. This will inevitably imply a shift in producers’ property rights relating to resource use. Second, voluntary agri-environmental measures will expand in order to promote and support more actively the role of the farmer as a natural resource manager, engaged in both farming and in other valued, but non-marketable, activities, and enabling him/her to earn an acceptable income from this portfolio of activities without the need for that income to be “supported” by pure income transfers. It is precisely because some of the AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 217

resource management activities required of farmers are non-marketable that this will involve an important role for policies that substitute for markets and, hence, for targeted agri-environmental measures. More work is needed to develop policies that are cost-effective when producer heterogeneity, incentive mechanisms and transaction costs are taken into account. If this scenario materialises, rather than just creating more isolated and independent programmes for specific environmental benefits and services, serious attention should be given to the development of a co-ordinating and consolidating framework in order to rationalise the various environmentally inspired initiatives and ensure that the whole is at least as effective as the sum of its parts. More systematic co-ordination of agri-environmental policies both at the design stage, and, even more importantly, at the implementation stage should in no way compromise the need for specific details of programmes to respect local conditions.

218 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Notes 1.

See, for example, Chapter 2; OECD, 1998d and 2001d; Shortle and Abler, 1999; McRae, et al, 2000; Claassen, et al., 2001.

2.

This chapter follows the literature in focusing on the negative environmental impact of intensive arable production. This is not to deny that arable cropping can also have positive environmental effects in some regions and with certain cropping systems.

3.

The only useful example is that of New Zealand’s major policy reform of the mid-1980s. Among the consequences was a 40% fall in fertiliser use, which did not regain its former level for 10 years. Pesticide use was still below its pre-reform level after the same period (OECD, 2000a).

4.

This management approach to set-aside was also found by Winter and Gaskell (1998) in a representative sample of 552 farmers in Great Britain.

5.

Babcock and Hennessy (1996) found that –for all reasonable levels of risk aversion– when producers have crop insurance, optimal fertiliser application declines, indicating that insurance and fertiliser use are substitutes.

6.

In Hennessy’s paper and other studies using his approach, it is assumed that the only source of income variability is price risk.

7.

Small farms with less than 92 tonnes of cereal and oilseed output were excluded from set-aside payments.

8.

However, EU eligibility for agri-environmental support is conditional on respect of usual GFP throughout the farm. In the future, receipt of agri-environmental payments would be subject to cross-compliance requirements on the whole farm.

9.

For example, for Claassen, et al. (2001) “the term ‘green payment’ refers to a subset of agri-environmental payment programs that have both environmental and farm income objectives”. As an illustration, the authors cite a farm income support payment conditional on the farmer implementing conservation practices, such as conservation tillage or nutrient management. By contrast, Horan, et al. (1999) define a green payment as “any payment to producers based on either specific actions taken to reduce non-point pollution or on the probable environmental results of such actions”. In this second definition, the income support objective has disappeared, and the type of environmental effect targeted is limited by the reference to non-point pollution.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 219

10.

Buller, et al. (2000), in assessing the EU’s voluntary agri-environmental schemes, conclude: “[voluntary] agri-environmental schemes have not challenged dominant agricultural models, but have rather sustained certain often more marginal farming systems and the countryside they produce from changes that are to a large degree consequent upon such dominant agricultural development trajectories (such as intensification and regional specialisation)". Because of the conflict between the strong productivist incentives of the CAP and the relatively soft, voluntary, environmentally oriented agri-environmental measures, Reg. 2078/92 (agri-environmental measures of the 1992 CAP reform) has had “a notable lack of impact on intensive areas” (EC, 1998).

11.

Claassen, et al. (2001) cite wetland conservation as an example of a cross-compliance objective that would be very costly and technically difficult to address through a voluntary programme.

12.

In a survey across eight EU countries, Falconer and Whitby (1999) found administration costs of voluntary agri-environmental measures introduced after 1992 that ranged from under 7% of the compensation payment, to over 87% of the compensation payment. They quoted administration costs for agricultural commodity regimes from other authors as ranging from under 2% to 20% of total public costs (that is, administration included in the total). Administration costs of the CAP arable and livestock direct payment schemes were under 5% in all countries, except Germany.

13.

Drake, et al. (1999) reported that the main reason given by farmers in eight different EU countries for non-participation in agri-environmental programmes was lack of information.

14.

For example, when market prices are high, converting wetlands or ploughing up HEL is economically more attractive to farmers (Claassen, et al., 2001).

15.

Baldock and Mitchell (1995) hypothesise that cross compliance may give an extra incentive to producers in this case.

16.

This proposition is advanced by Christensen and Rygnestad (2000). Similarly, when categorising US cross-compliance provisions as involuntary – although, in theory producers can opt out of commodity programmes – Claassen, et al. (2001) state that producers see support payments as unconditional entitlements, which are built into their financial calculations, not as compensation for meeting the compliance conditions. In fact, this is an empirical question on which hard evidence seems to be lacking so far.

17.

If compliance conditions relate to statutory requirements, this argument is weak, in that it applies to all legal requirements that impinge on businesses and individuals. Dwyer, et al. (2000) suggest ways in which homogeneous

220 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

compliance conditions could be made more flexible in order to take farmer heterogeneity into account (to a limited degree): farmers could be allowed to choose from several sets of conditions – or a points system could be used, under which farmers could combine options until they reach the required number of points. 18.

A 1997 USDA survey found 1 674 different sets of practices in approved conservation plans under the conservation compliance provision. There was much regional variation, although “plans involving conservation cropping sequences, conservation tillage, crop residue use, or some combination of these practices were applied on 54% of land subject to the regulations” (Claassen and Horan, 2000).

19.

This philosophy was set out by the EC (EC, 1998): “The underlying rationale for integrating environmental concerns into agriculture rests on two principles: –Firstly, farming as any economic sector, should attain a basic standard of environmental care without payments. This should be contained within the scope of good farming practice (which includes many matters other than environment) and comprises observance of regulatory standards and an exercise of care which a reasonable farmer would employ. This basic standard is also referred to as the reference level: –Secondly, wherever society asks farmers to provide an environmental service beyond the reference level, and the farmer incurs cost or income loss, society must expect to pay for the service. This standard is also known as the target level. Cross compliance is most appropriate in ensuring adherence to the reference level.”

20.

The term “opting out” is used to mean both not claiming and not complying. This should be distinguished from non-compliance, which is claiming although not complying.

21.

For example, Koroluk, et al. (2000) reported that in Canada 60% of farmers use soil N testing, and of this 60%, 75% do so every 1 to 3 years. The same environmental assessment, however, reports that over 50% of the assessed area across the country registered increases of more than 5 kg/ha of residual nitrogen between 1981 and 1996.

22.

For 6 of the 8 farm types, N applications fell significantly after N testing was introduced, but the share of N taken up in the crop increased. Moreover, for three of the four farms continuing maize production there was an increase in net returns per acre. The authors conclude that the value of N testing varies from farm to farm, depending on cropping history and soil characteristics. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 221

23.

This process has been incorporated, to a limited extent into, the 2003 CAP reform.

24.

A progression over time from Option 1 towards Option 4 is rather similar to the progression from the CAP to CARPE (“Common Agricultural and Rural Policy for Europe”) that was advocated by the Buckwell group (EC, 1997), although using cross compliance in an intermediate stage was not envisaged in that proposal.

25.

It seems unrealistic to suppose that producers in Groups 3 and 4 would continue to participate, rather than opting out – i.e. simply not claim the payment. For this scheme to become a possibility in the real world, we have to assume that the compliance conditions are statutory requirements, so that opting out of the payment still does not remove the onus to comply. Alternatively, producers will not perceive opting out to be an option.

26.

The full programme costs are (N1 + N2)d + A(N1 + N2), where Ni is the number of hectares operated by producers in the i-th group. Incremental cost is A(N1 + N2).

27.

If this occurs, then effectiveness per dollar spent will also improve, since 2

4

i 1

i 1

¦ (BC i CC i ) ! ¦ (BC i CC i ) ,

(N1 N 2 )d < Nd

and

A (N1 N 2 ) d A(N) . 2

28.

This is because

BC1 CC1 ! ¦ (BC i CC i ) , N1d < (N1 N 2 )d and i 1

A (N1 ) d A(N1 N 2 ) . 29.

The complaint is sometimes made by non-farmers that environmental payments reward farmers for doing what they were already doing, or would have done without the payment.

30.

For example, if the environmental objective is reduction of soil erosion, the Universal Soil Loss Equation or the Wind Loss Equation could be used to do this (Claassen, et al., 2001).

31.

For many, these schemes would not qualify for the definition of “income support + cross compliance”, but instead would be classified as “green ticket” programmes.

32.

Claassen, et al. (2001) give details of the construction of the Environmental Benefit Index (EBI) which is used as a proxy for (bc) in constructing the ratio of benefits to cost (bc/b) with which to rank bidders for a CRP contract.

222 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

33.

See, for example, contributions to the seminar in the EU Concerted Action “Developing Cross Compliance in the EU”, 2-3 June 2003, www.bal.fal.de/en/.

34.

In this situation, unless there is also a positive fine for non-compliance, the expected penalty for opting in and not complying will always be less than the loss from opting out, so risk-neutral and amoral producers would be indifferent towards these two strategies (Latacz-Lohman and van der Hamsvoort, 1999).

35.

See, among other studies, Segerson (1988) and Malik (1993), who have looked at aspects of monitoring actions or results in payment schemes where compliance or individual action is unobservable to the regulator.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 223

This page intentionally left blank

Chapter 6 ENVIRONMENTAL IMPACTS OF MULTILATERAL AGRICULTURAL TRADE LIBERALISATION ON ARABLE CROPS 6.1.

Introduction The URAA marked an historic point in the reform of the agricultural trading system (OECD, 2001f; Diakosavvas, 2004). The URAA imposed disciplines on trade-distorting domestic policies, and quantitative constraints were agreed for market access, domestic support and export subsidies. Notwithstanding the progress that has been made in liberalising agricultural trade, the level of support to producers of certain arable crops remains high. In many OECD countries, there is a wide range of agricultural and trade policies that affect the arable crop sector – domestic support, import tariffs and export subsidies (Chapter 3). Thus, further trade liberalisation and reduction in support to agriculture could have a significant impact on the output of arable crops. These impacts may include changes in cropping pattern and trade, as well as changes in scale (i.e. an expansion or contraction of the whole crops sector), composition of output (e.g. more wheat but less rice), and changes in farming practices (e.g. more extensive and less intensive output systems). The state of the natural and environmental resources associated with arable crops may be affected through changes in the scale and composition of inputs used in production (e.g. land, water, fertilisers, pesticides, etc.). Moreover, as has been frequently pointed out, many of the environmental effects are location-specific. Vasavada and Nimon (2003) reviewed the empirical literature regarding the environmental impacts of trade liberalisation and found that the overall impact may depend on several – and possibly offsetting effects – and that co-ordination of trade and environmental policies may promote a more sustainable pattern of natural resource use. Tsigas, Gray and Hertel (2002) extended the GTAP framework to account for linkages between trade and environment. Using pollution and abatement data from the United States, three sources of pollution were considered for agriculture: soil erosion, toxic pesticide releases, and livestock waste: in terms of agri-environmental AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 225

policies, a soil conservation programme and regulations regarding pesticide use and livestock waste were considered. The authors concluded that, if environmental resources are priced correctly and environmental regulations are enforced, trade liberalisation will most likely lead to less pollution. Cooper, Johansson and Peters (2003) analysed the regional environmental impacts of multilateral agricultural trade liberalisation in the United States. In particular, the authors examined impacts on soil erosion, fertilisers and pesticides lost to water, and pollution from manure. Two economic models and an environmental model were used: the ERS/PSU World Trade Model determined changes in global trade and US output; the USMP spatial equilibrium model determined output and input changes at the US regional level; and a spatial environmental model determined changes in physical environmental measures. The authors concluded that for the United States as a whole, environmental impacts stemming from multilateral agricultural trade liberalisation would most likely be small (i.e. less than 1.4%). Changes in the relevant environmental indicators, however, are not uniform across the United States and there may be large differences between various regions and the United States taken as a whole. For example, soil erosion due to wind may increase by 1.3% for the United States as a whole, but it may increase by 9.3% in the Northern Plains region; phosphorus fertiliser loading to water may increase by 0.3% for the United States as a whole, but decline by 0.6% in the Southern Plains region. Lehtonen, Aakkula and Rikkonen (2004) assessed the ecological, economic and social sustainability impacts up to 2020 in Finland of four alternative agricultural policy scenarios. The scenarios analysed are: prolonged Agenda 2000, 2003 CAP reform; integrated rural and environmental policy (i.e. full decoupling); and full-scale agricultural trade liberalisation. A dynamic regional model of Finnish agriculture, the DREMFIA model, is used. The results show that a partial decoupling of agricultural support from commodity output and moderate reductions of commodity prices could be expected to yield environmental benefits. However, full decoupling and radical price reductions would be unlikely to bring about any additional environmental benefits, but would result in significant down-scaling and regional concentration of remaining agricultural activities. Moreover, there would be a clear trade-off between environmental benefits and output volume and intensity. In terms of nutrient leaching from land into water, which is considered to be the most severe environmental problem in Finland, the nitrogen balance would be reduced in all scenarios. The reduction is of the same magnitude in the 2003 CAP reform; integrated rural and environmental policy; and full-scale agricultural trade liberalisation scenarios. This outcome is possible because lower milk prices reduce the optimal nitrogen level of silage in the 2003 CAP scenario,

226 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

while in the full-decoupling and full-trade liberalisation scenarios the lower level of fertiliser use due to lower grain prices is partially offset by the concentration of livestock production and the reduction of grain area. On the other hand, the phosphorus balance may increase in the full-trade liberalisation scenario, due to both concentration of production in a few areas and imports of feedgrains. Using the same model and similar policy scenarios, Miettinen, et al. (2004) compared the diversity effects of alternative agricultural policy reforms in Finland. Their results indicate that if agricultural support is independent from output, the amount of fallow land would increase considerably. At the landscape level, this change decreases the diversity of arable crops and reduces the amount of set-aside, as a result of both 2003 CAP reform and free trade liberalisation scenarios in all regions of the country, except northern Finland. The effects on biological diversity may, however, not be harmful, since green fallow has some positive consequences, especially for the species density and richness of farmland birds. The cultivated area of winter wheat would decrease as a result of each scenario studied. This development is unfavourable from the point of view of biological diversity and nutrient leaching, since winter cereals offer vegetation cover for the soil during winter. The pesticide application area would be smaller under 2003 CAP reform and free-trade liberalisation scenarios than would be the case with the base scenario (i.e. Agenda 2000) as cereal, potatoes and sugarbeet areas decrease when agricultural support is decoupled from output. The results also suggest large regional variations, particularly in the agricultural nutrient balances in Finland. Sinabell and Schmid (2004) estimated some environmental effects of the 2003 CAP reform for Austria using the PASMA model (a regional, linear programming model). The authors found that, compared to a business-as-usual scenario (i.e. continuation of the Agenda 2000 reform in the year 2003), the 2003 CAP reform would lessen environmental pressure at the aggregate level and slightly accelerate structural adjustment in terms of reducing agricultural inputs and output, and stimulate the adoption of environmentally friendly management practices. In particular, decoupling would lead to significant declines in arable crops and beef output, while the nitrate balance at the national level would improve. This chapter analyses some of the environmental impacts of further multilateral trade liberalisation and reduction in support on arable crops, using various indicators of environmental quality. State-of-the-art quantitative techniques are used to quantify impacts on trade, output and input use for arable crop sectors in every economy in the model. Changes in the scale and intensity of input use for arable crops are analysed in order to

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 227

enable a discussion of the environmental implications. In particular, the study assesses the impacts on the following agri-environmental indicators:

x

use of chemical fertilisers;

x

pesticide use;

x

nitrogen uptake and off load; and

x

emissions of GHGs from arable crop production.

Two hypothetical multilateral agricultural trade liberalisation scenarios are considered. The first scenario assumes an extension of the URAA, while the second scenario assumes full multilateral agricultural trade liberalisation. This analysis, which is undertaken on a cross-country level, does not consider the alleviating influence of existing environmental policies and regulations. This cross-country quantitative analysis is supplemented with some country-specific disaggregated analysis. More specifically, with the collaboration of AAFC and the ERS of the USDA, aggregate country results are used as inputs into spatial, regional and environmental models in order to assess the environmental impacts of trade liberalisation at the regional level for Canada and the United States. These countries have been selected primarily due to the availability of the appropriate spatial environmental models which are now widely used for this type of policy analysis. Moreover, the diversity of crop farming systems in these countries, together with the use of various policy instruments affecting the arable crop sector, makes their choice an interesting case to study.

6.2.

Cross-country analysis 6.2.1.

The liberalisation scenarios

The environmental impacts are assessed against two hypothetical multilateral agricultural trade liberalisation scenarios. The first scenario assumes an extension of the URAA. This scenario incorporates changes within each of the major negotiation pillars – market access, export competition and domestic support. In particular, it is simulated that all food and agriculture tariffs (including direct price impacts of TRQs) are reduced by 36% in developed countries and by 24% in developing countries; domestic support is reduced by 20% in developed countries and by 14% in developing counties; and export subsidies are reduced by 36% in developed countries and by 24% in developing countries. The second hypothetical scenario involves the elimination of all food and agriculture policy measures (i.e. import tariffs and TRQs, domestic support and export subsidies) in all countries. In each scenario, the

228 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

simulated liberalisation is limited to the food and agriculture sector; for example, all food and agriculture import tariffs are reduced or removed, but those on industrial products remain unchanged. Although the two liberalisation scenarios reflect some of the elements of various country proposals submitted to the current round of WTO multilateral trade negotiations, they do not represent the negotiating position of any particular country. It could be argued that these two scenarios represent extreme cases of numerous proposals presented to WTO and they serve to define low and upper boundaries on the results. Thus, although the simulated results of the full-trade liberalisation scenario do not necessarily demonstrate the effects of domestic reforms (such as the 2003 CAP reforms) they are indicative of the upper bounds of environmental effects of such unilateral reforms.

6.2.2.

Methodology

The methodology is similar to that undertaken for the OECD dairy sector study (OECD, 2004a). A multi-country, global trade model is used to simulate the impacts of further trade liberalisation. The model is based on standard economic theory and it allows consideration of the economy-wide impacts of policies by explicitly accounting for upstream and downstream linkages, inter-sectoral competition for resources, and price and income changes. It has now become standard practice to analyse the impacts of multilateral trade policy liberalisation within global trade models that rely on applied general equilibrium (AGE) methodologies; some of the analyses that relied on AGE frameworks are: Francois, McDonald and Nordstrom (1996); Hertel, et al. (1996); Harrison, Rutherford and Tarr (1996); Anderson and Ingo (1999); Elbehri, et al. (1999); Tsigas (2001); Beghin, et al. (2002); and Rae and Strutt (2003). Several of these studies are based on the GTAP (Global Trade Analysis Project) framework (Hertel, 1997). The GTAP framework consists of a global database and a global trade model. The database includes information on trade; domestic output and use of each commodity; and land, labour, and capital employment, by sector. The database also contains information on trade and support policies. The model is based on assumptions that are common in the literature: perfect competition, and constant returns to scale. It is also assumed that the policy changes under consideration do not affect the aggregate level of resource employment; the policy changes are assumed to affect the sectoral allocation of resources.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 229

The analysis reported here is based on policy simulations using a modified version of the GTAP model. Key technical features of the standard and modified versions of GTAP are presented in Annex 6.A. Section 6.2.4 briefly discusses the sensitivity of simulated impacts to model specification. In particular, simulated impacts from the standard GTAP model are compared to the simulated impacts from GTAP model applied in this analysis. The comparison suggests that the two model specifications give quite similar conclusions. The simulated impacts on quantities of outputs and inputs are all reported in percentage changes. In some cases, however, large percent changes may refer to very small sectors and are thus of small consequence.1 The arable crop sectors which represent a very small share of world production (i.e. less than 0.05%) include: rice in Canada, New Zealand, EFTA, the ten new member states of the EU (EU10), Denmark, Finland and Sweden, and Germany; wheat in Korea; and oil-bearing crops in New Zealand.

6.2.3.

Simulated environmental impacts agricultural trade liberalisation

of

multilateral

6.2.3.1. Output supply, input use and input intensity At the global level, output impacts of partial trade liberalisation would be small: rice output would decline by 0.5%; wheat and other grains by 0.3%; and oilcrops output by 0.7%. The aggregate implications for production in the EU15 are that output of rice, wheat, other grains, and oilcrops would decline by 1.0%, 4.0%, 2.7%, and 9.1%, respectively. Table 6.1 summarises the implications of partial trade liberalisation on land use and chemical use for arable crops. In Japan, Mexico, the United States, EFTA, and in the EU15 regions, both land and chemical use would decline due to trade liberalisation. In New Zealand and the EU10 both land and chemical use would expand. In Australia and Canada the area of land used for growing arable crops would decrease, but chemical use would increase. In Korea land use would remain unchanged and chemical use would decrease. Partial trade liberalisation would lead to the largest expansion of aggregate arable crops output in New Zealand (6%), although the size of the arable crop sector is very small; the expansion in output is primarily due to an increase in area and to intensification of production.2, 3, 4

230 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 6.1. Arable crops: summary of land- and chemical-use impacts of partial trade liberalisation More New Zealand EU10

Japan Mexico, USA EFTA, EU15

Australia Canada

More Less

Land use

Chem ical use Less

Note: See Table 6.A1 for definition of regional groupings.

At the global level, even under the scenario of full multilateral agricultural trade liberalisation, output impacts would be small: rice output would decline by 4.1%; wheat by 0.3%; other grains by 0.9%; and oilcrops by 2.1%. The aggregate implications for EU15 production are that output of rice, wheat, other grains, and oilcrops would decline by 3.7%, 6.7%, 7.6%, and 26.7%, respectively. Table 6.2 summarises the implications of full trade liberalisation for land use and chemical use in the arable crop sector. In Australia, Japan, Mexico and the United States and in the EU15 as a whole both land use and chemical use would decline as a result of full trade liberalisation. In New Zealand and the EU10 both land and chemical use would expand. In EFTA and in Canada land use would move in the opposite direction from chemical use. In Korea land use would remain unchanged and chemical use would decrease. Aggregate arable crops output would rise by the highest percentage (16%) in New Zealand – although from a small base. This increase is attributable to a 4% increase in area and a 4% increase in other inputs. Table 6.2. Arable crops: summary of land- and chemical-use impacts of full trade liberalisation Chemical use More

EFTA

Less

Land use

Less

Australia Japan Mexico, USA EU15

More New Zealand EU10 Canada

Note: See Table 6.A1 for definition of regional groupings. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 231

In the United States, the simulated impacts of partial and full trade liberalisation suggest that the output of arable crops and the use of chemicals would decline. The intensity of chemical use, however, would also increase, but not more than 1%. For Canada, the simulated impacts suggest that arable crop output and use of chemicals would increase by less than 2%; the intensity of chemical use would also increase, but not more than 5%. Table 6.3 summarises the information presented thus far. In particular, Table 6.3 identifies the arable crop-specific impacts on output and chemical intensification that are larger than 10%:

x

The most striking feature of the table is that it is sparsely populated.

x

For several regions (e.g. Japan; the United States; Mexico, the EU10 all impacts are less than 10%.

x

For all regions, the partial trade liberalisation impacts on output are less than 10%.

x

Only under the full liberalisation simulation do some sectors increase by 10% or more in output.

The information in Table 6.C3 identifies crop/region combinations that might warrant further investigation to determine whether they pose environmental problems.

6.2.3.2. Nitrogen balance and pesticide use Figure 6.1 shows simulated impacts on the balance of nitrogen. Annex Table 6.C1 shows the components of gross nitrogen balances. Under partial trade liberalisation, nitrogen uptake increases by less than 1% in Canada, but fertiliser use increases by 1%, which implies that there might be a 1% larger nitrogen surplus in the soil. Full trade liberalisation would cause the nitrogen surplus to increase by 2% in Canada. Nitrogen surplus might also increase due to policy liberalisation in New Zealand. As Figure 6.2 suggests, nitrogen balances would decline in the EU15 as a whole. Using information on the total use of pesticides (OECD, 2003g) and simulation results, it was found that the use of pesticides in OECD economies, as a whole, would decline by 3.4% under partial trade liberalisation and by 11.2% under full trade liberalisation.

232 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 6.3. Output (O)- and chemical-intensity (I) impacts of trade liberalisation larger than 10%, by arable crop

Region

Partial liberalisation Other Oil Wheat grains crops

Rice

Rice

Australia Canada EU15 EU10 Japan Korea Mexico New Zealand USA EFTA Argentina Brazil China RoAm RoAs RoE TMMENA RoAf RoW

Full liberalisation Other Oil Wheat grains crops

O, I

I

O,I O O O

O

O

O

Note: See Table 6.A1 for definition of regional groupings.

Figure 6.1. Arable crops: gross nitrogen balance impacts of trade liberalisation for selected regions 20 10 0

(%)

-10 -20 -30 -40 -50

P artial libe ralis ation

EU10

EFTA

PSIG

ABNL

Germany

France

UK/Ireland

DFS

Mexico

USA

Canada

Korea

Japan

New Zealand

Australia

-60

F ull liberalisa tion

Note: See Table 6.A1 for definition of regional groupings.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 233

6.2.3.3. Greenhouse gas emissions The GHG emissions that are directly related to arable crops are methane emissions from rice cultivation and nitrous oxide emissions from agricultural soils (Chapter 2). The simulations suggest that partial trade liberalisation would cause global rice output to decline by less than 0.5%; full trade liberalisation would cause global rice output to decline by 4.1%. The regional pattern of rice impacts, however, is somewhat different between the two trade liberalisation scenarios. Thus, the impact of the two trade liberalisation scenarios on global methane emissions would not be proportional to changes in global rice output. Simulated impacts suggest that partial trade liberalisation would cause global methane emissions to decline by 1% and full trade liberalisation would cause a decline of 9%.5 Simulated impacts suggest that partial trade liberalisation would cause global nitrous oxide emissions from soils under arable crops to decline by 2% and full trade liberalisation a decline of 6%.

6.2.4.

Sensitivity analysis

Annex Figures 6.C1 to 6.C3 explore the sensitivity of simulated impacts to model specification. In particular, the figures compare selected impacts from partial liberalisation under the standard GTAP model and the revised model adopted in this analysis, i.e. the revisions implemented concerning output functions in agriculture. A comparison of impacts for arable crops as a whole suggests that the two models are not producing significantly different impacts. In almost all cases the impacts suggest the same direction of change and they are close in magnitude.

6.2.5.

Caveats

The simulated arable crop impacts of trade liberalisation suggest a complex set of environmental implications. For several regions, there would be little change in land and chemical use, aggregate output, and the rate of chemical application. Thus, the environmental implications for these regions probably would be minimal. For some regions, however, crop output and chemical use would expand, even though cropland would not expand. Thus, environmental issues that are associated with the degree of intensity of output might warrant more attention. Although GTAP is among the most sophisticated applied general equilibrium models currently available, and despite the methodological improvements of the model used in this study, it necessarily involves some simplifications and abstractions from the real world. First, as the model is static in nature, it cannot trace the adjustment path of the reduction of support. Second, although the GTAP agricultural policy data have been updated to 2001, they do not reflect subsequent important agricultural policy

234 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

developments. Some of these are: the entry of China and Chinese Taipei into the WTO (thus, the data overstate China’s oilseeds and grains tariffs); the 2002 FSRI Act in the United States; full implementation of the Agenda 2000 CAP reform in the EU; the more recent reform of the CAP agreed in 2003, as well as the enlargement of the EU to twenty-five members. These shortcomings are important and will have an impact on market developments. However, the broad conclusions of this analysis are not expected to be altered substantially because, as stated earlier, the full trade liberalisation simulation scenario considered here is an extreme scenario which encompasses the effects of such reforms as special cases. Further, as in the OECD pigs and dairy studies (OECD, 2004a; 2003f), detailed analysis of policy reforms in specific countries is beyond the scope of this analysis. OECD has analysed the 2003 CAP reform and of the 2002 FSRI Act (OECD, 2004c; 2003d). Box 6.1 summarises the results of some studies on the economic and market effects of the 2003 CAP reform and the EU accession.

6.3.

Regional environmental impacts of agricultural trade liberalisation 6.3.1.

Canada

Impacts from the GTAP analysis are incorporated into the Canadian Regional Agricultural Model (CRAM), which is similar methodologically to the USMP (see Annex 6.B). The CRAM allows for both inter-provincial and international trade in primary and processed products. One of its important features is that it takes into account the interdependency of crop and livestock output. The CRAM model has been used by AAFC to link to AEI models. Recently, it has been used to establish environmental policy outcome targets for federal/provincial implementation agreements as part of the Agricultural Policy Framework (Heigh and Junkins, 2005). For the purpose of this study, six agri-environmental indicators related to soil, water, air quality and biodiversity were calculated. The results are shown in Table 6.4.

Soil-related impacts The risk of soil erosion by water is a concern in all of Canada’s agricultural regions. In many areas, fine-textured, erodible soils are exposed to erosion by rainfall and run-off. The risk of water erosion is usually greatest on inherently erodible landscapes under intensive cultivation. Under both trade liberalisation scenarios, soil erosion by water (RWE) decreased mainly due to shifting small amounts of cropland into forage, since hay provides more protection against soil erosion than occurs with annual crops.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 235

Box 6.1. Selected studies on 2003 CAP reform and the EU10 The OECD Secretariat’s preliminary analysis of the 2003 CAP reform, which treats the EU as an aggregate of the EU15 member states, suggests that the composition of the support will be significantly modified but there will be no change in the level of support; a shift from cropland to pasture land will occur and there will be a reduction in the number of cows per hectare (OECD, 2004c). The results depend on the structure and the sectoral coverage of the models; on the assumptions made; on the parameters introduced in the modelling tools; and on the base year used. Moreover, certain aspects of the reform, such as the regional option, cross compliance and certain sectors (e.g. beef) are not taken into account. The report also points out that other aspects of the reform, including environmental impacts, are difficult to evaluate at this stage due to the fact that EU member states are given lots of flexibility in setting specific criteria, and implementation details are important in such evaluations. According to analyses undertaken by researchers from the Danish Research Institute of Food Economics and by the EC, the impact of CAP reform on the EU10 will be to moderate, but not reverse, the changes expected in their agricultural sectors resulting from the CAP. That is, projected increases in cereals and beef output in the EU10 are likely to occur, but will be less under CAP reform than under Agenda 2000. The same is true of output changes forecast for the EU15. Jensen and Frandsen (2003) carried out an analysis of the impacts of EU enlargement under multiple scenarios, including Agenda 2000 and three CAP reform options. They conclude that for 21 out of 22 commodity categories in the EU10 and 20 out of 22 for the EU15, changes in output pre- and post-enlargement under Agenda 2000 are of the same direction as those pre-enlargement and post-enlargement under the CAP reform scenarios. An EC study carried out in March 2003, based on the EC proposal for CAP reform of January 2003 (EC, 2003b), largely supports these conclusions. Arable crop output is expected to shift towards oilseeds production in the enlarged EU as a whole, as well as towards soft wheat and barley, which are expected to see improved market conditions due to accession. Decoupling increases farmers’ welfare through increases in both output efficiency and transfer efficiency, when compared with Agenda 2000. Following a request from the Dutch Ministry of Agriculture, Nature Management and Food Quality, LEI, using the GTAP model, analysed, inter alia, the potential impact of the 2003 CAP reform and the Harbinson Proposal on the Netherlands, EU14 and the EU accession countries (Lips, 2004). It finds that for arable crops, in the EU14 the 2003 CAP reform leads to a small decrease in output, while in the Netherlands the reform leads to a larger decrease in output, and a slight increase in the accession countries. Assumptions about the effects of decoupling and the difference between applied and bound tariff rates influence the results.

236 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 6.4. Regional land use and environmental impacts of trade liberalisation Partial liberalisation (%) Province

Cropland

British Columbia Alberta Saskatchewan Manitoba Ontario Quebec New Brunswick Prince Edward Island Nova Scotia Newfoundland Canada

SummerHayland GHG IROWCN RSN RWDE RWE fallow

-2.7 -0.4

-0.8 -0.9

1.3 1.6

1.1 0.8

n.a. n.a.

0.9 3.0

0.2 -0.3 -0.9 -0.7

-1.3 -0.9

0.8 1.6 2.3 0.8

1.7 1.0 1.4 2.3

n.a. n.a. -0.3 0.6

2.8 2.8 -0.3 0.6

-0.5

0.4

1.6

0.4

-0.8 -1.1 -2.0 -0.2

1.4 0.3 0.1 1.4

1.2 2.1 0.9 1.3

1.2 0.8 0.4 n.a.

-1.2

HA

-0.6 -0.6

0.4 0.3

-0.9 -0.2 -0.9 -0.3

0.2 0.3 0.6 0.3

0.4

-0.1

0.2

1.2 0.8 0.4 1.5

-0.3 0.1 n.a. n.a.

0.4 0.1 0.1 0.3

-0.2 0.4 -0.2

0.2

Full liberalisation (%) SummerHayland GHG IROWCN RSN RWDE RWE fallow

Province

Cropland

British Columbia Alberta

-7.3 -0.8

-0.1 -0.2

3.5 2.3

3.1 5.6

n.a. n.a.

1.5 5.8

-0.1 -1.0 -0.1 -0.3

-1.3 -0.1

3.5 4.4 0.2 0.4

8.2 3.7 3.2 2.4

n.a. n.a. 3.6 1.9

5.7 6.6 3.5 2.0

-0.2

0.1

2.2

0.9

-0.5 -0.8 -2.0 -0.4

0.9 0.2 0.1 2.3

1.6 2.8 0.8 4.6

1.7 1.0 0.4 n.a.

Saskatchewan Manitoba Ontario Quebec New Brunswick Prince Edward Island Nova Scotia Newfoundland Canada

-0.9

HA

-1.3 -0.9

1.0 0.4

-1.3 -0.7 -0.3 -0.2

0.4 0.7 0.1 0.1

1.0

-0.1

0.1

1.7 1.0 0.4 3.3

-0.4 0.0 n.a. n.a.

0.3 0.1 0.1

-1.4 -0.7 -1.6

-1.0

0.4

Notes: n.a. = not available. IROWCN = Indicator of Risk of Water Contamination by Nitrogen; RSN= Residual Nitrogen; RWDE = Risk of Soil Erosion by Wind; RWE = Risk of Soil Erosion by Water; HA = Habitat Availability.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 237

The risk of RWDE is applicable only in the Prairie Provinces where the land is relatively flat and there are few trees or obstructions to act as wind barriers. A risk of wind erosion indicator is used to monitor the extent of cultivated land at risk of wind erosion, particularly as a result of changes in management practices (McRae, et al., 2000). The indicator can also be viewed as an indirect measure of a change in soil quality. Because wind erosion is a process of soil degradation resulting in decreased soil quality, a declining erosion risk is considered positive in terms of soil quality. For the partial liberalisation scenario, wind erosion rates decreased in Alberta and Manitoba and increased slightly in Saskatchewan. Overall for the Prairie Provinces, the wind erosion rate slightly increased. For the full liberalisation scenario, wind erosion rates decreased in all three Prairie Provinces. Overall for the Prairie Provinces, the wind erosion rate decreased by 1%. This was mainly due to increases in hayland and decreases in cropland and summerfallow.

Water-related impacts The average change in RSN is 1.5% and 3.3% for partial liberalisation and full liberalisation scenarios, respectively. Changes in the RSN indicator for partial liberalisation were relatively small. For the full trade liberalisation scenario, residual nitrogen increased from 0.4% in Newfoundland to 5.8% in Alberta. Increases in residual nitrogen are mainly due to increases in fertiliser use and livestock production. Contamination of water by nitrogen from farms is a major environmental concern for the agriculture industry. Potential water contamination by agricultural nitrogen, in the form of nitrates, is directly related to two factors: the movement of water off farmland, either in overland flow or by leaching through the soil profile into groundwater; and the amount of surplus or residual nitrogen available. The IROWCN indicator measures the risk of contamination by nitrogen coming from farmland (McRae, et al., 2000). IROWCN was estimated for eastern Canada only, due to problems with the methodology for the arid regions in western Canada. The IROWCN increased for all provinces, with the exception of Ontario, where it decreased by 0.3% under the partial trade liberalisation scenario. With full trade liberalisation, risk is increased in all eastern provinces between 1% and 3.6% (Ontario). The increase is attributable primarily to increases in fertiliser use for all provinces and expansion in livestock production.

238 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Air quality Results of the analysis for the partial liberalisation scenario show a slight rise in the GHG indicator of 1.3%, with the biggest increase in Quebec. Higher GHG emissions, resulting from the increased use of fertiliser use and livestock were partially offset by the sequestration of carbon from new hayland. For the full trade liberalisation scenario, GHG emissions went up by 4.6%. This is mainly due to increased fertiliser use for wheat, other grains and oilseeds, and increased beef and pig production. The rise in GHGs is offset somewhat by increased soil organic carbon due to increased hayland and reduced summer fallow. The largest increases in GHG emissions are in the Prairie Provinces.

Biodiversity The types of species and their use of agricultural habitats are essentially constant over time. However, patterns of agricultural land use and cover may evolve over time in response to market conditions and other factors. The habitat index is sensitive to such patterns of agricultural land use, as they affect habitats for species. For agricultural land uses, summer fallow provides the lowest level of wildlife, followed by cropland, hayland and pasture. Pasture (pasture provides the greatest potential for wildlife habitat). The habitat availability indicator (HA) suggests that a decrease in summer fallow, and thus an increase in hayland, improves biodiversity in Canada. The HA indicator somewhat increased in both trade liberalisation scenarios.

Summary The quantitative analysis was completed by integrating an economic model with six existing AEI models. The intention of this analysis is to demonstrate the type and level of information that can be generated. The AEI models indicate how the trade liberalisation scenarios impact on the Canadian environment; specifically on air, soil, water and biodiversity. The overall regional analysis suggests that trade liberalisation has the expected environmental impacts. Air quality declines due to the increase of GHGs emissions by 4.6% for full trade liberalisation from baseline levels. An improvement in soil quality is represented by reductions in risk of wind and water erosion. The reductions in wind and water erosion vary according to province. Residual nitrogen increases by 3.3% at the national level for the full liberalisation scenario. Biodiversity, in terms of habitat availability, improves by 0.3% and 0.4% for partial trade liberalisation and full trade liberalisation scenarios, respectively. The changes of the risk of water contamination by nitrogen are lower for Central Canada than for Atlantic AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 239

Canada. These results suggest that regional variation is important, and hence the change in environmental pressures will also vary across regions as a result of trade liberalisation.

6.3.2

United States

Impacts obtained from the GTAP analysis are incorporated into a spatial regional and environmental model, which is available at the Economic Research Service of USDA (i.e. the US Regional Agricultural Programming Model (USMP) (see Annex 6.B). This model is a multi-commodity, spatial equilibrium model that allocates output practices regionally, based on relative differences in net returns among output practices, differentiated by rotation, tillage and fertiliser rates by region. The USMP model divides the continental United States into 45 regions, which are then aggregated into 10 farm production regions for ease of discussion (see Figure 6.2). Variables in the model include regional supplies, prices, and demands for 44 crop, livestock and poultry commodities and processed outputs, farm input use, farm income, government expenditures, participation in farm programmes, and predicted levels for nine agri-environmental indicators.

Figure 6.2. US farm production regions and USMP sub-regions

Production adjustments occur in terms of technique effects (such as changing management practices), composition effects (such as a changing product mix) and scale effects (such as in production levels), which have specific regional, agri-environmental implications. For instance, nitrogen fertiliser use in USMP can be reduced by decreasing acreage planted (scale

240 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

effect), shifting to output crops that use less nitrogen fertiliser (composition effect), or by reducing nitrogen fertiliser application rates on a given crop mix (technique effect). Crop output systems are differentiated according to rotation, tillage and fertiliser rate. Environmental indicators discussed here include soil erosion, losses of nitrogen and pesticides to ground and surface water, and soil carbon levels (see Annex 6.B.1). In general, anticipated environmental impacts from USMP for the US, following partial or total trade liberalisation on agricultural crops, are similar in sign and magnitude to the GTAP estimates (Figure 6.3). In some cases, however, the implications of those changes will vary depending on the specification of the indicator examined. For example, pesticides leaching to groundwater could be much lower than one might expect if looking simply at pesticide use. On average, all indicators are predicted to fall (implying that environmental quality improves) under the partial liberalisation scenario. Those changes are no more than 5% and most are less than 2%. In the full liberalisation scenario, the change in indicators similarly suggests that environmental quality would improve, with changes larger than those predicted for the partial liberalisation scenario, and also larger than those predicted by the GTAP analysis. The one exception involves the soil erosion indicators; while aggregate soil losses decline in the full liberalisation scenario, soil losses due to wind erosion could increase. Figure 6.3. Aggregate US environmental impacts, by GTAP liberalisation scenarios Excess N

N Run-off

N Leach

Soil Run-off

Wind Erosion

Pesticide Run-off

Pesticide Leaching

Carbon Emissions

5

0

-5

-10

-15

Partial

Full

-20

-25 % Change

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 241

Potential limitations of the national scale analysis are illustrated by examining changes in agri-environmental indicators at a less aggregated scale. At the regional level, changes in environmental conditions could be significantly different from what the national aggregate results would suggest, with some regions seeing relatively large decreases in environmental loadings, others seeing relatively small percentage changes and still others seeing increases (Figure 6.4). In this particular case, with the full liberalisation scenario, nitrogen indicators suggest increased loadings in the Pacific States. However, base levels are relatively small in that region, so the relatively large percentage changes belie total changes that are relatively small. Conversely, base levels for nitrogen in the Corn Belt are the highest of all regions, so even relatively small percentage changes can imply substantially higher absolute changes in nitrogen levels. The increase in nitrogen emissions from the Pacific States is driven primarily by increases in land use for agricultural production in that region. That result is driven, in turn, by price increases associated with reductions in land use and in output, in all other regions.

s U

.S .T

ot a

St c ci fi Pa

l

St

at es

at e

s ai n

Pl ai n

N Leach

ou nt

ut h So

N Run-off

M

er n

at es D

el ta

ut h So

St

ea st

Pl ai ns pp al ac hi a A

er n

t or th N

Be l

ta

te s

Co rn

ke S La

N

or th

ea st

Figure 6.4. Regional changes in selected water quality indicators under full liberalisation scenario

35

25

15 5

-5

-15

-25 % Change

Excess N

Figure 6.4 also illustrates the potential importance of matching agrienvironmental indicators as closely as possible to the factor of interest. Three different approaches to considering changes in nitrogen are considered: excess nitrogen applied to fields, which is similar to the nitrogen

242 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

balance indicator incorporated in the GTAP analysis; nitrogen run-off to surface water, which estimates differences in the amount of excess nitrogen likely to reach a water body (by accounting for factors such as slope and distance to waterways); and, nitrogen leaching to groundwater (which accounts for soil types in predicting leaching potential). Conceptually, the nitrogen risk indicator highlighted in the Canadian analysis, falls between the nitrogen balance and nitrogen run-off indicators. Notably, no consistent relationship among the percentage change in indicators is evident. Changes in indicators may be generally bigger or smaller in one region than another, but which of the indicators experiences the largest shifts is highly variable. For example, the change in nitrogen run-off is larger than in nitrogen leaching in some regions, while the opposite holds true in other regions. While regional shifts in the levels of agri-environmental indicators can be important in policy evaluation, assessing whether or not differences in regional changes in various indicators are really important in a policy context requires additional information on the value of improving (or decreasing) environmental quality in one region relative to that in another. For example, even a small percentage change in drinking water quality in a heavily populated region may be significantly more valuable than a relatively large change in drinking water quality in a region where few individuals would be exposed.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 243

Annex 6.A. The Applied General Equilibrium Trade Framework

In the GTAP model, each regional economy consists of several economic agents: on the final demand side of the model, a utilitymaximising household purchases commodities and saves part of its income, which consists of returns to primary factors (land, labour and capital) and net tax collections. On the supply side of the model, cost-minimising producers employ primary factor services and intermediate inputs to supply commodities. In the model, intermediate (and final demand) users of commodities are assumed to differentiate a commodity by its region of origin (i.e. the Armington specification is applied) (Armington, 1969). Labour and capital are perfectly mobile across sectors in the same region; land is employed in agriculture and is imperfectly mobile across agricultural sectors. Integrated into this treatment of supply, demand and trade is a set of domestic support and trade policies, which are modelled as ad valorem equivalents. These policies affect the market equilibrium computed by the model and, when they change, induce changes in producer and consumer behaviour in all regions. The GTAP model is solved using the GEMPACK suite of software (Harrison and Pearson, 2002).

Data and parameters All economic data, including that on agricultural trade and protection, are taken from the GTAP data, version 6.0 (Dimaranan and McDougall, 2004). In the GTAP data, all PSE components, excluding market price support, were classified into one of four domestic support categories: output subsidies, intermediate input subsidies, and payments to land and capital.6 The GTAP data also account for food and agriculture import tariffs and export subsidies. TRQs are not explicitly modelled; they are instead accounted for by price gaps which reflect the direct price impacts of TRQs.7 This GTAP database contains a number of substantial improvements compared with earlier versions, some of which are fundamental to the present analysis. First, macroeconomic data (i.e. GDP components) and bilateral trade data reflect 2001 conditions. Second, the database includes 2001 domestic support data for several OECD member economies (i.e. Australia, Canada, the EU, Hungary, Iceland, Japan, Korea,

244 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Mexico, Norway, Poland, the Slovak Republic, Switzerland, Turkey and the United States), as well as non-OECD members (i.e. Bulgaria, Romania, Slovenia, Estonia, Latvia, Lithuania and Russia). Third, 2001 tariff data from the Market Access Maps (MAcMaps) database are incorporated in the GTAP database (Bouët, et al., 2002).8,9 Fourth, the database includes agricultural export subsidy data for 2000, as notified to the WTO, for Switzerland, the Czech Republic, the EU, Hungary, Israel, Korea, Norway, Poland, the Slovak Republic, Turkey and the United States. Regarding domestic production quotas, this analysis explicitly models production quotas for dairy products in Canada, and for dairy products and sugar in the EU15 [see van Tongeren (2002); Jensen and Nielsen (2004); and Frandsen, et al. (2001)]. It is assumed that the GTAP data capture the presence of output quotas, and that quota rents are part of the difference between producer prices and market prices. To separate quota rents from producer support, the analysis relies on estimates of dairy quota rent rates taken from Jensen and Nielsen (2004, Table 2); estimates of sugar quota rent rates are taken from Frandsen, et al. (2001, Table 3).10 The partial liberalisation scenario includes a 20% expansion of dairy production quotas in Canada, and a 20% expansion of dairy and sugar production quotas in the EU15 regions. Dairy and sugar production quotas are completely removed in the total liberalisation scenario. The Armington trade elasticity estimates are based on recent econometric work reported in Hertel, et al. (2003). For this analysis, the GTAP trade elasticities for rice, sugar and raw milk have been reduced because liberalisation is not expected to affect international trade in these commodities. The OECD’s Agri-environmental Indicators database (OECD, 2001a; 2003b; and 2003g) is the main source of agri-environmental data, particularly on pesticide use and nitrogen uptake and offload. The data on emissions of GHGs from crop output come from the United Nations Framework Convention on Climate Change (UNFCCC, 2003).

Sectoral and regional specification Ideally, the policy analysis should cover most individual OECD countries as well as the major non-OECD arable crop exporters. However, practical considerations, including manageability of the model, impose some limits in this respect. Table 6.A1 specifies the sectoral and regional scope of the model.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 245

Table 6.A1. Sectoral and regional specification of analysis Sector Farming Rice Wheat Other grains Fruits, vegetables and nuts Oil crops Sugar cane, sugarbeet Plant-based fibres Other crops Cattle, sheep, goats, horses Other animal products Raw milk Wool, silk-worm cocoons

Acronym

RoAs

RoAm DFS UK&I

Country/Region Australia New Zealand China Japan Korea Rest of Asia Canada United States Mexico Brazil Argentina Rest of the Americas Denmark, Finland and Sweden UK and Ireland France Germany

Food processing Red meat products Other meat products Vegetable oils and fats

ABNL

Austria, Belgium, the Netherlands and Luxembourg

Dairy products Processed rice Sugar manufacturing Other processed foods Beverages and tobacco products

PSIG

Portugal, Spain, Italy and Greece

EFTA

Switzerland, Iceland, Norway and Lichtenstein

EU10

10 New EU member states (i.e. the Czech Republic, Hungary, Poland, Slovakia, Slovak Republic, Estonia, Latvia, Lithuania Malta and Cyprus)

RoE

Rest of Europe (i.e., Albania, Bulgaria, Herzegovina, Romania, Russia and the rest of the former Soviet Republics)

TMMENA

Turkey, Morocco and the rest of Middle East and North Africa

RoAf RoW

Rest of Africa Rest of World

Other manufacturing and services Forestry Fishing Minerals Textiles, clothing, footwear Petroleum, coal products Chemical, rubber and plastic products Other manufacturing Wholesale, retail trade Transportation services Other services

246 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

There are thirty traded sectors in the model. All the food and agriculture sectors that are available in the GTAP database have been identified in this analysis: there are twelve aggregate farm sectors (eight of which represent crops) and eight sectors representing processed foods. The sector for “oil crops” includes soybeans, ground-nuts, rape or colza seeds, sunflower seeds, sesame seeds, mustard seeds, safflower seeds, cotton seeds, copra, linseed, palm nuts and kernels, castor oil seeds, poppy seeds, and other oilseeds and oleaginous fruits. The sector for “other crops” includes un-manufactured tobacco, coffee, tea, cocoa, spices, cut flowers, trees, shrubs and bushes. It is necessary to identify all the food and agriculture sectors available because, even though the study focuses on arable crops, the simulated policy change concerns the whole of food and agriculture. The rest of the economy is represented by ten aggregate sectors. The GTAP sector for “chemical, rubber, and plastic products” includes fertilisers, pesticides and other agricultural chemicals, and has been identified in the model. This analysis discusses the implications for four arable crop sectors: rice, wheat, other grains, and oil crops.11

Input substitution The standard GTAP model allows for substitution in production among primary factors (i.e. land, labour and capital). However, intermediate inputs are used in fixed proportions in producing the various outputs, i.e. the Leontief assumption is applied for intermediate inputs. For some analyses, the latter assumption is restrictive, as farmer input-output decisions usually depend on relative prices. In this analysis, the assumption is relaxed to better model the impacts of domestic support reduction, and substitutability between purchased-farm inputs for crop and livestock production is allowed (see Dimaranan, Hertel and Keeney, 2003). The parameters are taken from the literature, particularly from the OECD PEM model (OECD, 2001e). The output function is specified as a nested constant-elasticity-ofsubstitution (CES) function. Specifically, it is assumed that in one sub-process, capital is a factor to substitute for labour; in another sub-process, agri-chemicals (i.e. fertilisers and pesticides) substitute for land (Hayami and Ruttan, 1970; Kislev and Peterson, 1982). Figure 6.A1 specifies the nested CES output functions for the crop and livestock sectors. The specification in Figure 6.A1 is based on a review of the literature. There is no substitution between capital and skilled labour, i.e. WKH GLUHFW HODVWLFLW\ RI VXEVWLWXWLRQ 1 is zero; but there are substitution possibilities between the capital-skilled labour composite and unskilled labour, i.e. WKH GLUHFW HODVWLFLW\ RI VXEVWLWXWLRQ 2 is larger than zero. This specification reflects the idea that each tractor requires a single operator;

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 247

however, if the relative wage for unskilled labour declines, farmers may decide to substitute unskilled labour for certain normally mechanised tasks. In the specification in Figure 6.A1, chemicals substitute for land in crop production, i.e. WKHGLUHFWHODVWLFLW\RIVXEVWLWXWLRQ 3 is larger than zero. In livestock production, feed (either produced on the farm or purchased) substitutes for land, i.e. grazing. Finally, there is limited substitutability between augmented capital and augmented land, i.e. the direct elasticity of VXEVWLWXWLRQ 4 is larger than zero. All other intermediate inputs are used in fixed proportions, i.e. WKHGLUHFWHODVWLFLW\RIVXEVWLWXWLRQ 5 is zero. Figure 6.A1. Specification of crop and livestock output: nested CES function Sector j output 5=0 4

2 3 1=0

Capital

Skilled labour

Unskilled labour

Land

Crop chemicals or livestock feeds

Other inputs

$UHYLHZRIWKHOLWHUDWXUHSURYLGHVYDOXHVIRUWKHGLUHFWHODVWLFLWLHV 2, 3 DQG 4.127KHHODVWLFLW\ 2 is 0.2, a value which is within the range of values given in OECD (2001e, Table A1.4, column Land & purchased). The HODVWLFLW\ 3 is IRU WKH FURS VHFWRUV LQ WKH PRGHO 3 is 0.4 for livestock VHFWRUV LQ WKH PRGHO 9DOXHV IRU 4 are either taken from OECD (2001e, Table A1.4, column Land & farm owned), RU 4 is 0.1. The Allen partial elasticities of substitution in crop and livestock production, i.e. the overall, output-constant elasticities of substitution among inputs, are a function of all the direct elasticities of substitution in Figure A.6.1 and the input cost shares.13

248 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Notes 1.

A consequence of the sectoral specification of the GTAP data and of this analysis is that all identified regions produce some of all commodities.

2.

In this chapter, the term “intensification” means that the rate of input use per hectare (e.g. kilograms of chemicals per hectare) increases; the term “extensification” means that the rate of input use per hectare declines.

3.

Individual crop impacts are determined by two factors. First, total agricultural output and input use would contract in the regions where agricultural support and protection are reduced. Second, in several regions in the model, a significant portion of support is in the form of land and capital subsidies. Liberalisation would make land and capital more expensive to farmers, and thus cause an increase in demand for labour (a capital substitute) and for chemicals (a land substitute). Thus, the ratio of chemicals to land may increase because land use may decline by more than chemical use. The relative size of the scale and substitution effects depends on the degree of liberalisation (i.e. how the relative prices of land and capital are changing), the size of the relevant elasticities (in Figure 6.A1), and the relevant input cost shares.

4.

Similar percentage impacts across countries/regions do not imply similar absolute impacts because they refer to different magnitudes.

5.

The GHG data made available by the UNFCCC (2003) do not cover all regions in this study.

6.

Even though all economic policies in GTAP are modelled as ad valorem price gaps, decoupled agricultural support is applied to factors of production that are generally used only in agriculture, e.g. land, and thus the impact of decoupled support on output is reduced.

7.

TRQs would be relaxed under liberalisation. Analysis of TRQs can rely either on explicit modelling of all the policy instruments involved (e.g. in-quota tariff rate, trigger quantity, and over-quota tariff rate) or on price gaps which summarise the direct price impact of the TRQ. An example of the former approach is Elbehri and Pearson (2000); an example of the latter is USITC (2002). The former approach is necessary when options about specific policy instruments are evaluated (e.g. a reduction in an over-quota tariff rate v. an expansion of the quota). However, the analytical benefits of explicitly modelling TRQs are reduced because of data limitations. For example, dairy TRQs represent a complex system of import restraints which apply to imports of several milk products, including fluid milk, cream, butter, cheese, powdered milk products, ice cream, infant formula, and animal feeds containing milk; while casein, which represents a large and growing share of imports, may be imported free of duty; and the model sector under AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 249

consideration is an aggregate of all of those industries. Thus, the latter approach is satisfactory for broad, economy-wide assessments of policy reform. 8.

The border measures in the GTAP database are intended to quantify applied tariffs as opposed to bound tariffs.

9.

The simulation results reported in this chapter are based on the assumption that intra-NAFTA and intra-EU25 agricultural trade has been completely liberalised by the time further multilateral trade liberalisation takes place, i.e. the GTAP database has been accordingly revised via a simulation with the standard GTAP model.

10.

Jensen and Nielsen (2004) report estimates of dairy quota rents for EU15 member states. In this analysis, the average EU dairy quota rent rate is applied to Canada.

11.

Despite the effort to identify as many regions and sectors as possible in the model, significant aggregation biases remain. For example, countries have been grouped with other dissimilar countries, and the GTAP sector "chemical, rubber, and plastic products" includes very dissimilar commodities. Thus, it is not easy to translate simulated impacts for an aggregate region or sector to impacts for their components.

12.

Brandão, Hertel and Campos (1992); Dimaranan, Hertel and Keeney (2003); Hertel and Tsigas (1987); and Salhofer (2000).

13.

Despite the effort to make more the modelling of agricultural production more realistic, there are some issues which have not been dealt with: water issues (which affect crop supply response) and crop rotation issues (which affect soil quality and biodiversity).

250 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Annex 6.B. Regional Models

6.B.1.

The US Regional Agricultural Programming Model (USMP)

The USMP model uses a positive math programming approach (Howitt, 1995) with nested constant elasticity of transformation function to allow non-linear substitution across the production activities (House, et al., 1999). The agriculture sector is assumed to be a spatially competitive market equilibrium system, but partial equilibrium in the sense that it does not compete with other sectors (e.g., manufacturing) for factors of production (e.g. land or labour). The supply side of the system is aggregated into production units specified for large geographic areas delineated by erosion class (HEL and non-HEL). Twenty-three inputs are included, as are production and consumption of 44 agricultural commodities and processed products, including major crop (corn, soybeans, sorghum, oats, barley, wheat, cotton, rice, hay, silage) and livestock (beef, dairy, pigs and poultry) enterprises. This model comprises approximately 75% of US agronomic production, and more than 90% of US livestock production (USDA, NASS, 1997). Production levels, land use, land-use management (e.g. crop mix, rotations, tillage, and fertiliser practices), and government programme participation are endogenously determined spatially according to a constrained optimisation approach, maximising consumer and producer welfare. For this analysis, environmental impacts related to nitrogen, phosphorus and pesticide loss to water, sheet, rill, and wind-related soil erosion, carbon emissions, and manure nutrient production are examined. With the exception of manure nutrient production1 and carbon emissions,2 the environmental parameters are estimated using EPIC, the Environmental Policy Integrated Climate Model (Mitchell, et al., 1998). For each crop production activity, EPIC simulates erosion (sheet, rill and wind), nutrient cycling and transport of pesticide as a function of crop management (rotation, tillage, and fertiliser rates) given historic weather, hydrology, soil temperature, and typography data. The estimates of field-level discharge used here represent mean values for a 67-year time horizon. The transport of nutrients, pesticides and sediment across the landscape is then calibrated to US Geologic Survey estimates of regional pollutant loads (Smith, Schwartz and Alexander, 1997) to derive in-stream measure of nutrient and sediment loading. Estimates are reported at the USDA Farm Production Region level. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 251

A detailed description of the model can be found in Cooper, Johansson and Peters (2003). Changes in commodity production levels in USMP are fixed at the levels predicted by GTAP at the national level. The model then solves for commodity prices, regional production allocations, associated input uses and agri-environmental outcomes. USMP models many more agricultural commodities than GTAP. In cases where GTAP categories did not match USMP commodity categories, production shifts were assumed to be the same for all commodities encompassed in the GTAP grouping. USMP is formulated so that it can replicate commodity prices, supply, use, and acreage for any given baseline year. The USDA baseline provides information on prices, production, acreage planted, yield and various government programmes by commodity at the national level. Information about the distribution of agricultural activity among USMP regions comes from the NRI and the Agriculture Census. The distribution of crop yields by region comes from county data, while estimates of crop yield production activity come from EPIC simulations. Estimates of erosion at national and regional levels come from the NRI, while estimates of erosion by production activity also come from EPIC simulations. Estimates of cost of production nationally, and by commodity-specific region come from the USDA baseline and Agricultural Resources Management Survey, while cost of production estimates for individual production activities with respect to crops comes from the USDA baseline for 2001 (USDA, 2003e).

Generation of USMP environmental indicators Environmental indicators contained in USMP production activities were generated using the EPIC model. EPIC is a crop biophysical simulation model that is used to estimate the impact of management practices on crop yields, soil quality and various environmental emissions at the field level. It utilises information on soils, weather and management practices (including specific fertiliser rates), and produces information on crop yields, erosion, and chemical losses (including nitrogen losses) to the environment. Management practices used in the EPIC management files were set consistent with agronomic practices for the 45 regions (see Figure 6.2). Crop yields and the array of environmental indicators associated with each crop-production activity represented in USMP were generated by running EPIC in four sequential steps. The first conditions the soil, while the next three are used to calculate short-term yield, average rate of emission, and long-term yield. The first or conditioning step allows EPIC to rectify any inconsistencies in the soil profile imported from SOILS5 database. It involves running EPIC

252 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

for a period of 5 years while keeping its soil erosion module turned off. It has the added advantage of making the soils profile at the beginning of the second step consistent with a field that has been subjected to the management practices being simulated. In step two, the short-term yield was calculated. EPIC is run for a period of 7 years, again with its soil erosion module turned off and short-term yield calculated as average yield during this period. By turning erosion off, any variation in yield will be due solely to variation in weather. In step three, the environmental indicators were calculated by running EPIC for 60 years, this time with soil erosion turned on. Total emissions for each indicator are tabulated and divided by the length of the simulation to obtain the annual rate of emission. Running the systems for 60 years does two things: it eliminates the dependence of the emissions from the sequence of weather for any particular period, and it provides a consistent base for making comparisons between systems. Therefore, all systems are run through two full weather cycles. At the same time, each management regime is run through at least 5 full management cycles. In step four, the long-term yields were calculated. EPIC is run as described in step two, the only difference between them being that EPIC uses the soil profile generated from the previous 60 year simulations. The difference between short- and long-run yield represents the change in soil productivity caused by the management system.

Description of indicators Nitrogen indicators The potential impacts of agricultural production changes on nutrients lost to the environment were categorised using several indicators. Excess nitrogen balance is essentially a nitrogen balancing exercise. Nitrogen sources include chemical fertilisers, manure fertiliser, and nitrogen fixed by legumes. From this, the nitrogen harvested in crop yield is subtracted, which leaves excess nitrogen left on the field which may leach or run-off into surrounding waters. Nitrogen loading is the amount of nitrogen in subsurface flow, nitrogen in solution, and nitrogen attached to sediment that is estimated to arrive in surrounding streams, rivers, and lakes. This is calibrated using baseline USGS forecasts of nitrogen delivery from agricultural sources (Smith, Schwartz and Alexander, 1997).

Erosion indicators Several indicators of erosion were included for comparison. Risk of sediment loss to water is represented in EPIC according to the standard AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 253

Universal Soil Loss Equation (USLE). The USMP representative soils choice factors include slope length and soil erodibility. The hydrologic group reflects the course fragment factor. The crop management factor is determined by the USMP crop system, and rainfall parameters are incorporated in the EPIC simulation for each of the 45 regions. Thus, all USLE variables are accounted for, with the exception of the erosion control factor. The absence of this factor could result in low erosion estimates where physical controls such as terracing are needed but have not been incorporated. Over the years much of this type of land has either been taken out of tilled crop production or has undergone treatment. Sediment loading applies a delivery ratio determined in calibrating phosphorus loading to USGS forecasts to the edge of field delivery of sediment estimated by EPIC. Wind erosion is determined by adjusting the wind erosion equation to account for crop residues and rotations at a daily level. Both the wind erosion and risk of sediment loss to water parameters are calibrated to the levels from the 1992 NRI, which have been aggregated to the USMP regions.

Pesticide indicators Indicators have been developed for leaching and run-off. For each cropping enterprise in USMP, the amount of active ingredient for the predominant pesticides applied has been calculated in pounds. For each pesticide, the active ingredient is then converted to "toxicity persistence units" (TPUs) (Barnard, et al., 1997). These refer to the sum of reference doses (maximum daily human exposure resulting in no appreciable risk) of the pesticides used for a particular cropping enterprise, multiplied by the number of days each of those pesticides remains active in the environment. As a point of reference, the number of TPUs in a pound of DDT is 4 443 million and in a pound of Borax it is 103 872. Edge of field estimates for pesticides in sub-surface flow, in solution, and attached to sediment are multiplied by a transfer coefficient equal to the sediment delivery ratio (above) to generate an estimate of the amount of pesticide toxicity that is discharged into lakes, streams and rivers.

Carbon emissions Carbon emissions are calculated according to the Intergovernmental Panel on Climate Change estimates (IPCC, 1996). These are applied to the USMP cropping enterprises. The values indicate the amount of carbon emitted when converting land from native pasture. Negative values indicate lower carbon emissions.

254 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

6.B.2.

The Canadian Regional Agricultural Model (CRAM)

The CRAM is a sector equilibrium model for Canadian agriculture which is disaggregated across both commodities and space (Horner, et al., 1992). CRAM is static, non-linear optimisation model that maximises producer plus consumer surplus. The basic commodity coverage is for grains and oilseeds, forage, beef, pigs, dairy and poultry (horticulture is excluded). Spatial features of the model include provincial level livestock and crop production, with the exception of the Prairie Provinces, where crop production is divided into 22 regions based on the Census of Agriculture boundaries. Supply response is determined by the relative profitability of alternative crops. It allows for both inter-provincial and international trade in primary and processed products. One of its most important features is that it takes into account the interdependency of crop and livestock output. It should be pointed out that, when using the CRAM model in conjunction with AEIs it is difficult, – sometimes impossible – to account for potential spatial differences. CRAM is a regional model disaggregated by provinces with the exception of the crop component for the Prairies, while most of the AEIs are based on Soil Landscape of Canada poygons (SLC) that are much smaller than the CRAM regions. Better analytical tools are needed to more realistically apply the production changes generated by CRAM to the SLC level for improved estimation of the impacts on AEIs. The impacts of the GTAP analysis were imposed on CRAM for partial and total trade liberalisation. For this exercise, only the output quantity and land-use changes from GTAP have been imposed and any price changes have been ignored. Assumptions were made about productivity increases for various crops.

Adjustments to CRAM for partial removal of tariffs and subsidies To impose the quantity changes from GTAP, the crop yields for wheat were increased by 4.6% and for durum by 4.8% to achieve the 5.19% increase in wheat production. For each 1% increase in the yields, fertiliser costs were increased by 2%. Given the preliminary nature of the analysis and the time constraints, this simple rule was imposed to at least acknowledge the fact that to achieve a yield increase, input use has to change. To impose the land-use change from GTAP, wheat areas were increased by 0.52%. Other grains included barley, oats and corn grain. Other grain yields and areas were increased by 0.25% and 0.30%, respectively, to achieve a 0.55% increase in production of other grains. Rapeseed and soybean areas were decreased by 4%. Beef production was increased by 2.58%. Pig production was increased by 0.6%, whereas dairy production was decreased by 1.69%. The livestock changes were imposed through AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 255

adjusting animal numbers. Although poultry is included in “other livestock”, for the preliminary analysis, it was not changed in CRAM as there was some uncertainty as to how the production quota was dealt with in GTAP.

Adjustments to CRAM for total removal of tariffs and subsidies Crop yields for wheat and durum were increased by 15% to achieve the 10.80% increase in wheat production. Wheat and durum areas were decreased by 3.61%. For each 1% increase in the yields, fertiliser use was increased by 2%. Other grains included barley, oats and corn grain. Other grain yields were increased by 6% and crop areas were decreased by 1.18% to achieve a 4.65% increase in production of other grains. Rapeseed and soybean areas were decreased by 8.78 % and yields were increased by 7.6% to achieve a 1.80% decrease in oilseed production. Beef production was increased by 11.09%. Pig production was increased by 20.42%, whereas dairy production was decreased by 5.22%. Poultry production was not changed. The Canadian Economic and Emissions Model for Agriculture (CEEMA) is used to assess GHG emissions (Kulshreshtha, et al., 2002). CEEMA consists of a GHG module, which links CRAM output to GHG coefficients to estimate emissions of CO2, CH4 and N2O. Total emissions are calculated on a CO2-equivalent basis based on 100-year global warming potential conversion factors of 21 for CH4 and 310 for N2O. The model uses current state-of-the-science data of GHG emissions to estimate total emissions from primary agriculture. In order to provide a complete indicator for agricultural systems, the calculations include all of the IPCC sources for the agriculture sector, plus sinks from the land use, land-use change and forestry sector and on farm-fuel use from the transportation sector. A systems approach is taken, meaning both direct and indirect emissions are accounted for all three major GHGs.

Residual Nitrogen The Residual Nitrogen Indicator (RSN) is an estimate of the quantity of nitrogen remaining in the field after harvesting (McRae, et al., 2000). It is the difference between the amount of nitrogen that is available to the growing crop from all sources and the maximum amount removed in the harvested portion of the crop under average conditions. The crop nitrogen requirement is estimated as the amount recommended for achieving economically optimal production. The RSN is calculated by:

x

Estimating the amount of nitrogen available from the three major agricultural sources of nitrogen: mineral fertiliser, animal manure,

256 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

and legume nitrogen fixation. In semi-arid regions, inputs also include crop residues and mineralisation of soil nitrogen during periods of summer fallow.

x

Estimating the amount of nitrogen removed in the harvested portion of the crop, based on a combination of recommended levels and standard tables of the portion removed in harvest.

x

Calculating the difference between these two amounts to give a value for residual nitrogen.

Nitrogen levels were determined from recommended rates of fertiliser application rather than from crop yields, to reflect the actual situation in which farmers must decide at an early stage of crop growth how much nitrogen to apply. Crop yield is only partly controlled by management inputs; uncontrollable growing season conditions exert a major influence. Where the levels of available nitrogen are less than, or equal to, crop recommendations, the ratio of nitrogen remaining to nitrogen available corresponds to standard published information and reflects the overall ability of the crop to use nitrogen. Where nitrogen is present in excess, the ratio increases. The indicator itself does not give any insight into the environmental effects of various levels of residual nitrogen in different agricultural settings. Surplus nitrogen may pose a risk to the environment, but this risk is also sensitive to other factors, such as soil type and climatic conditions. For example, the movement of nitrogen from farmland into the broader environment is related to the movement of water. In the dry regions of the interior of British Columbia and the Prairies, the movement of nitrogen in water is limited, occurring mainly during storms and periods of heavy runoff. The environmental risks of having residual nitrogen in the soil are greater in humid areas of the country, such as central and eastern Canada. Thus, residual nitrogen was also used in the assessment of the next indicator – Risk of Water Contamination by Nitrogen.

Risk of Water Contamination by Nitrogen The IROWCN is based on estimates of the potential concentration of nitrate-nitrogen in water leaving farmland. The potential concentration of nitrogen in water leaving farmland is determined by dividing the amount of nitrogen by the amount of water available to dilute this nitrogen (called excess water). The quantity of nitrogen that is potentially available to move off farmland, called residual nitrogen, was calculated as described above for the RSN. The amount of water that is potentially available to move off farmland was calculated by devising a moisture budget based on 30-year averages for AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 257

precipitation (moisture input) and potential evapotranspiration (moisture output). The difference between these two values was used as the estimate of water surplus or water deficit. The capacity of the soil to hold available water was also an important factor in the water budget.

Habitat availability Agriculture has reduced the quantity of natural habitats, mainly through conversion of the natural landscape and changes in land use, such as the drainage of wetlands and the removal and fragmentation of forest cover. It can also affect the quality of wildlife habitats through various land management practices, such as fertilisation, pesticide use and intensive grazing. However, some wildlife species are able to thrive where a native habitat has been replaced by an agricultural habitat, or where agricultural lands contain such habitats as wetlands, grasslands and wooded areas. To assess agriculture’s impacts on wildlife habit, matrices were developed that relate habitat types found on agricultural land (e.g. cropland, pasture, woodlands, wetlands) to the ways in which individual species use agricultural habitats (e.g. for foraging, feeding, nesting, breeding). The matrices were developed from the literature and expert opinion, and data on area of agricultural habitats were obtained from the Census of Agriculture. The indicator model used in this analysis can be interpreted as the level of habitat availability on agricultural lands (adapted from McRae, et al., 2000).

258 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Notes 1.

Manure nutrient parameters are developed from Kellogg, et al. (2000).

2.

Carbon emissions are calculated according to the Intergovernmental Panel on Climate Change estimates (IPCC, 1996). The values indicate the amount of carbon emitted when converting land from native pasture.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 259

Annex 6.C. Selected Data Table 6.C1. Gross nitrogen balances and trade liberalisation impacts Uptake coefficients

Region

Australia New Zealand Japan Korea Canada US Mexico DFS UK&I France Germany ABNL PSIG EFTA EU10

Arable crops average ---------------------- KG/t ----------------------22 20 19 39 21 21 18 35 21 21 25 19 68 21 18 22 20 50 19 31 28 42 34 10 20 17 38 23 14 24 19 40 19 19 17 35 19 19 17 31 20 25 27 20 38 26 25 21 50 27 21 17 35 25 22 25 22 39 30 21 16 41 20 26 24 22 41 25

Rice

Wheat

Other grains

Oil crops

Simulation impacts Partial trade liberalisation Full trade liberalisation Uptake

Chemical Balance use

Uptake

Chemical Balance use

--------------------------------- % change ---------------------------------0 0 0 -1 -1 0 2 7 5 3 19 16 -2 -4 -2 -11 -27 -16 0 -1 0 -2 -4 -2 0 1 1 0 2 2 -1 -1 0 -4 -5 -1 0 -1 -1 -3 -10 -7 -1 -4 -3 0 -6 -6 0 -4 -3 1 -5 -6 -1 -4 -3 -2 -9 -7 -1 -4 -3 -1 -8 -7 -3 -33 -29 -6 -60 -55 -1 -5 -4 -5 -18 -13 -3 -9 -6 -5 -14 -9 0 1 0 1 2 1

Source: OECD (2003b), Eurostat, and OECD Secretariat calculations.

260 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Figure 6.C1. Arable crop output: partial trade liberalisation impacts, comparison of standard and revised GTAP models

30 20

(% )

10 0 -10 -20

Standard GTAP model

R oAf

R oW

R oE

TM M E N A

E FTA

1 0 EU a cc

P S IG

ABNL

Fra n c e

G e rm a n y

D FS

UK&I

R oAm

B ra z il

A rg e n tin a

M e x ic o

USA

RoAs

C anada

K o re a

Jap a n

C h in a

N e w Z e a la n d

A u s tra lia

-30

Revised GTAP model

Note: For the definition of regional groupings see Table A.6.1.

Figure 6.C2. Arable crop land use: partial trade liberalisation impacts, comparison of standard and revised GTAP models 30 20

(% )

10 0 -10 -20

Standard GTAP model

RoW

R oAf

TM M E N A

RoE

1 0 E U a cc

E FTA

P SIG

ABNL

Ge rm an y

Fra nce

UK&I

D FS

RoAm

A rg en tin a

B ra zil

M e xico

U SA

Canada

R o As

K o re a

Ja p a n

C h in a

N e w Ze a la n d

A u stralia

-30

Revised GTAP model

Note: For the definition of regional groupings see Table A.6.1.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 261

Figure 6.C3. Arable crop chemical use: partial trade liberalisation impacts, comparison of standard and revised GTAP models 40 30 20

(% )

10 0 -10 -20 -30

Standard GTAP model

Revised GTAP model

Note: For the definition of regional groupings see Table A.6.1.

262 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

R oAf

R oW

TM M E N A

R oE

1 0EU acc

E FTA

P S IG

ABN L

Fra n c e

G e rm a n y

U K&I

D FS

R oAm

A rg e n tin a

B ra z il

M e xic o

USA

RoAs

Canada

K o re a

C h in a

Jap a n

N e w Ze a la n d

A u s tra lia

-40

Chapter 7 AN ANALYSIS OF THE TRADE EFFECTS OF AGRI-ENVIRONMENTAL PAYMENTS AND REGULATIONS ON ARABLE CROPS 7.1.

Introduction The new and more comprehensive agri-environmental programmes discussed in Chapter 4 raise questions about the possible negative effects on trade, including arable crop imports and exports. It is theoretically sound and permissible under current trade rules to pursue such agri-environmental policy interventions (Anderson, 1992; Ervin, 1999). Indeed, correcting for missing environmental markets (i.e. positive and negative externalities and public goods) or reducing government policy distortions improves social welfare. However, the choice of approach adopted for making the corrections and reducing the distortions is critical. Agri-environmental programmes, if not designed and implemented in cost-effective ways, may lead to losses in national and/or global welfare. Hence, an examination of their effects on production and trade, and of ways to improve their efficacy is warranted. Crop production, trade, environmental impacts and environmental policy are linked in mutually dependent fashion. That is, new crop production and transport patterns resulting from trade liberalisation may cause environmental changes that in turn induce agri-environmental policy responses. For example, following liberalisation, a shift to a new crop mix requiring higher fertiliser or pesticide use may degrade water conditions and prompt environmental policy reforms to address the new pressures. Subsequently, the new agri-environmental policies may affect production patterns and trade flows, which redefine the supply pressures influencing future agri-environmental policy. In addition, changes in agri-environmental policies that are not in response to trade concerns may have production, and hence, trade impacts. This chapter focuses primarily on the impacts of agrienvironmental policy on trade. It is clear that agricultural trade and environmental linkages are interwoven and dynamic. The main objectives of this chapter are: to conduct a cross-country analysis of 1) the trade effects of agri-environmental payments on the arable sector; 2) the extent to which environmental regulations affect the factor costs for arable crop producers and trade; and 3) to offer some practical suggestions for enhancing the effectiveness of agri-environmental policies AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 263

related to arable crops in achieving their environmental objectives without “distorting”1 trade flows. The first section briefly reviews the major agrienvironmental policies related to arable crops. Relevant theory and empirical analyses of the trade impacts of agri-environmental programmes are reviewed in order to lay the groundwork for assessments of the probable trade and factor cost effects. A discussion of ways to improve the effectiveness of agri-environmental programmes concludes the chapter. The analysis primarily draws on the scientific literature to assess the trade, cost and policy efficiency issues. Limited empirical analysis is conducted to illustrate the potential trade effects under selected agri-environmental regulatory cost and payments for selected arable crops in selected OECD countries.

7.2. Overview of agri-environmental policies for arable crop agriculture 7.2.1.

Payment programmes

The degree to which arable crop production and trade flows may be affected, and potentially distorted, depends on three characteristics of the payment programmes. The first is the total amount of the payments made under the programme, and therefore the potential effect on crop supplies. The second is whether eligibility for payment is linked to the production of specific arable crops. The third is whether the amount of payment varies directly with the hectarage planted or production level of the crop. The two latter charateristics are major factors in determining a programme’s degree of “decoupling” from production. It is often assumed that agri-environmental payments should be fully decoupled from the farmer’s decision about planting a particular crop to avoid potential production and trade distortions. Such “decoupled” payments differ from schemes used to foster the adoption of environmentally friendly technologies, such as delayed mowing for protection of biodiversity, which may or may not affect production levels. Hodge (2000) argues that the assumed desirability of decoupled payments reflects an “input model” of agri-environmental impacts in which crop production and the environmental service(s) are competitive across the relevant range of production levels. For example, fertiliser applications beyond plant utilisation levels will leach or run-off into ground and surface waters and degrade quality for human use and natural system functioning. However, some agri-environmental services may complement certain types of agricultural production over specified ranges. That is, if the desired environmental service is tightly coupled or “joint” with the production of certain arable crops, then the agri-environmental programme payment may

264 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

unavoidably affect production and trade. In these complementary situations, it may be theoretically sound and cost-effective to pay farmers to increase the level of certain forms of crop production and, simultenoulsy, to increase environmental service supply over a limited range (Hodge, 2000). Hodge discusses the cost-effectiveness of using an output-based approach for making payments, which depends to a large extent on the relative transaction costs of implementing the scheme (i.e. whether there are other easily observable and less expensive measures on which to base payments).

7.2.2.

Regulatory approaches

Generally speaking, agriculture has traditionally been subject to fewer environmental regulations than other industrial sectors, mainly due to the difficulties and costs of tracing non-point pollution sources. However, regulatory approaches are increasing and have the potential to influence trade in certain markets. Pesticide regulations are the most notable national regulations, whereas water and air quality regulations are more likely to occur at state/provincial and local levels (Brouwer and Ervin, 2002). The cross-compliance concept originated in the United States and has spread to many other OECD countries (Chapter 4). The compliance schemes generally require farmers to implement certain practices. While the ex ante costs of the compliance measures can be calculated using data on technical standards, the actual costs are very difficult to measure because they depend on the farmer’s behavioural responses to the requirements, which are often unobservable and will depend on the degree of enforcement. In the EU, the degree of enforcement has been specified by law through audits. Moreover, the impact of compliance on a farm’s finances will be coupled with other agricultural payments, so that the final effect of the measures may be to lower the net value of the other payments (Chapter 5). Perhaps for these reasons, there have been few empirical estimates of the net costs of imposing compliance requirements on arable crop producers. A recent inventory of environmental regulatory standards related to agriculture was conducted in the EU, the United States, Canada, Australia and New Zealand (Brouwer and Ervin, 2002). The country analyses did not find evidence to suggest that compliance with environmental regulations has been, or will become, a driving force in the location of crop production (Brouwer and Ervin, 2002). However, the same conclusion could not be drawn for livestock, on which environmental regulations impose higher costs than for crop production. Moreover, Ribaud, et al., (2003) suggest that indirect effects of environmental regulations related to animal manure and slurry may lead to changes in the extent and location of the production of certain crops.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 265

Another important lesson from the multi-country analysis is that it is insufficient to analyse the impacts of agri-environmental regulations at national level alone. Local, state/provincial and regional governments are, to an increasing extent, implementing environmental regulations on agricultural producers through the use of compulsory standards. For example, the state of California has implemented new regulations to control cropland run-off pollution (ENS, 2004). Prior to 2004, Californian farmers enjoyed an exemption from the state’s water quality law (which requires entities to apply for permits to discharge run-off into public waters and submit plans to reduce pollution). However, in order to continue receiving the waiver, they are now required to test and report run-off during irrigation and storms and report the findings to their regional water authority. Although fines are not planned as part of the monitoring effort, the monitoring exercises will involve considerable expense, reported at USD 10 000 for startup and USD 2 000 per year. If monitoring data reveal high levels of pollution, the potential exists for state authorities to require pollution control actions that involve significant additional expense, unless subsidies are provided. The implication of this initiative (and the growing number of other non-federal regulatory actions) is that using only the costs of the national programmes for the analysis will result in an underestimation of the potential production cost and trade effects.

7.2.3.

Other measures

The final category includes advisory and institutional measures, including research and development (R&D), technical assistance and extension, and product information. The R&D, technical assistance and extension approaches have been the foundation of advancing agricultural development in OECD countries. Analyses generally have estimated significant and high rates of return to agricultural research for marketed outputs (Alston and Pardey, 1996). However, the efficacy of agricultural research in supplying non-market environmental services, through either avoiding potential damage, or increasing positive benefits, is less clear. The absence of scarcity values for non-market environmental services may lead to an undersupply of R&D that addresses many public agri-environmental objectives (Ervin, 1998). The lack of incentives may have contributed to the inadequacies of current environmental research related to agriculture (i.e. it is reactive and narrow in scope and scientific content) (Robertson et al., 2003). The effectiveness of technical assistance, extension and product information in furthering the supply of agri-environmental services is similarly unclear. However, with the exception of product information, none of these forces is expected to exert a significant influence on the factor costs and trade flows of arable crops.

266 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

7.3.

Agri-environmental programmes and trade: theory and models A rich literature has evolved on agri-environmental programmes in OECD countries (e.g. Batie and Horan, 2004). However, only very little examines the linkages of trade with agri-environmental conditions and policies. OECD has contributed to the sparse literature on agricultural trade and environmental issues (e.g. OECD, 2000a; 2000b). An even smaller number of studies has attempted to estimate the empirical effects of agrienvironmental policies on trade. Very few have analysed cross border effects, such as the importation of plant species that may degrade domestic environmental resources (Roberts, Josling and Orden, 1999). The management of transboundary environmental relationships differs in that policy approaches require bilateral or multilateral negotiations with significant transaction costs (Pearson, 2000). Nonetheless, the available studies provide a basis for building an analytical framework to estimate trade impacts and infer potential agri-environmental policy reforms. The potential effects of agri-environmental programmes on the arable crops trade are of four types. First, payments made to farmers for agrienvironmental purposes may affect the level or composition of crop production in ways that alter trade flows between countries. Second, the costs and requirements for farmers to comply with compulsory environmental requirements may raise production and other expenses to such an extent that their competitiveness in international markets is diminished. Third, if too severe, the regulatory requirements may induce some agricultural operations to relocate to other countries. Finally, some agri-environmental programmes may impose restrictions on imports to protect domestic natural resources, such as phytosanitary measures to ban the importation of crops that could become invasive species. This study focuses on the first and second categories – the impacts of agrienvironmental payments on trade and the effects of regulations on factor costs and competitiveness. A large number of studies has examined the linkages between environmental policies and trade, but these are mostly concerned with sectors other than agriculture (Pearson, 2000). Many of the investigations focused on “competitiveness” questions (i.e. whether the imposition of environmental regulations raises the costs to domestic firms to such an extent that they become economically disadvantaged in competing for export markets with firms from other countries required to meet less stringent environmental regulations). A recent interpretation of the effects of environmental regulations concludes: To summarize, the studies through 1990 suggest that there had been no significant overall loss of U.S. competitiveness (industrial location) due to AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 267

strong environmental regulations. Some, not all, of the more recent studies are consistent with the competitive loss/industrial relocation hypothesis, but the effects appear small and may be the result of normal industrialization in developing countries, or the result of relatively closed trade regimes and not the result of differential environmental regulations assessment. (Pearson, 2000) Pearson cautions that the prospect of large increases in environmental regulations, such as for the control of greenhouse gas emissions, could alter the basic conclusion of little effect on competitiveness. Although some analyses were hindered by weak data, especially actual compliance costs, the general finding that environmental regulations have no significant trade effect has been fairly robust. However, most of the studies were conducted at aggregate scales that may not reveal significant adjustments by particular sectors in selected countries. Moreover, increased regulation could change that conclusion in the future, but, for now, environmental regulations on non-agricultural industries generally have not significantly affected their trade flows.

7.3.1.

Welfare theory

Anderson (1992) has developed the standard welfare economics of trade and the environment. The basic conclusion for the small country case is “… liberalizing trade in the good whose production is pollutive improves the small country’s environment and welfare if, following the policy change, the country imports the good; but should this good be exported, the environment is worsened and so welfare in this small country may or may not increase in the absence of a pollution tax” (Anderson, 1992). The conclusion for a consumption good is “… liberalizing trade in a good whose consumption is pollutive improves the country’s environment and welfare if the country exports that good, but would worsen the environment and therefore may reduce welfare if the good is imported unless a pollution tax close enough to the optimal rate is in place” (Anderson, 1992). The welfare effect of the large country case is indeterminate and depends on the magnitude of the price and quantity shifts and the differences between the marginal private and social cost curves from the pollution effects. However, the overarching conclusion is clear “… the fundamental point remains that free trade is nationally and globally superior to no trade so long as the optimal pollution tax is in place” (Anderson, 1992). This final conclusion also implies that all types of environmental policies must be “optimal” to ensure that trade liberalisation will lead to national or global net welfare gains.2 The definition of the term “optimal” will be explored in the final section of this chapter.

268 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Models of the relationships between production, trade and the environment generally assume that the inputs used to produce outputs (e.g. land used for wheat production) directly and negatively affect the desired environmental service (e.g. soil erosion and degraded run-off) (Anderson, 1992; Antle and Just, 1991). In this formulation, there is an obvious trade-off between producing more of the output with more inputs and producing more of the environmental service. Analyses of agricultural trade and the environment usually adopt this “input” model to illustrate that imposing an “optimal” pollution tax on output or input reduces supply and places upward price pressure on output. However, as noted above, if the environmental service and output are joint to a significant degree, an “output” model approach is needed (Hodge, 2000). In some cases, the agricultural output and environmental service are negatively linked, which leads to the same conclusion as the "input" model. However, the “output” model more often assumes that certain levels and types of production, such as extensive grazing, provides positively linked production and environmental services. Hodge uses this output model to show that decreases in agricultural support prices will alter the mix of environmental services associated with production, although the outcome may not necessarily be beneficial. This positively linked output model has important implications for the evaluation of agri-environmental programme designs that are consistent with multilateral trading rules. The graphical analyses in Anderson (1992) must be converted into quantitative models of specific markets in order to estimate potential impacts. Larson (2000) uses some basic duality relationships between profit and cost functions to decompose potential supply impacts into components that are easy to understand and estimate. Larson, et al. (2002) and Larson and Scatasta (2003) illustrate how this empirically tractable approach can be used to evaluate the impacts of domestic environmental policies on agricultural exports and imports (see Annex 7.A). The attractiveness of their approach is that the required information is often available, or can be readily estimated, from secondary data. With estimates of the variables in each case, policy makers can analyse the potential impacts of new agri-environmental regulations on their export and import markets. The quality of the estimates depends on the quality of the data for the required parameters and on the capability of the modelling framework to capture the salient features of the system under analysis. However, the above models do not apply directly to agri-environmental payments, which are the largest and fastest-growing form of programme intervention. It may be possible to adapt the basic approach to certain types of payments. For example, if the programme pays farmers to institute less intensive production practices that produce joint outputs of an arable crop AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 269

and certain environmental services, the effect is to lower the average cost of producing the crop. In this case, the average cost model can be reversed to simulate the effects of a decrease in costs on production and trade. The effects of land retirement programmes are more complex and can be conceptualised as expanding the land-use choices confronting farmers with another option for enrolling eligible lands and receiving payment for retirement rather than cropping. This formulation makes it clear that land retirement competes with crop supplies, and cannot be effectively decoupled from production decisions and supply.

7.4.

Effects of agri-environmental programme payments on trade 7.4.1.

Trade and agricultural policy context

OECD countries use a wide variety of programmes that provide payments to farmers with the aim of protective the environment (Chapter 4). Such payments are generally either contingent on altering a land-use practice to provide a positive environmental benefit (e.g. wildlife habitat), or reducing a negative environmental cost (e.g. soil erosion causing off-site water pollution). To avoid trade disputes, agri-environmental payments must conform to the following “Green Box” criteria of the URAA:

x

Eligibility for such payments shall be determined as part of a clearly-defined government environmental or conservation programme and be dependent on the fulfilment of specific conditions under the government programme, including conditions related to production methods or inputs.

x

The amount of payment shall be limited to the extra costs or loss of income involved in complying with the government programme.

These restrictions are quite general and do not ensure that qualifying agri-environmental programmes are intended to effectively address missing markets (e.g. externality or public good situations). If the programmes are “approximately optimal” in the sense of cost-effectively addressing such missing market problems, they can be considered “non-distorting”. Nonetheless, such payments also may change production and trade in arable crops. For example, payments to alter land-use practices, including temporary or permanent retirement of cropland, may affect land allocations (e.g. reduce cropland supply) in such a way that national production generally decreases. If the retirement programme reduces supply sufficiently, world prices will increase and affect consumer and producer welfare in domestic and foreign countries. As such, these agrienvironmental payment programmes cannot be totally decoupled from

270 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

production levels. Since agri-environmental payments related to arable crops are increasing in many OECD countries, larger potential supply effects can be expected. As an example, the United States’ CRP generally satisfies the two Green Box criteria, i.e. explicit conservation and environmental objectives, and payments (i.e. rental rates) equal to the costs of implementing the required practices (i.e. income lost from crop production and annual maintenance to minimise adverse environmental effects, such as noxious weed control) (Smith, 2002). Note that the CRP is decoupled in one respect. Eligibility to participate in the programme does not depend on the land having been in production of a specific arable crop; thus, the crop mix should not be systematically changed from market allocations. The enrolled CRP land is inversely coupled to land productivity (yields). That is, lands with lower productivity are more likely to be retired because their rental rates are lower, assuming environmental benefits are not directly related to the land’s crop productivity (USDA, 2003c). The ultimate effects of such payment programmes on production and trade must consider all farmer responses to the impacts on land allocations and prices. For example, Wu (2000) estimated that for every 40.5 hectares of cropland retired under the United States’ CRP in the late 1990s, another 8.1 hectares of non-cropland were converted to crop production. This “slippage” effect is hypothesised to occur due to farmers’ responses to higher crop prices and their attempts to substitute other areas of land for those that have been retired. Wu argues that the slippage effect decreased water and wind erosion control benefits from the CRP. A subsequent analysis, using the same data, concludes that Wu’s findings were the result of spurious correlation and that there is no convincing evidence of slippage (Roberts and Bucholtz, 2005). The authors argue that Wu’s estimates of slippage did not account for the endogenous nature of CRP land decisions (i.e. farmer responses to the CRP depend on regional land quality, and the data are not suited to capture price feedback effects). Although the Roberts and Bucholz analysis did not find evidence that slippage occurred for the CRP during that period, it does not mean that slippage cannot occur in the future or under land retirement programmes in other OECD countries. Analyses that estimate the effects of agri-environmental payments on arable crop production and trade also must consider other agricultural policies that shape the total supply response. Agricultural support policies may or may not offset the production changes, depending on the policy mechanisms used. For example, a study on the PFCP in the United States, estimated only a slight increase in supplies (USDA, 2003c). If agrienvironmental payments do not generally alter famers’ decisions to grow specific arable crops (and thus influence production levels), then their AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 271

potential trade effects also would be reduced. Decoupling is feasible if the programme intent is to alter “input” levels to reduce external costs. However, as explained above, if the programme objective is to supply environmental services that are linked directly with production of the crop over certain ranges, then the payments will inevitably influence production. If the programme is intended to alter production methods to more environmentally friendly technologies, irrespective of the particular crop or production level, then the payments are less likely to alter the composition of trade and more likely to achieve efficiency in terms of both production and environmental effects. Of course, a land retirement programme will have very different impacts from a payment programme designed to retain land in crop production but under environmentally friendly production practices. For example, in the United States, land retirement programmes have been extended under the 2002 FSRI Act. In the EU, programmes to foster less intensive production and to reduce environmental pressure are in various stages of implementation, including organic farming aid schemes.

7.4.2.

Previous analyses

Very few studies have investigated the trade impacts of agrienvironmental payments. Diakosavvas (2003) conducted one of the first empirical investigations into the aggregate production and trade effects of such payments. Data from country notifications of Green Box expenditures for agri-environmental programmes over the 1995-98 period were assembled to portray the differences across countries and the time period. The notification data generally portray an increasing trend of payments in most countries, which suggest a growing potential to influence farmer production decisions. Most of the environmental payments are linked to area planted, livestock numbers, or input use, and therefore hold the potential to be economically distorting. Diakosavvas (2003) used two empirical procedures to test the relationship between the payments and production and trade. The first is a meta-production function approach that attempts to capture the aggregate input-output relationship of the agricultural sector for all countries in the database. The model has four conventional explanatory variables: labour, land, fertiliser and machinery, along with one policy shifter to capture the effect of agri-environmental payments as reflected by WTO Green Box notifications data. Using pooled cross-sectional and time series data, a Cobb-Douglas production function was estimated in "intensive" form to reduce statistical problems. The overall relationship between output and the independent variables was statistically significant. The conventional input variables displayed the expected positive signs as well as the environmental

272 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

payments variable, suggesting that increasing payment levels are significantly associated with increasing output levels. The environmental payment variable showed a smaller effect than the conventional inputs. A potential complication in interpreting the findings is that the environmental payments variable is likely to be correlated with overall agricultural support and may be capturing some or all of that larger effect. The second procedure used a gravity trade model to estimate the effects of the payments on bilateral trade flows between countries. Again, the estimates indicate that the Green Box agri-environmental payments by exporting countries are positively and significantly associated with increasing agricultural exports. As for the production function analysis, the potential exists that the payments variable is also reflecting some of the influence of overall agricultural support. Diakosavvas concludes that his tests are only a first consideration in determining whether agrienvironmental payments affect production and trade. Further analysis is necessary to determine the extent to which the crop or animal outputs are joint with environmental services. A second study by Lubowski, Platinga and Stavins (2004) examined the largest agri-environmental payment programme in OECD countries, the United States’ CRP. The CRP expends approximately USD 2 billion per year and has approximately 13.7 million ha enrolled, which accounts for nearly 9% of the nation’s cropland (approximately the size of the entire state of Iowa). This temporary land retirement programme reduces the level of national production of arable crops by the amount that would be planted on the enrolled land net of any slippage effect. Despite its potential impacts, very few studies of the impacts of the CRP on crop production and prices have been conducted, perhaps because of the detailed data needed for enrolled land and the complexity of interactions with other agricultural policies. As mentioned above, agri-environmental policies need to be analysed in connection with other agricultural policies. This comprehensive analysis estimated the impacts of the CRP on potential trade distortions, in conjunction with the effects of other agricultural policies (e.g. direct federal payments) on land. In summary, the results indicate that the positive impact on crop acres and output of direct payments to farmers were more than offset by cropland retirement under the CRP. The authors estimate that the value of aggregate crop production in the United States in 1997 would have been 2% higher in the absence of both the CRP and direct federal farm payments (Lubowski, Platinga and Stavins). Therefore, the special combination of the CRP with direct payment programmes essentially cancels out production and trade impacts for the United States. However, as the authors note, this aggregate analysis does not

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 273

reveal potential shifts in crop composition due to the mix of the CRP and agricultural support policies. The two empirical investigations of agri-environmental payments arrive at quite different conclusions. The Diakosavvas study finds that the level of payments is directly related to output and influences trade flows. The Lubowski, Platinga and Stavins study estimates that the CRP exerts a drag on the output-enhancing effect of other agricultural support payments and essentially counteracts any distortion in aggregate production and trade. Some of the difference may be attributed to the different levels of aggregation in the models, which hides specific programme effects in the Diakosavvas study. A third study, by Cooper, Peters and Claassen (2003) used a multiple commodity spatial partial equilibrium model of the US agricultural sector to examine the trade impacts of three generic agri-environmental schemes that provide farmers with incentive payments to encourage farm management activities that reduce erosion. It finds that, for the three agri-environmental payment scenarios evaluated, the maximum change in exports ranges from a 7% decrease (wheat) to a 1% increase (soybeans). Further study is clearly needed.

7.4.3.

Simulating potential trade effects of agri-environmental payments

The Larson, et al. (2002) approach can be adapted to shed light on the potential impacts of payments by assuming that the payments decrease the average variable costs of production. Note that this assumption implies that such payments are on a regular annual basis that lower the annual costs, or are used to cover the initial costs of installing practices or equipment that subsequently lower the variable costs of production. If the payments simply offset the added cost of implementing new practices or only partially compensate for required changes of practice, then they would not affect the variable costs of production, and conceptually no production effect would occur. Table 7.1 shows the simulated changes in production, imports and exports for selected countries and crops under the assumption of 1%, 3% and 5% average cost decreases and fixed world prices. In addition to the cost increase, the values in the cells depend on the domestic supply elasticity, and the ratios of domestic production over imports or exports (see Annex 7.A3). Tables 7.A1, 7.A2 and 7.A3 present the values used for the required parameters. It should be noted that the simulation estimates do not relate to a particular programme, but only to the potential impact if such payments were made for the particular crop by the specified country. Those country-crop combinations were chosen arbitrarily and were influenced by

274 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

available data for the key parameters. One implication of the simulation analysis is that as payments grow to 5% of costs, changes in production approach or exceed 5% and induce larger impacts in the trade flow for the country. Of course, the estimates only reflect the actions of a single country in isolation from other countries’ actions. To estimate actual trade impacts, the simultaneous actions of third countries would need to be considered.

7.5.

Effects of agri-environmental regulations on factor costs and trade Few empirical analyses of environmental compliance costs for agriculture and their potential effects on competitiveness have been conducted. The national studies reported by Brouwer and Ervin (2002) conducted extensive searches for cost impacts, but could uncover only partial estimates for a limited number of issues in particular countries. Comprehensive evaluations are not commonly available, even for major national regulatory programmes such as pesticide management and conservation compliance. As already noted, the task is more challenging than just assessing the impacts of national programmes. The effects of regulatory standards at all levels of government should be included, but subnational data are very difficult to obtain, given the diversity of programmes and lack of reporting.

Table 7.1. Estimated changes in arable crop output (Y), imports (I) and exports (E) due to agri-environmental payments Cost decrease

1% Y

I

3% E

Y

I

5% E

Y

I

E

% change Wheat

Australia

1.0

n.a.

1.5

3.0

n.a.

4.5

5.0

n.a.

7.4

Maize

Mexico

.72

-2.3

n.a.

2.2

-6.9

n.a.

3.6

-12

n.a.

Rice

Japan1

.81

-10

13

2.4

-31

39

4.1

-51

65

Soybeans

Canada1

1.8

-4

5.0

5.5

-12

15

9.1

-20

25

Notes: n.a. = not available. 1. Japan and Canada import and export similar quantities of rice and soybeans, so both potential trade-flow changes are calculated. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 275

7.5.1.

Previous analyses

It is worth noting first that a number of studies have investigated the effects of environmental regulations on firm costs and the international competitiveness of industrial sectors other than agriculture. The effects on firm costs vary considerably across sectors, reflecting the different production and distribution processes used, but average approximately 2-5% of total costs. Given this relatively small percentage and the fact that other trading countries tend to impose similar environmental requirements, empirical analyses have found that the regulations generally have exerted insignificant or negligible effects on trade. Two early studies examined the effects of environmental regulations on crop agriculture in OECD countries and came to contrasting conclusions. Tobey (1991) estimated the potential for different crops to generate pollution and correlated the estimates with the revealed comparative performance of crops in the world market. He found that the crops that perform well in world markets also have the largest pollution potential. Therefore, stringent programmes to control that pollution could affect their trade performance. However, he concluded that trade competitiveness effects are likely to be quite modest for three reasons. First, most competing exporters have introduced similar agri-environmental programmes, which implies that the relative trade competitiveness effects have not been significant. Second, developing countries do not hold large market shares in most of the commodities. Finally, the competitiveness effects of agri-environmental programmes are likely to be swamped by larger forces such as labour costs and exchange rate fluctuations. In the second study, Diakosavvas (1994) assumed that each crop’s relative pollution potential was directly correlated to the proportion of pollution abatement costs (PAC) in total production costs, and empirically tested for trade distortions. The author’s estimates showed statistically significant correlations between trade distortions and the assumed PAC measure. Since many OECD countries used voluntary and compensatory agricultural pollution control programmes during the analysis period, the assumed relative PAC expenses may not fully capture the actual compliance costs. The study findings highlight the potential trade impacts if the voluntary and compensatory agri-environmental policies give way to more regulatory approaches. The Tobey and Diakosavvas studies had to make assumptions about the nature (e.g. coverage and stringency) of agrienvironmental programmes due to lack of detailed information. Improved data on programme implementation is needed to assure the viability of such assumptions.

276 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

A study by Grote, et al. (2000) analysed the impact of agrienvironmental standards on the international competitiveness of farmers and agro-processors in Germany compared to competitors in Brazil and Indonesia during 1998-99. The investigation was based on production-level data collected by experts in the study countries. The researchers found that in Germany, which has more stringent technical security and environmental standards than Brazil and Indonesia, 0.3% to 4.4% was added to the total production costs for typical rapeseed and grain producers. The typical German poultry producer faced 2.7% higher production costs due to relatively more demanding environmental and animal welfare regulations. In the oilseed-processing industry, environmental compliance costs amounted to 5% of total production costs in Germany, 0.5 to 1% in Brazil, and 0.4 to 1.1% in Indonesia. The cross-country differences in environmental compliance costs at the farm and processing level were found to be considerably smaller than the differences in total production costs, which would suggest that other factors, including producer support, wage levels, land rents and capital costs, are more important determinants of overall production costs and international competitiveness. Also, the study revealed that many of the cost-increasing environmental standards in Germany (for example, with respect to building codes), are of little relevance in Brazil and Indonesia, because of differing climatic conditions and resource endowments. However, the results of Grote, et al. should be interpreted with care, as their analysis was based on a relatively small sample of agricultural products and processing facilities. Research by the United Kingdom’s National Farmers’ Union (Wilkinson, 1998) found that the vast majority of the 224 regulations aimed specifically at agriculture in the UK was intended to prevent or control crop and livestock diseases (55% of all sector-specific regulations); to ensure food safety (26%); to protect the environment (7%); to enhance animal welfare (6%) and to achieve other policy objectives (7%), such as seed quality and integrity. More than 80% of the regulations applied to livestock operations, while about a quarter concerned arable farming or horticulture (some regulations related to both livestock and crops). Furthermore, the study estimated the impact of environmental regulations on farming costs, based on information from parliamentary compliance-cost assessments. The results showed that individual regulations, such as the Crop Residues Burning Regulation or the EU Nitrates Directive, impose costs on affected farmers varying from 1.2% of their value of output (Crop Residues Burning), to 2% (Nitrates Directive for dairy farmers). However, the impact was found to be much less pronounced at the national level, as not all farmers are affected to the same extent by particular regulations. For example, the Nitrates Directive restricts emissions of nitrates in designated AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 277

nitrate-vulnerable zones, but does not impose similar constraints on farming practices outside these zones. A recent inventory of agri-environmental regulatory standards was conducted in the EU, the United States, Canada, Australia and New Zealand to provide more detailed information, including likely effects on farm costs (Brouwer and Ervin, 2002). Unlike other assessments, the exercise attempted to include provincial/state and local regulations in addition to national programmes, in order to capture the “cumulative” regulatory impact on producers. In many cases the inventory relied on expert judgment because of inadequate data, especially reliable quantitative cost estimates. The nature of the constraints was summarised by subjectively assessing the significance of the agri-environmental issues across countries (Table 7.2). Note that nutrient, pesticide, sediment pollution, irrigation and salinisation are all associated with arable crops and were judged significant (2 or 3 stars) in one or more of the countries studied. The editors judged that the environmental compliance costs for agriculture are likely to fall in the range of 3%-4% of gross revenue, similar to non-agricultural sectors. However, they caution that the compliance cost data are weak and incomplete, especially at the sub-national level, and that variances within countries may cause significant competitiveness effects for some crop sectors. The five country/region studies reported in Brouwer and Ervin suggested that regulations on the livestock sector posed the largest potential threat to competitiveness for the various countries. OECD recently conducted major assessments of the pig and dairy sectors that compare the costs of manure management across OECD countries (OECD, 2003f; 2004a). The pig sector study found that regulatory measures to address water pollution, odour, ammonia and greenhouse gas emissions are increasing in severity and complexity. The analysis concluded that manure management regulations are not significantly different between countries, and that regulatory costs only partially explain differences in trade competitiveness and are a minor consideration in location decisions. The dairy analysis compared the costs of manure management across OECD countries and estimated that the gross costs (not the added costs caused by the environmental regulations) varied from approximately 2% to 4% of production costs per cow. The study concluded that this range of costs would not exert significant effects on dairy trade flows among OECD countries.

278 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 7.2. Issues of concern in the five countries/regions

Issue

EU

United States

Canada

Australia

New Zealand

Nutrient enrichment by nitrates and phosphates

***

***

*

*

*

*

***

*

*

**

Pesticides (including drift and applicator safety)

***

***

*

**

*

Irrigation

**

**

*

***

**

Salinisation

*

*

*

***

–

Soil contamination

*

*

*

*

*

Soil erosion

**

***

*

***

***

Ammonia

**

–

*

–

–

Odour and nuisance

***

**

***

–

–

Crop burning

*

*

–

–

–

Biodiversity, landscape

***

**

*

***

***

GMOs

***

*

*

**

***

Animal welfare

***

*

*

*

*

Hormones (and animal feed ingredients)

**

–

*

*

*

Pesticide residues in food

**

***

**

**

**

Hygiene rules for dairy farming

**

**

**

*

*

Veterinary and animal diseases

**

*

*

*

*

Sediments

Notes: – No issue. * Issue identified as a problem, but not of major concern. ** Issue identified as a problem, and of significant concern. *** Major issue with high priority in policy. Source: Brouwer and Ervin, 2002.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 279

The present analysis could not employ the analytical approach used in the studies on the pig and dairy sectors due to lack of reliable data on the costs of environmental regulations for arable crops. Therefore, a sensitivity analysis approach was used to illustrate the impacts of potential regulations for various country-crop combinations, as explained below. Larson, et al. (2002) conducted case studies of the effects of environmental regulations on agricultural exports in several countries and report empirical estimates for Cyprus and Tunisia. They used available data or estimates of the needed parameters to model export impacts as described above. The Cyprus case evaluated the impacts of probable increases in the cost of irrigation water and fertilisers (e.g. nitrogen) on the production and exports of potatoes. The Tunisian case also evaluated the impacts of expected increases in irrigation water prices on fruit production and exports. The authors examined a range of water price increases from 20%-60% and a 40% increase for fertiliser in Cyprus. The findings illustrate the importance of the specific country and regulatory conditions affecting the parameters for the estimation methodology. For example, the impacts on Cypriot potato exports due to a 20%-60% increase in water prices were estimated at -0.6% to -3.4%, while a 50% water price increase in Tunisia resulted in an estimated decrease in exports from 3% to 4.9% for citrus, but a 14%-26% decrease for dates. The large difference in the citrus and date impacts primarily depends on the higher cost share of water in date production. The Larson, et al. (2002) analysis cautions that broad generalisations about the effects of environmental regulations on exports are inappropriate. Detailed empirical analyses are necessary to understand the individual country situations. Their findings show that variations in individual country and crop situations are the key to assessing potential trade impacts. However, their analyses do not capture the multilateral agri-environmental policy interactions. For example, the studies reported in Brouwer and Ervin (2002) suggest that major competing exporters often adopt similar types and levels of regulations for their agricultural producers. If this is accurate, then the effect is to raise the aggregate export supply curve and increase world price, thereby lessening the potential export impacts in any one country.

7.5.2.

Simulating potential trade effects of agri-environmental regulations

Given the lack of reliable compliance cost data, the potential effects on the arable crops trade are illustrated using simulation analyses with the modelling approaches developed by Larson, et al. (2000) and Larson and Scatasta (2003). The simulations are constrained by incomplete data and the simple modelling framework used.

280 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 7.3 presents the assumed scenarios that are analysed for selected arable crops, countries, types of regulation and level of cost impact. The countries were selected based on their production and trade positions in the respective crops and on the availability of key parameters (e.g. output supply and input demand elasticities). Four arable crops were included – wheat, maize, rice and soybeans. Three types of regulation were examined: (a) the percentage increase in the price or cost of a composite factor input (columns 3 and 4); (b) the percentage increase in average variable costs (column 5) and (c) the cost of product regulation (column 6). Three levels of cost increase were analysed for each type of regulation, and three factor shares were analysed for the composite input price increase. The input price and average cost increase levels were chosen to illustrate the sensitivity of the estimates to variations in these parameters. The three price and cost levels should not be interpreted as equally likely to occur. Recent studies suggest that likely increases generally fall toward the lower end of the ranges (e.g. Brouwer and Ervin, 2002). Annex 7 explains the modelling approach used, and Tables 7.A1, 7.A2 and 7.A3 give the assumed values for the parameters used in making the estimates of production and trade impacts reported below. Table 7.3. Regulatory cases Analysis features

Small country

Large country

Small country

Large country

Regulation type

Composite input price

Composite input price

Average cost

Product regulation

Cost impacts Factor shares

10, 50 and 200% 5, 10, and 20%

10, 50 and 200% 5, 10, and 20%

1, 3 and 5% n.a.

10, 20 and 50% 5, 10 and 20%

Australia Mexico Korea/Japan Canada

EU US None US

Australia Mexico Korea/Japan Canada

EU US None US

Wheat Maize Rice Soybeans Note: n.a. = not available.

Table 7.4 presents estimates for the case of regulating a production input under a fixed crop price (small country) assumption. As with all of the simulation analyses, the largest percentage changes are for the country’s minor trade flow and therefore represent small absolute quantity shifts. The production and major trade flow percentage changes are relatively small for AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 281

the 10% cost increase case, and then increase linearly for higher cost increases. Although not reported here, the impacts become larger and more significant as the factor share rises to 20%. Therefore, environmental controls on inputs that comprise a larger share of production expense will exert larger effects. Table 7.4. Simulated changes in output (Y), imports (I) and exports (E) due to input regulation: small country Cost increase (10% factor share)

10% Y

I

50% E

Y

200%

I

E

Y

I

E

% change Wheat

Australia

-6

n.a.

-.9

-3.0

n.a.

-4.5

-12.0

n.a.

-17.9

Maize

Mexico

-4

1.4

n.a.

-2.2

6.7

n.a.

-8.6

27.7

n.a.

Rice

Japan1

-5

6.1

-7.8

-2.4

30.6

-38.8

-9.7

~100

~-100

Soybeans

Canada1

-1.1

2.4

-3.0

-5.5

12.0

-15.1

-21.8

48.0

-60.2

Note: n.a. = not applicable. 1.

Japan and Canada import and export similar quantities of rice and soybeans, so both potential trade flow changes are calculated.

Table 7.5 presents the simulated impacts for the large country case of input regulation. Very small to modest production and export impacts are estimated until the factor price increase reaches 200%. A primary difference from the small country case is that the world price increase compensates for the loss in production and exports resulting from the imposition of the environmental regulation on the pollutive factor. Table 7.6 shows the effects of environmental regulations that are not factor-specific but they increase average variable production costs. Recall that the trade effects on agri-environmental payments (Section 7.4.3) were modelled as a decrease in average costs, so the effect of non-specific regulations is the reverse. The simulated impacts suggest that a 1% increase in average cost causes approximately equivalent impacts to a factor price increase of 10% (Table 7.4). The analysis also suggests that average cost increases of 3-5% can cause moderate production and trade impacts, assuming no equivalent agri-environmental actions by other countries that could combine to lift world prices and moderate or offset the trade impacts.

282 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 7.5. Simulated output (Y), import (I) and export (E) impacts from input regulation: large country Cost increase (10% factor share)

10% Y

I

50% E

Y

200%

I

E

Y

I

E

% change

Wheat

EU

1

-1.0

3.0

-.51

-5.1

15.0

-2.5

-20.6

60.2

-10.2

Maize

US

-.65

n.a.

-.38

-3.2

n.a.

-1.9

-12.9

n.a.

-7.6

Soybeans

US

-.48

n.a.

-.20

-2.4

n.a

-1.0

-9.6

n.a.

-4.0

Note: n.a. = not applicable. 1. The EU imports and exports similar quantities of wheat, so both potential trade flow changes are calculated.

Table 7.6. Simulated output (Y), import (I) and export (E) impacts of agrienvironmental regulations that increase average variable cost Cost increase

1% Y

I

3% E

Y

I

5% E

Y

I

E

% change Wheat

Australia

-1.0

n.a.

-1.5

-3.0

n.a.

-4.5

-13.6

n.a.

-7.4

Maize

Mexico

-.70

2.3

n.a.

-2.2

6.9

n.a.

-3.6

11.5

n.a.

Rice

Japan

-.81

10

-13

-2.4

31

-39

-4.1

51

65

Soybeans

Canada1

-1.8

4.0

-5.0

-5.5

12.0

115

-9.1

20

-25

1

Note: n.a. = not applicable. 1. Japan and Canada import and export similar quantities of rice and soybeans, so both potential trade flow changes are calculated.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 283

Table 7.7 presents the simulated impacts of imposing a product standard (e.g. a pesticide or residue limit), on the maize trade between the United States and the rest of the world (e.g. EU). The effect of the product standard is assumed to operate through increasing factor costs. Therefore, the same cost increase scenarios as those applied in Tables 7.4 and 7.5 are used. For this exercise, different cost increases are assumed for the United States and other countries. In case 1, other countries/regions (e.g. the EU) are assumed to have a cost advantage in implementing the product regulation of 90% of the amount the United States would have to pay to implement the regulated systems. The case 2 entries reverse the cost advantage to the United States. As expected, the trade adjustments depend on the relative ability of each trading partner to adjust in a cost-effective way to the regulations. Larger impacts on the United States maize exports are estimated for the case where the EU holds the cost advantage in implementing the product regulations and for the 50% and 200% factor-cost increase scenarios. Larson and Scatasta (2003) show that product standards for both domestic and foreign products could induce more domestic production, as may occur in the EU when it holds the assumed cost advantage. Table 7.7. Simulated Output (Y), Consumption (B), and Import (I) impacts from product regulations Cost increase

10% Y

B

50% I

Y

B

200% I

Y

B

I

% change EU Maize

EU cost increase = 90% of US

-0.20

EU Maize

US cost increase = 90% of EU

-0.60

7.6.

-0.40

-0.30

-0.2

0.20

-2.2

-1.0

0.40

-7.6

-4.0

-0.20

-0.3

-1.7

-1.0

-1.1

-6.9

-4.2

Suggestions for enhancing the effectiveness of agrienvironmental policies on arable crops The simulations illustrate that agri-environmental payment and regulatory programmes in OECD countries could conceptually cause significant effects on arable crop factor costs and trade.3 The analyses estimate the impacts when only one country implements the programmes, so

284 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

actual import or export impacts may be muted to the extent that other major trading countries implement similar programmes. In fact, analyses to date suggest that many OECD countries have implemented similar agrienvironmental standards, and that may be why significant aggregate trade impacts have not been detected in most previous analyses.

7.6.1.

Reactive or proactive policy approach?

The expansion in agri-environmental payment and regulatory programmes in most OECD countries indicates larger production and trade impacts for the arable crop sector. As noted, the United States’ EQIP and CSP are forecast to grow significantly during the next decade. Another example is the growth of payments under the EU Rural Development Regulation (No. 1257/99) to promote the adoption of less input-intensive cropping systems, including organic farming systems. Pesticide regulations, which have exerted significant effects on factor costs in arable crop production, will most likely continue to be strengthened. Simultaneously, the trend towards greater trade liberalisation in agriculture appears equally strong. A reactive policy approach is to monitor the trade impacts to detect significant changes before taking policy action. However, assessments of the trade effects of agri-environmental programmes often fall short on theoretical, modelling and empirical grounds due to the complexity of the processes under investigation. For example, a recent paper compared the predictions made by several economic models of the effects of NAFTA with the actual levels of production and trade that occurred subsequent to its implementation (Carpentier, 2002). The general finding of this comparison was that the economic models did not do a very good job of sectoral forecasting of actual levels of production and trade. Several analytical difficulties contribute to inaccurate estimates of the responses by complex economic systems. First, altering world prices under trade liberalisation agreements triggers production and import shifts in all countries participating in open global markets and results in important feedback effects that are often not included in theory or modelling. Second, production and trade responses differ across countries because production systems and agricultural policies vary. Third, many countries are in the first stages of liberalisation, and there is a lack of data and virtually no experience on which to base accurate estimates of production adjustments. Fourth, temporal changes, such as policy and technology innovation and adoption induced by trade liberalisation, are very difficult to forecast. Finally, ex ante modelling exercises are often based on the expected AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 285

outcomes of on-going trade and other policy processes. The actual outcomes may be quite different from the predictions made for the modelling analysis. Weaknesses in such ex ante assessments can lead to serious risks for trade and environmental quality. The policy challenge is clear – ways must be found to achieve progress on each social objective with minimal negative effect on the other. This challenge returns to the key theoretical conclusion from the study of the competitiveness and trade implications of environmental programmes – both open trade and optimal environmental policy are requisite to maximising national and global economic welfare (Anderson, 1992). Programmes that cost-effectively remedy external environmental costs and provide public environmental benefits from arable crop farming are needed to reduce potential trade effects. This proposition suggests assigning a high research priority to answering the following question: Are “effective” agri-environmental policies in place to address the ecological and environmental effects of agricultural trade liberalisation?4 This proactive focus reverses the perspective of most agricultural trade and environment analyses that attempt to estimate the impacts of existing environmental or trade policies (Ervin and Fox, 2004). The following section is devoted to reviewing some the lessons from research and experience with agri-environmental policies in OECD countries in order to identify practical avenues for moving towards an affirmative answer.

7.6.2.

Some lessons from analysis and experience

Several important lessons about ways to improve the environmental, economic and social performance of agri-environmental policies have emerged (Claasen, et al., 2001; Ervin, et al., 1998; Ervin, 1999; OECD, 2001d and 2000b). These are discussed below as steps in a general policy process, but each country’s conditions may require a different sequence and combination.

a)

Minimise conflicts between agricultural and agri-environmental policies

A first step in preparing the foundation for effective agri-environmental programmes is to remove conflicting influences from other agricultural policies. Claassen, et al. (2001) relate how the US 1985 FSA reduced the conflicts between federal commodity, loan and crop insurance policies and soil conservation and water quality objectives through the CRP and conservation-compliance programmes. The compliance programmes provided a model for other countries, and several have now implemented their own versions (Chapter 4). For example, the EU countries of Denmark,

286 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Greece, Finland, France, Italy, Spain, the Netherlands and the United Kingdom have introduced forms of voluntary cross compliance under the Agenda 2000, although such efforts have been characterised as “patchy”. Other OECD countries, including Korea, Norway and Switzerland, have also implemented compliance schemes for arable crops with varying degrees of coverage of environmental issues and strengths of sanction. As discussed in Chapter 5, the cost effectiveness of an environmental cross-compliance provision linked to direct payments relative to the incremental cost of the cross compliance scores very highly in the evaluation. However, this policy approach inevitably involves some compromise between income support and environmental objectives. Hanley and Whitby (2003) illustrate the importance of reducing conflicts between agri-environmental programmes and production and income support. They cite two examples, one in which farmers who enter the United Kingdom’s Habitat Improvement Scheme would lose GPB 500 per hectare compared to the AAPS, and another in which a farmer in a species-rich grassland area in a river valley would receive 10 times as much for converting to potatoes as from the Countryside Stewardship Scheme.

b)

Target programme resources to achieve the highest priorities at the lowest costs

The heterogeneity of environmental and social conditions usually means that geographically uniform eligibility, payments or regulations rarely produce cost-effective results. Evidence from OECD countries affirms this notion. The United States substantially improved the environmental performance of the CRP by better targeting the programme resources to areas and issues considered higher social priorities (Claassen, et al., 2001; Smith, 2002). Ideally, national targeting schemes should include multiple environmental problems to minimise the potential effect of making some environmental problems worse. The consideration of multiple environmental targets also makes more farms eligible for the payments, which can serve social equity goals established by elected political representatives. Other countries have used targeting schemes to improve the cost-effectiveness of their agri-environmental programmes, for example the United Kingdom’s ESAS (Hanley and Whitby, 2003). Targeting schemes can also incorporate the potential for payments to produce enduring land-use change that takes into account potential environmental services over the long-run period possible.

c)

Grant farmers flexibility in meeting programme requirements

This guideline enables farmers to select the lowest-cost method of achieving the environmental performance target in their specific resource AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 287

and management settings (Claassen, et al., 2001). Mirroring the rationale for targeting environmental programmes to priority areas because of heterogeneous conditions, granting farmers flexibility recognises that each arable crop production unit faces a unique set of natural resource, human and other conditions. The farmer is in the best position to discern the most cost-effective way for his/her operation to achieve the environmental objective. Flexible approaches also allow the farmer to develop innovative approaches to solving problems. The Australian or New Zealand Landcare programmes and the EQIP the United States emphasise the importance of farmer flexibility in creating durable land management solutions (OECD, 2003e). Fostering the supply of environmental services through market-like mechanisms can improve programme flexibility and cost-effectiveness. For example, tradeable pollution permit programmes designed to achieve pollution reductions at the lowest cost are one of the most important environmental policy innovations. Markets are powerful institutions that play key roles in resource allocation. They signal the strength of preferences for goods and services by means of prices reflecting relative demands. On the supply side, they transmit information via prices and costs to efficiently organise production and distribution. Most environmental services cannot be produced and distributed efficiently through private markets because of nonrival and/or non-exclusive characteristics. However, some of the roles of markets may be simulated in public environmental programmes. For example, schemes that allow landowners to bid for participation in land retirement programmes, such as the United States’ CRP, have helped maximise environmental benefits per dollar spent (Smith, 2002). Tradeable permit schemes also have the potential to help solve some agrienvironmental problems, but the difficulty of identifying non-point sources has hampered progress (Chapter 4).

d)

Use the level of government, i.e. local, state/provincial, national or international, that is most cost-effective for the particular environmental problem

The proper division of responsibility, authority and resources among levels of government is a subject of ongoing political discussion in many OECD countries. In the United States, the “devolution debate” was originally concerned with issues other than environmental policy, but this changed in the early 1990s. Political groups increasingly intent upon shifting the balance of power away from national government argued that the rise in centrally managed environmental programmes since the 1970s was costly and inefficient. Because of highly diverse agri-environmental situations, it is difficult to devise policies at the national level – or even at state/provincial

288 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

capital level – that effectively influence farm-level behaviour in such a way that programme objectives will be maximised. It is also possible that subnational programmes can be more cost-effective in administration because of increased knowledge and closer proximity to the problems. The administrative and transaction costs are often omitted or only partially captured in agri-environmental programme evaluations. Recent assessments in Canada and the United States have found that state/provincial and local governments are increasingly implementing agrienvironmental regulations (Fox and Kidon, 2002; Carpentier and Ervin, 2002). However, given the magnitude of the environmental challenge in agriculture, it is equally clear that a large portion of the financial resources must come from national sources where the majority of revenues are collected. Federal government also has a role to play in targeting resource problems at local level and in managing environmental problems that cross sub-national political boundaries. Options for achieving the goal of addressing national public goals while accounting for local needs could include partnership between federal, state, local and/or private participants.

e)

Stimulate R&D that fosters the development of cropping systems that maintain profit and contribute to environmental objectives

An essential, but often overlooked, strategy to improve the long-run cost-effectiveness of agri-environmental programmes is R&D policy. Agricultural research has probably not been sufficiently responsive to many agri-environmental concerns (Ervin, 1998). Incomplete or non-existent markets for many environmental services and natural resources hamper the effectiveness of price incentives to stimulate either public or private R&D. Missing markets for certain environmental resources create external benefits and costs that are not captured or paid by the sellers or buyers of agricultural products. In addition, financial payments under voluntary programmes mostly reward the adoption of existing technologies – an approach that often fails to reward and therefore stimulate the development of new research and technology. Failing appropriate scarcity signals for such environmental assets, R&D responses may concentrate on remediation rather than forwardlooking investigations to avoid pollution or excessive resource degradation. Past research on agri-environmental issues has largely been reactive, i.e. conducted after a problem has arisen. Without more incentive to anticipate emerging problems and consider ecosystem-scale relationships, the most cost-effective solutions will not be discovered. The weaknesses of current research on agri-environmental issues have been chronicled (Robertson, et al., 2003) and future research on the interconnectedness of ecological systems, the essential dynamic roles of ecosystem services, AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 289

non-linear and threshold responses to accumulating pressures from agricultural production would be beneficial. An ideal R&D system would foster precautionary R&D that anticipates the effects of these increasing pressures and creates technologies and management systems that maximise the long-term potential of agriculture to produce food and environmental services.

290 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Annex 7.A. Equations Used to Estimate the Trade Effects of Agri-Environmental Programmes

Larson (2000), Larson, et al., (2002) and Larson and Scatasta (2003) develop several cases that can be used to estimate the impact of environmental regulations on the imports and exports of arable crops. To illustrate one case from Larson and Scatasta (2003), consider the impact of domestic environmental regulations on an input or factor, e.g. fertiliser or water, used in the production of a homogeneous product, such as a bulk arable crop, on the domestic country’s imports of the crop, under the assumption of fixed import prices (i.e. the small country case). The regulation is assumed to increase the input’s price from w’ to w’’. The impact on imports is found by creating an expression for the elasticity of output supply with respect to the input price (n yw ) , and calculating the initial percentage change in domestic production ( dY / Y ) from the policy change ( dw / w) using the following equations:

n yw

wY w ww Y

ª wX º ª C º c « « » n xy n yp ¬ C »¼ ¬ pY ¼

(1)

dY Y

n yw

dw w

(2)

dI I

dY (Y / B) Y ( I / B)

(3)

p = output price, w = factor price, X = production factor, C = variable production cost, I = imports, and B = where Y = output,

domestic demand for the output. This calculation requires six pieces of information:

§ wX · ¸ © C ¹

1. Regulated input’s cost as a share of total production cost: ¨

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 291

§ C · ¸¸ © pY ¹

2. Total costs as a share of total revenues: ¨¨

3. Domestic supply elasticity with respect to output price: n yp

wY p wp Y

4. Returns to scale of the industry as reflected by the elasticity of the cost c

minimising input demand with respect to output level: n xy

wX c X wY Y

w ’’ w ’ w’ Y (Y / B) 6. Share of domestic production relative to imports: I ( I / B) 5. Regulatory impact on production costs:

dw w’

If the importing country is large in the sector, then a shift in domestic demand could lead to import price changes. Two additional pieces of information are needed to incorporate import price adjustments: 7. The import supply elasticity with respect to price ( n sp ) 8. The domestic demand elasticity with respect to price (n Bp ) In this case, if the import price increases, some of the costs of the domestic environmental regulation will be passed along to import prices, which will tend to mitigate the impact of the regulations on import quantities. Some agri-environmental regulatory programmes do not apply to a specific input. In such situations, the average costs of production increase, rather than the effective price of the pollutive input. Larson and Scatasta (2003) show that the import impacts for these cases can be estimated more simply than for the input regulating case with just three pieces of information: domestic supply elasticity with respect to output price (n yp ) , current market price ( p ) and average variable production cost increase due to regulation (m). More specifically,

dY Y

n yp (m / p ) .

A similar modelling approach can be used to analyse export impacts. The impact of environmental regulations on exports of a crop by a small country requires the following pieces of information (Larson, et al., 2002):

292 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

§ wX · ¸ © C ¹

1. Regulated input’s cost as a share of total production cost: ¨

2. Profitability of the industry, as reflected by percentage of profits over

§ pY C · ¸ © C ¹

production costs: ¨

3. Domestic supply elasticity with respect to output price: ( n yp ) 4. Returns to scale of the industry as reflected by the input demand elasticity (cost minimising) with respect to output level: n

c xy

wX c Y wy X

5. Regulatory impact on production costs: ( dw / w ’ ) 6. Share of domestic production relative to exports: (Y / E ) If the country is large and can influence export prices, two additional pieces of information are required: 7. The elasticity of export demand with respect to export price, ( n Dp ) 8. The domestic demand elasticity with respect to product price, (n Bp ) Finally, in the case of product regulations that target both domestic and foreign producers, e.g. pesticide residue and GM crop labelling, the analysis must include a second stage that accounts for the additional costs that are imposed on import suppliers. Larson and Scatasta (2003) show that such product standards for both domestic and foreign products could induce more domestic production, depending upon the particular supply and demand elasticities. The following lists the basic equations used to calculate the trade impacts for those cases. The reader is referred to the original articles for the derivations of the equations.

7.A1

Small country import impact of agri-regulation on factor that increases the factor price (marginal cost) Consider the case of a small country that imposes an environmental regulation on an input, e.g. fertiliser or water, used in the production of a homogeneous product, such as a bulk arable crop, on the country’s imports of the crop, under the assumption of fixed import prices (i.e. the small country case). The import impact is found by creating an expression for the

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 293

elasticity of output supply with respect to the input price n yw , and calculating the initial percentage change in production (dY / Y ) from the policy change ( dw / w) using equations (1), (2) and (3).

7.A2

Large country imports The import impacts if the country is large and influences world price is found by:

n pw

ª º p « n yw » « »!0 B S « nBp n yp nSp » Y P¼ ¬

(4)

n yw

p n yw n yp * n pw

(5)

dB B

n Bp n pw

dI I

ª dB 1 º ª dY (Y / B) º « B ( I / B) » « Y ( I / B) » ¬ ¼ ¬ ¼

dw w

(6)

(7)

p

where n Bp is the elasticity of domestic demand with respect to price p and n yw is the cross price elasticity that holds price p fixed.

7.A3

Small country trade impact of agri-environmental regulation that increases average variable cost Certain agri-environmental regulations may not directly affect an input but increase the average variable cost of production. In this case, the import or export impact is found as:

C

dY Y

C ( w, y ) my

n yp

m p

(8) (9)

294 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

dI I

dY yB Y ( I / B)

(10)

dE E

dY 1 Y (E / Y )

7.A4

(11)

Product regulation case In the case of product regulations that target both domestic and foreign producers, e.g. pesticide residue and GM crop labelling, the analysis needs to include a second stage that accounts for the additional costs that are imposed on import suppliers. The product regulation is assumed to operate through a restriction on the input which raises its price. However, the input price increase for domestic and foreign producers may vary as captured by the parameter T below.

n pw

ª n p yw ( S / Y )n Sw (Tw / W ) º « »!0 «¬ n yp n Bp ( B / Y ) n Sp ( S / Y ) »¼

(12)

where S = import supply, W = input prices for foreign producers, w = input prices for domestic producers, n SW = import supply elasticity with respect to the price of the regulated input, and T captures the difference in foreign and domestic input price adjustments such that dW Tdw .

n yw

p n yw n yp * n pw

(13)

where Y

Y ( p ( w, W , r ), w, r ) represents domestic supply that is now influenced by W and r is the price of the other composite input used in production. If foreign supply exhibits constant returns to scale, i.e. constant world price, then ( n pw ) is now just the percentage increase in marginal costs of foreign producers for a 1% increase in the price of the regulated input, and equation (12) is not needed. The change in domestic consumption and imports are calculated using equations (6) and (7).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 295

Table 7A1. Parameters used in simulation analysis: small-country regulation and payment cases Australia

Mexico

Wheat

Maize

0.6

0.6

0.6

0.6

0.6

nyp

1

0.72

0.81

0.81

1.82

C n xy

1

1

1

1

1

Y /I

244.63

3.21

59.69

12.61

2.2

Y /E

1.49

284.13

1119.91

15.97

2.76

C / pY

Korea

Japan Rice

Canada Soybeans

Sources: OECD (2001a) and FAOSTAT, 2001. The total variable cost share of revenue, C/pY, parameter could not be found and was estimated based on ranges in other studies, e.g. Larson, et al. (2002).

296 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Table 7A2. Parameters used in simulation analysis: large country regulation case EU

United States

United States

Wheat

Maize

Soybeans

C / pY

0.6

0.695

0.58

n yp

1.75

1.31

1.11

C n xy

1.2

1.05

1

n Bp

-0.33

-0.34

-1.67

n Sp

1.4

1.59

1.72

n Dp

-1

-0.69

-0.76

B /Y

0.94

0.83

0.63

S /Y I /B

5.39

1.54

1.24

0.35

0.00167

0.00256

Sources: FAOSTAT (2001); OECD (2001a 86'$ 7KH &[\UHWXUQVWRVFDOH DQG elasticity parameters could not be found and were estimated based on other analyses.

6SLPSRUWVXSSO\

Table 7A3. Parameters used in simulation analysis: product regulation case EU

United States Maize

0.6

0.695

n yp

2

1.31

C n xy

1.25

1.05

n CXS

1.05

1.25

n Bp

-0.33

-0.34

n Sp

1.31

1.59

B /Y S /Y I /B

1.05

0.83

5.92

0.154

0.26

0.00167

C / pY

Sources: See Table 7A.2.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 297

Notes 1.

A universally accepted definition of “distorted” trade flows does not exist. One interpretation is that trade “distortions” occur when the prices for arable crops do not reflect fully the true social scarcity values of the market and non-market resources used in, or affected by, production. Such distortions occur in environmental regulatory or payment situations where the public agrienvironmental action either imposes controls or offers support with the result that the marginal social costs of environmental protection do not equal the marginal social benefits.

2.

For example, if trade liberalisation leads to an overall increase in the transport of agricultural commodities, and if effective agri-environmental programmes are not in place to control the resulting national and transboundary pollution emissions, then a net welfare gain cannot be assured.

3.

Whether agri-environmental programmes affect production costs or trade is a legitimate policy question. However, the broader social welfare question is: “How can a country maximise welfare from arable crop production, consumption, trade, and agri-environmental management?” If the evaluation extends beyond country borders to transboundary and global effects, then optimal environmental policies are required in those spheres as well.

4.

The term “cost-effective” is used rather than “efficient” because estimating the total economic value (i.e. use, bequest and non-use components) of many environmental effects is beyond the scope of current science and data. Additionally, public environmental policy often specifies a performance target, e.g. hazard reduction, but rarely is it framed to maximise net benefits. However, the validity of arguments remains the same conceptually if “efficient” is substituted for “cost effective”.

298 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

BIBLIOGRAPHY

Armington, P.S. (1969), “The Geographic Pattern of Trade and the Effects of Price Changes,” IMF Staff Papers, XVI, July, pp. 179-199. ABARE [Australian Bureau of Agricultural and Resource Economics] (2003), Australian Grain Industry 2003: Performance and Outlook, Canberra. ABARE (2001), Alternative Policy Approaches to Natural Resource Management: Background Report to the Natural Resource Management Taskforce, February, Canberra. Abler, D.G. and J.S. Shortle (1992), “Environmental and Farm Commodity Policy Linkages in the US and the EC”, European Review of Agricultural Economics, Vol. 19, pp. 197-217. Ackrill, R.W., S.J. Ramsden and J.M. Gibbons (2001), “CAP Reform and the Rebalancing of Support for Cereals and Oilseeds: a Farm-level Analysis”, European Review of Agricultural Economics, Vol. 28, pp. 207-226. Agriculture and Agri-Food Canada [AAFC] (1996), “Farm Resource Management Indicator: Soil Cover and Land Management Component”, Agri-Environmental Indicator Project Report 18, Ottawa. Allen, K. (ed.) (1990), Agricultural Policies in a New Decade, Resources for the Future, Washington, D.C. Alston, J. and P. Pardey (1996), Making Science Pay: The Economics of Agricultural R&D Policy, American Enterprise Institute Press, Washington, D.C. Alvarez-Buylla, E. (2004), “Ecological and Biological Aspects of the Impacts of Transgenic Maize, including Agro-Biodiversity”, February, paper prepared for the Secretariat of the Commission for Environmental Cooperation of North America [CEC] as part of the Article 13 initiative on “Maize and Biodiversity: the Effects of Transgenic Maize in Mexico”.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 299

Anderson, K. and M. Ingco (1999) “Integrating Agriculture into the WTO: the Next Phase”, paper presented at the World Bank’s conference on “Developing Countries and the Millennium Round”, 19-20 September, Geneva. Anderson, K. (1992), “The Standard Welfare Economics of Policies Affecting Trade and the Environment”, pp. 25-48 in K. Anderson and R. Blackhurst (eds), The Greening of World Trade Issues, University of Michigan Press, Ann Arbor, Michigan. Antle, J. and R. Just (1991), “Effects of Commodity Program Structure on Resources and the Environment”, in N. Bockstael and R. Just (eds), Commodity Policy and Resource Use in Agricultural Systems, Springer-Verlag, New York. AUDIT [National Land and Water Resources Audit] (2001), Australian Water Resources Assessment 2000: http://audit.ea.gov.au/ANRA/water/docs/national/water. Babcock, B.A. and D.A. Hennessy (1996), “Input Demand under Yield and Revenue Insurance”, American Journal of Agricultural Economics, Vol. 78, pp. 416-427. Baldock, D., J. Dwyer and J.M. Sumpsi Vinas (2002), Environmental Integration and the CAP, a report to the Directorate-General for Agriculture of the European Commission, IEEP, London. Baldock, D. and K. Mitchell (1995), Cross-Compliance within the Common Agricultural Policy: a Review of Options for Landscape and Nature Conservation, IEEP, London. Barbier, J. and J. Mouret (1998), “Product Quality and Environmental Impacts in Rice Farming”, Cahiers Options Méditerranéennes, Vol. 24, No. 3. Barnard, C., S. Daberkow, M. Padgitt, M. Smith and N. Uri (1997), “Alternative Measures of Pesticide Use”, The Science of the Total Environment, Vol. 203, pp. 229-244. Batie, S. and R. Horan (eds) (2004), The Economics of Agri-Environmental Policy, Vol. I and II, Ashgate Publishing, Aldershot. Batie, S. and A. Sappington (1986), “Cross-compliance as a Soil Conservation Strategy: a Case Study”, American Journal of Agricultural Economics, Vol. 68, pp. 880-885.

300 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Beghin, J.C., D. Roland-Holst and D. van der Mensbrugghe (2002), Global Agricultural Trade and the Doha Round: What are the Implications for North-South?, Working Paper 02-WP 308, June, Center for Agricultural and Rural Development, Iowa State University, Iowa. van den Bergh, J. and J. Holley (2001), “An Environmental-Economic Assessment of Genetic Modification of Agricultural Crops”, Discussion Paper 025/3, Tinbergen Institute, Amsterdam. Bergström, L. and W. Jokela (2001), “Ryegrass Cover Crop Effects on Nitrate Leaching in Spring Barley Fertilised with 15NH415NO3”, Journal of Environmental Quality, Vol. 30, No. 5, pp. 1 659-1 667. den Biggelaar, C., R. Lal, K. Wiebe, H. Eswaran, V. Breneman and P. Reich (2003), “The Global Impact of Soil Erosion on Productivity”, Advances in Agronomy, Vol. 72, May. den Biggelaar, C., R. Lal, K. Wiebe and V. Breneman (2001), “Impact of Soil Erosion on Crop Yields in North America”, Advances in Agronomy, Vol. 72, pp. 1-52. Bloem, J., et al. (1994), “Dynamics of Microorganisms, Microbivores and Nitrogen Mineralization in Winter Wheat Fields under Conventional and Integrated Management”, Agriculture Ecosystems and Environment, Vol. 51, pp. 129-143. Blogowski, A. and N. Pingault (2002), "La réforme de la PAC de 1992: bilan dune décennie d'adaptation des exploitations de grandes cultures", Notes et études économiques, no. 16, avril, pp. 35-54. Boatman, N., et al. (1999), “The Environmental Impact of Arable Crop Production in the European Union: Practical Options for Improvement”, report prepared for the Directorate-General for Environment of the European Commission, Brussels, November. Bouët, A., et al. (2002), “Market Access for GTAP: a Bilateral Measure of Merchandise Trade Protection,” paper presented at the 5th Annual Conference on GEA, 5-7 June, National Tsing Hua University, Chinese Taipei. Brandão, A.S.P., T. Hertel and A. Campos (1992), “The Implications of International Trade Liberalization for the Brazilian Agriculture: a General Equilibrium Analysis”, manuscript, The World Bank, Washington, D.C.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 301

Bromley, D.W. and I. Hodge (1990), “Private Property Rights and Presumptive Policy Entitlements: Reconsidering the Premises of Rural Policy”, European Review of Agricultural Economics, Vol. 17, pp. 197214. Brookes, G. (2003), “The Farm Level Impact of using Bt Maize in Spain”, paper presented at the 7th International Consortium on Agricultural Biotechnology Research – International Conference on Public Goods and Public Policy for Agricultural Biotechnology, 29 June-3 July, Ravello. Brouwer, F. and D. Ervin (eds) (2002), Public Concerns, Environmental Standards and Agricultural Trade, CABI Publishing, Oxford. Brouwer, F, J. Dwyer and D. Baldock (2002), “Public Concerns and the Regulation of Agricultural Production: European Union”, in F. Brouwer and D. Ervin (eds). Brouwer, F. and S. van Berkum (1996), CAP and the Environment in the European Union, Wageningen Pers, Wageningen. Buller, H., G.A. Wilson and A. Höll (eds) (2000), Agri-environmental Policy in the European Union, Ashgate Publishing, Aldershot. Burrell, A. (1989), “The Demand for Fertilisers in the United Kingdom”, Journal of Agricultural Economics, Vol. 40, pp. 1-20. Carpentier, C.L. (2002), “Trade Liberalization Impacts on Agriculture: Predicted vs. Realized”, draft paper distributed at the Meeting of the Commission for Environmental Cooperation, Montreal, 17-18 January. Carpentier, C.L. and D. Ervin (2002), “Public Concerns and the Regulation of Agricultural Production: USA”, in F. Brouwer and D. Ervin (eds). Cason, T.N., L. Gangadharan and C. Duke (2003), “A Laboratory Study of Auctions for Reducing Non-point Source Pollution”, Journal of Environmental Economics and Management, Vol. 46, pp. 446-471. Caswell, M., et al. (2001), “Adoption of Agricultural Production Practices: Lessons Learned from the US”, Department of Agriculture Area Studies Project, Report No. 792, Economic Research Service (ERS)/USDA, Washington, D.C. Christensen, L. (2002), “Soil, Nutrient, and Water Management Systems used in U.S. Corn Production”, Agriculture Information Bulletin No. 774, April, ERS/USDA, Washington, D.C.

302 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Christensen, T. and H. Rygnestad (2000), “Environmental Cross Compliance: Topics for Future Research”, Working Paper No. 1/2000, SJFI, Copenhagen. Claassen, R., K. Wiebe and L. Hansen (2004a), “Farmers’ Choices and the Role of Environmental Indicators”, in R. Francaviglia (ed.). Claassen, R., V. Breneman, S. Bucholtz, A. Cattaneo, R. Johansson and M. Morehart (2004b), “Environmental Compliance in U.S. Agricultural Policy: Past Performance and Future Potential”, Agricultural Economic Report No. 832, ERS/USDA, June, available at: www.ers.usda.gov/publications/aer832/. Claassen, R., et al. (2001), “Agri-Environmental Policy at the Crossroads: Guideposts on a Changing Landscape”, Agricultural Report No. 794, ERS/USDA, Washington, D.C. Claassen, R. and R.D. Horan (2000), “Environmental Payments to Farmers: Issues of Program Design,” Agricultural Outlook, June-July, pp. 15-18. Cooper, J., R. Johansson and M. Peters (2003), “Some Domestic Environmental Effects of U.S. Agricultural Adjustments under Liberalized Trade: a Preliminary Analysis”, paper presented at the International Agricultural Trade Research Consortium’s [IATRC] conference on Agricultural Policy Reform and the WTO: Where are we Heading?, 23-26 June, Capri. Cooper, J., M. Peters, and R. Claassen (2003), “Effects of Agri-Environmental Payment Policies on Agricultural Trade”. paper presented at the Annual Meeting of the American Agricultural Economics Association, 27-30 July, Montreal. Croll, T. and R. Hayes (1988), “Nitrate and Water Supplies in the United Kingdom”, Environmental Pollution, Vol. 50, pp. 163-187. Crosson, P. (2004), “The Economics of Soil Erosion and Maintaining Soil Biodiversity”, in R. Francaviglia (ed.). Curey, M., D. Sumner and R. Howitt (2000), “The Value of Tradable Credits for Rice Straw Burning”, AIC Issues Brief No. 12, May, University of California. DEFRA [Department of Environment, Food and Rural Affairs, United Kingdom] (2003), Farm-Scale Evaluation Results: www.defra.gov.uk/environment/gm/fse.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 303

DEFRA (2002), The Effectiveness of Changing Farmer Practice on the Reduction of Diffuse Pollution of Nitrogen and Phosphorus, London. DEFRA (1997), Survey of Irrigation of Outdoor Crops in 1995: www.defra.gov.uk/esg/work_htm/publications/cs/census/analyses/For_19 97/irrig_95.pdf. Derpsch, R. (2000), Frontiers in Conservation Tillage and Advances in Conservation Practice, Food and Agriculture Organization of the United Nations [FAO], Rome. Diakosavvas, D. (2004), “The Uruguay Round Agreement on Agriculture in Practice: How Open Are OECD Markets?”, in M. Ingco and L.A. Winters (eds), Agriculture and the New Trade Agenda, Cambridge University Press, Cambridge. Diakosavvas, D. (2003), “Production and Trade Effects of the WTO Green Box Agri-environmental Policies in OECD Countries: an Empirical Investigation”, paper presented at the IATRC conference on Agricultural Policy Reform and the WTO: Where are we Heading?, 23-26 June, Capri. Diakosavvas, D. (1994), “The Impact of Environmental Policies on Agricultural Trade”, Journal of International Development, Vol. 6, No. 2, pp. 207-218. Dimaranan, B. and R. McDougall (2004), “GTAP Database Version 6.0”, GTAP Centre, Purdue University, West Lafayette, Indiana, April. Dimaranan, B., T. Hertel and R. Keeney (2003), “OECD Domestic Support and the Developing Countries”, paper presented at the 6th Annual Conference on GEA, 12-14 June, Scheveningen, The Hague. Dimitri, C. and C. Greene (2002), Recent Growth Patterns in the US Organic Foods Market, AIB-777, ERS/USDA, Washington, D.C. Drake, L., P. Bergström and H. Svedsäter (1999), “Farmers’ Attitudes and Uptake”, Chapter 5 in G. van Huylenbroeck and M. Whitby (eds). Dwyer, J., D. Baldock and S. Einschütz (2000), Cross-compliance under the Common Agricultural Policy, a Report to the Department of Environment, Transport and the Regions [DETR], IEEP, London. Dyer-Leal, G. and A. Yúnez-Naude (2003), “NAFTA and Conservation of Maize Diversity in Mexico”, paper presented at the 2nd North American

304 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Symposium of CEC on Assessing the Environmental Effects of Trade, 14 February, Mexico City. EC [European Commission] (2003a), “CAP Reform Summary”, accessed in July: www.europa.eu.int/comm/agriculture/capreform/index_en.htm. EC (2003b), “Reform of the Common Agricultural Policy: a Long-Term Perspective for Sustainable Agriculture – Impact Analysis”, EC, Brussels, March. EC (1998), “Evaluation of Agri-environmental Programmes”, Commission Working Document VI/7655/98, D-G 6, EC, Brussels. EC (1997), “Towards a Common Agricultural and Rural Policy for Europe”, Reports and Studies, Vol. 5, Directorate-General for Economic and Financial Affairs, EC, Brussels. EEA [European Environment Agency] (2003a), “Assessment and Reporting on Soil Erosion. Background and Workshop Report”, Technical Report No. 94/2003, EEA, Copenhagen. EEA (2003b), Europe’s Environment: the Third Assessment, EEA, Copenhagen. Elbehri, A. and K. Pearson (2000), “Implementing Bilateral Tariff Rate Quotas in GTAP using GEMPACK,” GTAP Technical Paper No. 18, Department of Agricultural Economics, Purdue University. Elbehri, A., M. Ingco, T. Hertel and K. Pearson (1999), “Agriculture and WTO 2000: Quantitative Assessment of Multilateral Liberalization of Agricultural Policies”, paper presented at the Conference on Agriculture and the New Trade Agenda in the WTO 2000 Negotiations, 1-2 October, Geneva. Engel, K., T. Frenzel and A. Miller (2002), “Current and Future Benefits from the use of GM Technology in Food Production”, Toxicology Letters, No. 127, pp. 329-336. ENS [Environmental News Service] (2004), “California Farmers Struggle with New Rules on Runoff Pollution” (accessed 29 April): www.enn.com/news/2004-04-29/s_23295.asp. Ervin, D. (1999), “Toward GATT-Proofing Environmental Programmes for Agriculture,” Journal of World Trade, Vol. 33, No. 2, pp. 63-82.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 305

Ervin, D. (1998), “Shaping a Smarter Environmental Policy for Farming”, Issues in Science and Technology, Vol. XIV, No. 4, pp. 73-79. Ervin, D. and R. Welsh (2005), “Environmental Effects of Genetically Modified Crops: a Differentiated Risk Assessment Model”, in J. Wesseler (ed.), Environmental Costs and Benefits of Transgenic Crops in Europe: Implications for Research, Production, and Consumption, Springel Publishing, Dordrecht. Ervin, D. and G. Fox (2004), “The Environmental Effects of Trade: Have We Been Asking the Right Questions?”, Working Paper, Portland State University. Ervin, D., J. Kahn and M. Livingston (eds), (2003), Does Environmental Policy Work: The Theory and Practice of Outcomes Assessment, Cheltenham, Edward Elgar. Ervin, D., et al. (2000), Transgenic Crops: an Environmental Assessment, Henry A. Wallace Center for Agricultural and Environmental Policy, Winrock, Arlington, November. Ervin, D., et al. (1998), “Agriculture and the Environment: a New Strategic Vision”, Environment, Vol. 40, No. 6, pp. 8-15 and 35-40. Eswaran, H. and P. Reich (2001), World Soil Resources Map Index, Natural Resources Conservation Service, USDA, Washington, D.C.: www.nrcs.usda.gov/technical/worldsoils/mapindx. Evans, R. (1996), Soil Erosion and its Impact in England and Wales, Friends of the Earth, London. Faeth, P. (1993), “Evaluating Agricultural Policy and the Sustainability of Production Systems: an Economic Framework”, Journal of Soil and Water Conservation, March/April, pp. 94-99. Falconer, K. and C. Saunders (2002), “Transaction Costs for SSSIs and Policy Design”, Land Use Policy, Vol. 19, pp. 157-166. Falconer, K. and M. Whitby (1999), “The Invisible Costs of Scheme Implementation and Administration”, Chapter 4 in G. van Huylenbroeck and M. Whitby (eds) (1999). FAO [Food and Agriculture Organization of the United Nations] (2003), World Agriculture Towards 2015/2030, FAO, Rome.

306 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

FAO/WHO [World Health Organization] (1999), Guidelines for the Production, Processing, Labelling and Marketing of Organically Produced Foods, Codex Alimentarius Commission, CAC/GL 32-1999, FAO/WHO, Rome. Fernadez-Cornejo, J. and W. McBride (2002), “Adoption of Bioengineered Crops”, Agricultural Economic Report, No. 810, ERS/USDA, Washington, D.C. Foreman, L. (2001), “Characteristics and Production Costs of U.S. Corn Farms”, Statistical Bulletin No. 974, ERS/USDA, Washington, D.C. Foster, M. (2003), “GM Canola: What are the Economics under Australian Conditions?”, Australian Grains Industry, ABARE, Canberra. Foster, C. and N. Lampkin (2000), Organic and In-conversion Land Area, Holdings, Livestock and Crop Production in Europe, FAIR3-CT961794, October, Welsh Institute of Rural Studies, Aberystwyth University. Fox, G. and J. Kidon (2002), chapter on “Canada” in F. Brouwer and D. Ervin (eds), pp. 141-179. Francaviglia, R. (ed.) (2004), Agricultural Impacts on Soil Erosion and Soil Biodiversity: Developing Indicators for Policy Analysis – Proceedings from an OECD Expert Meeting, Rome, Italy, March 2003, Istituto Sperimentale per la Nutrizione della Pianta, Rome. Francois, J., B. McDonald and H. Nordstrom (1996), “The Uruguay Round: a Numerically-based Qualitative Assessment”, Chapter 9 in W. Martin and L.A. Winters (eds). Frandsen, S.E., H.G. Jensen, W. Yu and A. Walter-Jørgensen, (2001), “Modelling the EU sugar policy – a preliminary study of policy reform scenarios,” paper prepared for the 77th EAAE Seminar/NJF Seminar No. 325, 17-18 August, Helsinki. Fraser, I.M. (1995), “An Analysis of Management Agreement Bargaining under Asymmetric Information”, Journal of Agricultural Economics, Vol. 46, pp. 20-32. Frink, C., P. Waggoner and J. Ausubel (1998), “Nitrogen Fertiliser: Retrospect and Prospect”, paper presented at the National Academy of Science of the United States of America [NAS] Colloquium on “Plants and Population: Is there Time?”, 5-6 December, Irvine, California.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 307

Fuglie, K.O. and D.J. Bosch (1995), “Economic and Environmental Implications of Soil Nitrogen Testing: a Switching Regression Analysis”, American Journal of Agricultural Economics, Vol. 77, pp. 891-900. Fukuda, H., J. Dyck and J. Stout (2003), Rice Sector Policies in Japan, Report 0303-01, March, ERS/USDA, Washington, D.C. Gianessi, L., S. Sankula and N. Reigner (2003), “Plant Biotechnology: Potential Impact for Improving Pest Management in European Agriculture: a Summary of Nine Case Studies”, National Center for Food and Agricultural Policy [NCFAP], December, Washington, D.C. (for the full report, see www.ncfap.org). Gren, I.M. (1994), “Regulating Farmers’ Use of Pesticides in Sweden”, Chapter 8 in J.B. Opschoor and R.K. Turner (eds), Economic Incentives and Environmental Policies, Principles and Practice, Kluwer Academic Publishers, Dordrecht, pp. 153-173. Grote, U., C. Deblitz, T. Reichert and S. Stegmann (2000), “Umweltstandards und internationale Wettbewerbsfähigkeit: Analyse und Bedeutung – insbesondere im Rahmen der WTO”, study prepared for the Ministry of Food, Agriculture and Forestry, Bonn. Guyomard, H., M. Baudry and A. Carpentier, (1996), “Estimating Crop Supply Response in the Presence of Farm Programmes: Application to the CAP”, European Review of Agricultural Economics, Vol. 23, pp. 401-420. Hanley, N. and M. Whitby (2003), “Alternative Criteria for Judging the Success of Agro-environmental Policy in the UK”, pp. 145-168 in D. Ervin, J. Kahn and M. Livingston (eds). Harrison, W.J. and K.R. Pearson (2002), An Introduction to GEMPACK, Release 8.0, GEMPACK Document No. GPD-1, Sixth Edition, Monash University, October. Harrison, G.W., T.F. Rutherford and D.G. Tarr (1996), “Quantifying the Uruguay Round”, Chapter 8 in W. Martin and L.A. Winters (eds). Hayami, Y. and V.W. Ruttan (1970), “Factor Prices and Technical Change in Agricultural Development: The United States and Japan, 1880-1960,” Journal of Political Economy, Vol. 78, No. 5, pp. 1 115-1 141. Heath, M. and M. Rayment (2003), “Using Data to Develop Biodiversity Indicators for Agriculture”, in OECD (2003c).

308 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Heigh, L. and B. Junkins (2005), Quantitative Analysis of the Impact of Agricultural Management Strategies on Environmental Indicators, AAFC, Ottawa. Heimlich, R. (1991), “Soil Erosion and Conservation Policies in the United States”, in N. Hanley (ed.) (1991), Farming and the Countryside: an Economic Analysis of External Costs and Benefits, CABI Publishing, Oxford. Hennessy, D.A. (1998), “The Production Effects of Agricultural Income Support Policies under Uncertainty”, American Journal of Agricultural Economics, Vol. 80, pp. 46-57. Hertel, T., D. Hummels, M. Ivanic, and R. Keeney (2003), “How Confident Can We Be in CGE-based Assessments of Free Trade Agreements?”, GTAP Working Paper No. 26. Hertel, T. (ed.) (1997), Global Trade Analysis: Modeling and Applications, Cambridge University Press, New York. Hertel, T., W. Martin, K. Yanagishima and B. Dimaranan (1996), “Liberalizing Manufactures Trade in a Changing World Economy”, Chapter 7 in W. Martin and L.A. Winters (eds). Hertel, T., K. Stiegert and H. Vroomen (1996), “Nitrogen-land Substitution in Corn Production: a Reconciliation of Aggregate and Firm-level Evidence”, American Journal of Agricultural Economics Vol. 78, pp. 30-40. Hertel, T. and M. Tsigas (1987), “General Equilibrium Analysis of Supply Control in U.S. Agriculture”, Staff Paper No. 87-25, Department of Agricultural Economics, Purdue University, Indiana. Hirad, S., R. Nelson, N. Andrews and I. Shaw (2003), “Comparing the Effects of Different Approaches to Liberalising World Grain Markets”, paper presented at the 47th Annual Conference of the Australian Agricultural and Resource Economics Society, 12-14 February, Fremantle. Hodge, I. (2000), “Agri-environmental Relationships,” The World Economy, Vol. 23, No. 2, pp. 257-274. Hofer, E. (2000), “The Swiss Approach to Environmental Protection”, in J.E. Petersen and K. Shaw (eds).

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 309

Hoffmann, M., et al. (2000), “Leaching of Nitrogen in Swedish Agriculture: a Historical Perspective”, Agriculture Ecosystems and Environment, Vol. 80, pp. 277-290. Hong-Sang, K. (2004), “Irrigation Development and Water Management Systems in Korea”, in R. Francaviglia (ed.). Hoppe, R., et al. (2001), Structural and Financial Characteristics of U.S. Farms: 2001 Family Farm Report, Report No. 768, ERS/USDA, Washington, D.C. Horan, R.D., M. Ribaud and D.G. Abler (2001), “Voluntary and Indirect Approaches for Reducing Externalities and Satisfying Multiple Objectives”, Chapter 3 in J.S. Shortle and D.G. Abler (eds), Environmental Policies for Agricultural Pollution Control, CABI Publishing, Oxford. Horan, R.D., J.S. Shortle and D.G. Abler, (1999), “Green Payments for Nonpoint Pollution Control”, American Journal of Agricultural Economics, Vol. 81, pp. 1 210-1 215. Horner, G.L., J. Corman, R.E. Howitt, C.A. Carter and R.J. MacGregor (1992), “The Canadian Regional Agricultural Model: Structure, Operation and Development”, Policy Branch Technical Report 1/92, Agriculture Canada, Ottawa. House, R., H. McDowell, M. Peters and R. Heimlich (1999), “Agriculture Sector Resource and Environmental Policy Analysis: An Economic and Biophysical Approach,” in G. R. Bock and J. A. Goode (eds), Environmental Statistics: Analyzing Data for Environmental Policy, Novartis Foundation, London. Howitt, R. (1995), “Positive Mathematical Programming,” American Journal of Agricultural Economics, Vol. 77, No. 329-342. Hrubovcak, J., et al. (1999), Green Technologies for a More Sustainable Agriculture, Report No. 752, ERS/USDA, Washington, D.C. Hur, S. et al. (2004), “Research on the Impact of Soil Erosion on Agricultural Lands in Korea”, in R. Francaviglia. (ed.). van Huylenbroeck, G. and M. Whitby (eds) (1999), Countryside Stewardship: Farmers, Policies and Markets, Elsevier Science Ltd., Amsterdam.

310 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

IEEP [Institute for European Environmental Policy, United Kingdom], et al. (2000), The Environmental Impacts of Irrigation in the European Union, a report to the Directorate-General for Environment of the EC, Brussels, March. IFA/IFDC/FAO [International Fertilizer Industry Association/International Fertilizer Development Center/FAO], Fertiliser Use by Crop, various issues, Rome, available on-line at: www.fertilizer.org/ifa/statistics/crops/fubc5ed.pdf. Intergovernmental Panel on Climate Change (IPCC) (1996), Climate Change 1995: Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses, Cambridge University Press, Cambridge. James, C. (2003), “Global Status of Commercialised Transgenic Crops: 2003”, ISAAA Briefs No. 30, ISAAA, Ithaca, New York: www.agbiotechnet.com. Jensen, H.G. and C.P. Nielsen (2004), “EU dairy policy analysis: Assessing the importance of quota rent estimates,” paper prepared for the 7th Annual Conference on Global Economic Analysis, 17-19 June, Washington, D.C. Jensen, H.G. and S. Frandsen (2003), Impact of the Eastern European Accession and the 2003 Reform of the CAP: Consequences for Individual Member Countries, Danish Research Institute of Food Economics (FOI) Frederiksberg. Kellogg, R., C. Lander, D. Moffitt and N. Gollehon (2000), Manure Nutrients Relative to the Capacity of Cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal Trends for the United States, ERS/USDA, Washington, D.C. Kislev, Y. and W. Peterson (1982), “Prices, Technology, and Farm Size”, Journal of Political Economy, Vol. 90, No. 3, pp. 578-595. Klonsky, K. and F. Jacquet (2003), “How Well do Green Payments Fit into the Green Box? Lessons from EQIP”, paper presented at the IATRC conference on Agricultural Policy Reform and the WTO: Where are we Heading?, 23-26 June, Capri. Komen, H.C.M. and J.H.M. Peerlings (2001), “Endogenous Technology Switches in Dutch Dairy Farming under Environmental Restrictions”, European Review of Agricultural Economics, Vol. 28, pp. 117-142. Koroluk, R., D. Culver, A. Lefebvre and T. McRae (2000), “Management of Farm Nutrient and Pesticide Inputs,” Chapter 5 in T. McRae, C.A.S. Smith et L.J. Gregorich (eds). AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 311

Kulshreshtha, S., G. Ravinderpal, B. Junkins, R. Desjardins, M. Boehm and M. Bonneau (2002), "Canadian Economic and Emissions Model for Agriculture (CEEEA 2.0): Technical Documentation", CSALE Working Paper 12, University of Saskatchewan, Saskatoon. LaFrance, J.T. (1992), “Do Increased Commodity Prices Lead to More or Less Soil Degradation?”, Australian Journal of Agricultural Economics, Vol. 36, pp. 57-82. Larson, B. (2000) “Specific Environmental Regulations and Exports”, Land Economics, Vol. 76, pp. 534-549. Larson, B. and S. Scatasta (2003), “Modeling the Impacts of Environmental Policies on Agricultural Imports,” Working Paper, Department of Agricultural and Resource Economics, University of Connecticut, Storrs. Larson, B., et al. (2002), “The Impact of Environmental Regulations on Exports: Case Study Results from Cyprus, Jordan, Morocco, Syria, Tunisia, and Turkey”, World Development Vol. 30, No. 6, pp. 1 0571 072. Latacz-Lohmann, U. and C.P.C.M. van der Hamsvoort (1999), “A Theoretical Analysis of Environmental Cross Compliance within the Common Agricultural Policy”, paper presented at the AES Annual Conference, 24-26 March, Belfast. Latacz-Lohmann, U. and C.P.C.M. van der Hamsvoort (1997), “Auctioning Conservation Contracts: a Theoretical Analysis and an Application”, American Journal of Agricultural Economics, Vol. 79, pp. 407-18. Lehtonen, H., J. Aakkula and P. Rikkonen (2004), “Alternative Agricultural Policy Scenarios, Sector Modelling and Indicators: a Sustainability Assessment”, Discussion Paper DP 2004/1, MTT Economic Research, Helsinki. Lewandrowski, J., J. Tobey and Z. Cook (1997), “The Interface between Agricultural Assistance and the Environment: Chemical Fertiliser Consumption and Area Expansion,” Land Economics, Vol. 73, pp. 404-427. Lin, W., et al. (2000), “Supply Response Under the 1996 Farm Act and Implications for the U.S. Field Crops Sector”, Technical Bulletin No. 1 888, ERS/USDA, Washington, D.C.

312 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Lips, M. (2004), The CAP Mid Term Review and the WTO Doha Round: Analyses for the Netherlands, EU and Accession Countries, LEI, Report 6.04.03, February, The Hague. Livezey, J. and L. Foreman (2004), “Characteristics and Production Costs of US Rice Farms”, Statistical Bulletin, No. 974-7, March, ERS/USDA, Washington, D.C. Loucks, R. (2003), “Displacement of Fertilizers and Pesticides in Wheat Culture in North America”, March, paper prepared for Environmental Cooperation of North America. Lu, H., et al. (2003), “Predicting Sheetwash and Rill Erosion over the Australian Continent”, Australian Journal of Soil Research, Vol. 41, No. 6, pp. 1 037-1 062. Lubowski, R., A. Plantinga and R. Stavins (2004), “What Drives Land Use Change in the United States?”, Working Paper, Resources for the Future, Washington, D.C. Malik, A. (1993), “Self-reporting and the Design of Policies for Regulating Stochastic Pollution”, Journal of Environmental Economics and Management, Vol. 24, pp. 241-257. Martin, W. and L.A. Winters (eds) (1996), The Uruguay Round and the Developing Countries, Cambridge University Press, New York. McHugh, M. (2004), “Soil Erosion in the UK: Assessing the Impacts and Developing Indicators”, in R. Francaviglia (ed.). McRae, T. and T. Weins (2003), “Using Biological and Land Use Information to Developing Indicators of Habitat Availability of Farmland”, in OECD (2003c). McRae, T., C.A.S. Smith and L.J. Gregorich (eds) (2000), Environmental Sustainability of Canadian Agriculture: Report of the Agri-Environmental Indicator Project, Agriculture and Agri-Food Canada, Ottawa: www.agr.gc.ca/policy/environment/pdfs/aei/fullreport.pdf. Meilke, K.D. and A. Weersink (1990), “The Impact of Support Programs on Crop Area Response,” Canadian Journal of Agricultural Economics, Vol. 38, pp. 871-885. Meissner, R., et al. (1998), “Lysimeter Studies in East Germany Concerning the Influence of Set-aside of Intensive Farmed Land on the Seepage

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 313

Water Quality”, Agriculture Ecosystems and Environment, Vol. 67, pp. 161-173. Miettinen, A., R. Hietala-Koivu and H. Lehtonen (2004), “On Diversity Effects of Alternative Agricultural Policy Reforms in Finland: an Agricultural Sector Modelling Approach”, Agricultural and Food Science, Vol. 13, No. 3, pp. 229-46. Mims, A.M., P.A. Duffy and G. Young (1989), “Effects of Alternative Acreage Restriction Provisions on Alabama Cotton Farms”, Southern Journal of Agricultural Economics, December, pp. 85-94. Mitchell, G., R. Griggs, V. Benson and J. Williams (1998), Environmental Policy Integrated Climate Model, Washington, D.C., available at: www.brc.tqmus.edu/epic (accessed in September). Moro, D. and P. Sckokai (1999), “Modelling the CAP Arable Crop Regime in Italy: Degree of Decoupling and Impact of Agenda 2000”, Cahiers d’Economie et Sociologie Rurales, No. 53, pp. 49-74. Morris, C. and M. Winter (1999), “Integrated Farming Systems: the Third Way for European Agriculture?” Land Use Policy, Vol. 16, pp. 193-205. MTT [Economic Research] (2003), Finnish Agriculture and Rural Industries 2003, Agri-food Research Finland, Helsinki. Nakashima, Y. (2001), “Multifunctionality: Applying the OECD Framework: a Review of Literature in Japan”, paper presented at the OECD workshop on: Multifunctionality: Applying the OECD Analytical Framework, 2-3 July, Paris. Nelson, G. and A. Pinto (2001), “GMO Adoption and Non-market Effects” in G. Nelson (ed.), Genetically Modified Organisms in Agriculture: Economics and Politics, Academic Press, London. Nelson, G. and A. Pinto (1999), “GMO Adoption and Non-market Effects” in G. Nelson, T.E. Josling, D. Bullock, L. Unnevehr, M. Rosegrant and L. Hill., The Economics and Politics of Genetically Modified Organisms in Agriculture: Implications for WTO, Bulletin 809, November, University of Illinois at Urbana-Champaign. NRC [National Research Council] (2002 and 2003), Environmental Effects of Transgenic Plants: the Scope and Adequacy of Regulation, National Academy Press, Washington, D.C.

314 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

NRCS [Natural Resources Conservation Service] (2003), National Resources Inventory: 2001 Annual NRI, July, Washington, D.C. OECD (2004a), Agriculture, Trade and the Environment: The Dairy Sector, OECD, Paris. OECD (2004b), Risk Effects of PSE Crop Measures, OECD, Paris. OECD (2004c), Analysis of the 2003 CAP Reform, OECD, Paris. OECD (2004d), Biomass and Agriculture: Sustainability, Markets and Policies, OECD, Paris. OECD (2003a), Organic Agriculture: Sustainability, Markets and Policies, OECD, Paris. OECD (2003b), OECD Nitrogen Balance Database, OECD, Paris: www.oecd.org/agr/env/indicators.htm. OECD (2003c), Agriculture and Biodiversity: Developing Indicators for Policy Analysis, Proceedings from an OECD Expert Meeting, Zurich, Switzerland, November 2001, OECD, Paris. OECD (2003d), Agricultural Policies in OECD Countries: Monitoring and Evaluation 2003, OECD, Paris. OECD (2003e), Agri-environmental Policy Measures: Overview of Developments, OECD, Paris. OECD (2003f), Agriculture, Trade and the Environment: The Pig Sector, OECD, Paris. OECD (2003g), OECD Pesticide Use Data, OECD, Paris. OECD (2002). Risk Related Non-price Effects of the CAP Arable Crop Regime: Results from a FADN Sample, AGR/CA/APM(2002)14/FINAL, OECD, Paris. OECD (2001a), Environmental Indicators for Agriculture: Methods and Results, Volume 3, OECD, Paris. OECD (2001b), Adoption of Technologies for Sustainable Farming Systems: Wageningen Workshop Proceedings, OECD, Paris. OECD (2001c), Business Approaches to Agri-environmental Management: Incentives, Constraints and Policy Issues, OECD, Paris.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 315

OECD (2001d), Improving the Environmental Performance of Agriculture. Policy Options and Market Approaches, OECD, Paris. OECD (2001e), Market Effects of Crop Support Measures, OECD, Paris. OECD (2001f), The Uruguay Round Agreement on Agriculture: an Evaluation of its Implementation in OECD Countries, OECD, Paris. OECD (2000a), Domestic and International Environmental Impacts of Agricultural Trade Liberalisation, COM/AGR/ENV(2000)75/FINAL, OECD, Paris. OECD (2000b), Assessing the Environmental Effects of Trade Liberalisation Agreements, OECD, Paris. OECD (1998a), Agricultural Policy Reform and the Rural Economy in OECD Countries, OECD, Paris. OECD (1998b), Agricultural Practices that Reduce Greenhouse Gas Emissions: Overview and Results of Survey Instrument, COM/ENV/EPOC/AGR/CA(1998)149/FINAL, OECD, Paris. OECD (1998c), Co-operative Approaches to Sustainable Agriculture, OECD, Paris. OECD (1998d), Agriculture and the Environment: Issues and Policies, OECD, Paris. OECD (1997), The Environmental Effects of Agricultural Land Diversion Schemes, OECD, Paris. OECD (1994), Towards Sustainable Agricultural Production: Cleaner Technologies, OECD, Paris. Orlick, C., L. Bauer and S. Jeffrey (1995), “An Annotated Bibliography of Crop Rotation and Tillage Literature with Special Emphasis on the Canadian Prairies”, Staff Paper 95-08, Department of Rural Economy, University of Alberta. Oude Lansink, A. and J. Peerlings (1996), “Modelling the New Cereals and Oilseeds Regime in the Netherlands”, European Review of Agricultural Economics, Vol. 23, pp. 161-178. Oygarden, L. and A. Gronlund (2004), “Indicators for Soil Erosion in Norway” in R. Francaviglia (ed.).

316 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Ozanne, A., T. Hogan and D. Colman (2001), “Moral Hazard, Risk Aversion and Compliance Monitoring in Agri-environmental Policy”, European Review of Agricultural Economics, Vol. 28, pp. 329-347. Painter, K. and D.L. Young (1994), “Environmental and Economic Impacts of Agricultural Policy Reform: an Interregional Comparison", Journal of Agricultural and Applied Economics, Vol. 26, pp. 451-462. Pearson, C. (2000), Economics and the Global Environment, Cambridge University Press. Petersen, J.E. and K. Shaw (eds) (2000), Environmental Standards in Agriculture: Proceedings of a Pan-European Conference on Meeting of Environmental Standards under Agenda 2000, IEEP, London. Pieri, C., et al. (2002), No-till Farming for Sustainable Rural Development, The World Bank, Washington, D.C. Plano Nacional da Água (2002), Portugal: www.inag.pt/inag2002/port/a_intervencao/planeamento/pna/pna_indice. html. Prasuhn, V. and P. Weisskopf (2004), “Current Approaches to Measure, Monitor and Model Agricultural Soil Erosion in Switzerland” in R. Francaviglia (ed.). Rae, A.N. and A. Strutt (2003), “Doha Proposals for Domestic Support: Assessing the Priorities”, paper presented at the IATRC conference on Agricultural Policy Reform and the WTO: Where are we Heading?, 23-26 June, Capri. Rainelli, P. and D. Vermersch (1997), “Thematic Network on CAP Environment in the EU”, in F. Brouwer and P. Lowe (eds), CAP and the Rural Environment in Transition: a Panorama of National Perspectives, Wageningen Pers, Wageningen. Reichelderfer, K. and W.G. Boggess (1988), “Government Decision Making and Program Performance: the Case of the Conservation Reserve Program”, American Journal of Agricultural Economics, Vol. 70, pp. 1-11. Ribaudo, M., N. Gollehon, M. Aillery, J. Kaplan, R. Johansson, J. Agapoff, L. Chirstensen, V. Breneman and M. Peters (2003), “Manure management for water quality: costs to animal feeding operations of applying manure nutrient to land”, Agricultural Economic Report No. 824, ERS/USDA, Washington D.C.: www.ers.usda.gov/publications/aer824/. AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 317

Roberts, M. and S. Bucholtz (2005), “Slippage in the Conservation Reserve Program or Spurious Correlation? A Comment”, American Journal of Agricultural Economics, Vol. 87, No. 1, pp. 244-250. Roberts, D., T.E. Josling and D. Orden (1999), A Framework for Analysing Technical Trade Barriers in Agricultural Markets, ERS/USDA, Washington, D.C. Robertson, G., et al. (2003), “Rethinking the Vision for Environmental Research for U.S. Agriculture,” Bioscience, December. Röling, N. and M.A.E. Wagemakers (eds) (1998), Facilitating Sustainable Agriculture: Participatory Learning and Adaptive Management in Times of Environmental Uncertainty, Cambridge University Press, Cambridge. Röling, N. and J. Jiggins (1998), “The Ecological Knowledge System”, in N. Röling and M.A.E. Wagemakers (eds), pp. 283-311. The Royal Society (1998), Genetically Modified Plants for Food Use, Carlton House Terrace, London. Salhofer, K. (2000), “Elasticities of Substitution and Factor Supply Elasticities in European Agriculture: a Review of Past Studies”, Discussion Paper No. 83-W-2000, September, Institut für Wirtschaft, Politik und Recht, Universität für Bodenkultur, Vienna. Saunders, C. (1996), “Conservation Covenants in New Zealand”, Land Use Policy, Vol. 13, pp. 325-329. Scott, A. (2003), “Developing Habitat Accounts: an Application of the United Kingdom Countryside Surveys”, in OECD (2003c). Segerson, K. (1988), “Uncertainty and Incentives for Nonpoint Pollution Control”, Journal of Environmental Economics and Management, Vol. 15, pp. 87-98. Serecon Management Consulting and Koch Paul Associates (2001), An Agronomic and Economic Assessment of Transgenic Canola, Report to the Canola Council of Canada, January. Shortle, J.S. and D.G. Abler (1999), “Agriculture and the Environment”, Chapter 11 in J.C.J.M. van den Bergh (ed.) (1999), Handbook of Environmental and Resource Economics, Edward Elgar, Cheltenham, pp. 159-176.

318 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

Sinabell, F. and E. Schmid (2004), “The Effects of the EU’s Common Agricultural Policy Reforms on the Choice of Management Practices”, paper presented at the OECD Experts’ Meeting on Farm Management Indicators and the Environment, 8-12 March, Palmerston North. Smith, P. (2002), “Setting Goals, Making Decisions, and Assessing Outcomes in Conservation Programs Administered by the USDA”, in F. Brouwer and D. Ervin (eds), pp. 130-144. Smith, R., G. Schwarz and R. Alexander (1997), “Regional interpretation of water-quality monitoring data”, Water Resources Research, Vol. 33, No. 2781-2798. Socolow, R. (1998), “Nitrogen Management and the Future of Food: Lessons from Management of Energy and Carbon”, paper presented at the NAS Colloquium, “Plants and Population: is there Time?”, 5-6 December, Irvine, California. Stoate, C., et al. (2001), “Ecological Impacts of Arable Intensification in Europe”, Journal of Environmental Management, Vol. 63, pp. 337-365. Tobey, T. (1991), “The Effects of Environmental Policy towards Agriculture on Trade: Some Considerations”, Food Policy, Vol. 16, No. 2, pp. 90-94. Tobey, J.A. and K.A. Reinert (1991), “The Effects of Domestic Agricultural Policy Reform on Environmental Quality”, Journal of Agricultural Economics Research, Vol. 43, pp. 20-28. van Tongeren, F. (2002), “Forward-looking Analysis of Reforms of EU Dairy Policy”, paper presented at the 10th EAAE Conference, 2831 August, Zaragoza. Torres, G., et al. (2001), “Conservation Agriculture in Europe: Current Status and Perspectives”, paper presented at the 1st World Congress on Conservation Agriculture: a Worldwide Challenge?, European Conservation Agricultural Federation and FAO, 1-5 October, Madrid. van Tran, D. (1998), “World Rice Production: Main Issues and Technical Possibilities”, Cahiers Options Méditerranéennes, Vol. 24, No. 2. Traxler, G. (2003), “The Economic Impacts of Biotechnology-based Technological Innovations”, ESA Working Paper No. 03-XX, FAO, Rome.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 319

Tsigas, M. (2001), “How would Food Markets be Affected by Liberalizing Trade in Processed Foods?”, Economics Working Paper No. 2001-08-A, August, USITC, Washington, D.C. Tsigas, M., D. Gray and T.W. Hertel (2002), “How to Assess the Environmental Impacts of Trade Liberalization”, paper presented to the 5th Annual Conference on GEA, 5-7 June, National Tsing Hua University. UNFCCC [United Nations Framework Convention on Climate Change] (2003), Greenhouse Gas Inventory Database, accessed 29 October: http://ghg.unfccc.int. USDA [United States Department of Agriculture] (2003a), Environmental Quality Incentive Program, Draft Benefit Cost Analysis: www.nrcs.usda.gov/programs/Env_Assess/EQIP/EQIP%/20BCAnal211_ 03.pdf. USDA (2003b), Agricultural Resources and Environmental Indicators, ERS/USDA, Washington, D.C.: www.ers.usda.gov/publications/arei/arei2001. USDA (2003c), “Decoupled Payments: Household Income Transfers in Contemporary U.S. Agriculture”, Agricultural Economic Report No. 822, USDA, Washington, D.C. USDA (2003d), Prospective Plantings, National Agricultural Statistics Service, Agricultural Statistics Board, USDA, Washington, D.C.: http://usda.mannlib.cornell.edu/reports/nassr/field/pcp-bbp/pspl0303.txt. USDA (2003e), USDA Agricultural Baseline Projections to 2010, World Agricultural Outlook Board, Staff Report WAOB-2001-1, Office of the Chief Economist, Washington, D.C. USDA (2001), Commodity Costs and Returns, ERS/USDA, Washington, D.C. USDA (2000), Agricultural Chemical Usage: 1999 Field Crops Summary, ERS/USDA, Washington, D.C. USDA (1999a), Genetically Engineered Crops for Pest Management, ERS/USDA, Washington, D.C.: www.econ.ag.gov/whatsnew/issues/biotech. USDA (1999b), Impacts of Adopting Genetically Engineered Crops in the U.S. – Preliminary Results, ERS/USDA, Washington, D.C.: www.ers.usda.gov/emphases/harmony/issues/genengcrops/genengcrops. htm.

320 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

USDA (1997), Census of Agriculture – 1997, National Agricultural Statistical Service (NASS), Washington, D.C.: www.nass.usda.gov/census. USITC [United States International Trade Commission] (2002), The Economic Effects of Significant U.S. Import Restraints: Third Update, June, Washington, D.C. Vasavada, U. and W. Nimon (2003), Environmental Effects of Further Trade Liberalization in Agriculture, ERS/USDA, Washington, D.C.: www.ers.usda.gov/briefing/wto/PDF/environmentandtradeliberalization. pdf. Vasavada, U., S. Warmerdam and W. Nimon (2001), Green Box Policies and the Environment, ERS/USDA, Washington, D.C.: www.ers.usda.gov/briefing/wto/environm.htm. Vatn, A. L.R. Bakken, H. Lundeby, E. Romstad, P.K. Rørstad, A. Vold and P. Botterweg (1997), “Regulating Nonpoint-source Pollution from Agriculture: an Integrated Modelling Analysis”, European Review of Agricultural Economics, Vol. 24, pp. 207-229. Wailes, J. (2004), “Implications of the WTO Doha Round for the Rice Sector”, paper presented at the FAO conference: Rice in Global Markets and Sustainable Production Systems, 12-13 February, Rome. Wailes J., et al. (1991), Japan’s Rice Market: Policies and Prospects for Trade liberalisation, Arkansas. Walford, N. (2002), “Agricultural Adjustment: Adoption of and Adaptation to Policy Reform Measures by Large-scale Commercial Farmers”, Land Use Policy, Vol. 19, pp. 243-257. Westcott, P., E. Young and M. Price (2002), “The 2002 Farm Act: Provisions and Implications for Commodity Markets”, Agriculture Information Bulletin No. AIB778, ERS/USDA, Washington, D.C. Wetterich, F. (2003), “Biological Diversity of Livestock and Crops: Useful Classification and Appropriate Agri-environmental Indicators”, in OECD (2003c). Wilkinson, D. (1998), “The Commercial Effect of Environmental & Animal Welfare Regulations on UK Agriculture” in Is UK Agricultural Competitive? A European Perspective, National Farmers’ Union, London.

AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

– 321

Winter, M. and P. Gaskell (1998), “The Agenda 2000 Debate and CAP Reform in Great Britain: Is the Environment being Sidelined?”, Land Use Policy, Vol. 15, pp. 217-231. Wolfenbarger, L. and P. Phifer (2000), “The Ecological Risks and Benefits of Genetically Engineered Plants”, Science, Vol. 290, pp. 2 088-2 093. WTO [World Trade Organization] (2002), Tariff and Other Quotas, Background Paper by the Secretariat, Committee for Agriculture, Special Session, TN/AG/S/5, 21 March, Geneva. Wu, J. (2000), “Slippage Effects of the Conservation Reserve Program,” American Journal of Agricultural Economics, Vol. 82, No. 4, pp. 979-992. Wu, J. and K. Segerson (1995), “The Impact of Policies and Land Characteristics on Potential Groundwater Pollution”, American Journal of Agricultural Economics, Vol. 77, pp. 1 033-1 047. Yadav, S.N., et al. (1997), “Do Farmers Overuse Nitrogen Fertilizer to the Detriment of the Environment?”, Environmental and Resource Economics, Vol. 9, pp. 323-340. Yagi, K. (1997), “Methane Emissions from Paddy Fields”, Bulletin of the National Institute of Agro-Environmental Sciences, No. 14, Tsukuba, pp. 96-207. Young, P. (1986), “Nitrate in Groundwater and the Effects of Ploughing on Release of Nitrate” in J.F.L.G. Solbe (ed.) (1986), Effects of Land Use on Fresh Waters, Ellis Horwood, Chichester, pp. 221-237. Yussefi, M. and H. Willer (eds) (2003), The World of Organic Agriculture 2003: Statistics and Future Prospects, 5th revised edition, International Federation of Organic Agriculture Movements [IFOAM], TholeyTheley.

322 – AGRICULTURE, TRADE AND THE ENVIRONMENT – THE ARABLE CROP SECTOR – ISBN-92-64-00996-5 © OECD 2005

caecilia_histo_gen_a_16x23.fm Page 3 Thursday, June 23, 2005 9:44 AM

0&$%16#-*$"5*0/4 SVF"OESÏ1BTDBM 1"3*4$&%&9 13*/5&%*/'3"/$& 1 *4#/o/P