OECD Studies on Environmental Innovation Environmental Policy, Technological Innovation and Patents (OECD Studies on Environmental Innovation)

Environmental Policy, Technological Innovation and Patents Technological innovation can help realise environmental objectives in a less costly manner than would otherwise be the case. Thus, understanding the role that technological innovation can play in achieving environmental objectives is important for policy debates. However, the relationship between environmental policy and technological innovation remains an area in which empirical evidence is scant. In an attempt to bridge this gap, the OECD has examined the relevant issues, using patent activity as a measure of technological innovation.

The full text of this book is available on line via these links: www.sourceoecd.org/environment/9789264046818 www.sourceoecd.org/scienceIT/9789264046818 Those with access to all OECD books on line should use this link: www.sourceoecd.org/9789264046818 SourceOECD is the OECD online library of books, periodicals and statistical databases. For more information about this award-winning service and free trials, ask your librarian, or write to us at [email protected].

ISBN 978-92-64-04681-8 97 2008 06 1 P

XXXPFDEPSHQVCMJTIJOH

-:HSTCQE=UY[]V]:

Environmental Policy, Technological Innovation and Patents

Three case studies have been undertaken: abatement technologies for wastewater effluent from pulp production, abatement of motor vehicle emissions, and development of renewable energy technologies. On the basis of patent data, the nature, extent, and causes of innovation in each of these areas have been explored. While a particular focus has been placed on the role of environmental policy in bringing about the innovation documented, it is recognised that other factors play a key role in inducing innovation that has positive environmental implications.

OECD Studies on Environmental Innovation

OECD Studies on Environmental Innovation

OECD Studies on Environmental Innovation

Environmental Policy, Technological Innovation and Patents

OECD Studies on Environmental Innovation

Environmental Policy, Technological Innovation and Patents

Frontmatter Page 2 Thursday, October 23, 2008 10:31 AM

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT The OECD is a unique forum where the governments of 30 democracies work together to address the economic, social and environmental challenges of globalisation. The OECD is also at the forefront of efforts to understand and to help governments respond to new developments and concerns, such as corporate governance, the information economy and the challenges of an ageing population. The Organisation provides a setting where governments can compare policy experiences, seek answers to common problems, identify good practice and work to co-ordinate domestic and international policies. The OECD member countries are: Australia, Austria, Belgium, Canada, the Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. The Commission of the European Communities takes part in the work of the OECD. OECD Publishing disseminates widely the results of the Organisation’s statistics gathering and research on economic, social and environmental issues, as well as the conventions, guidelines and standards agreed by its members.

This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the Organisation or of the governments of its member countries.

Also available in French under the title:

Études de l’OCDE pour l’innovation environnementale Politique environnementale, innovation technologique et dépôts de brevets Corrigenda to OECD publications may be found on line at: www.oecd.org/publishing/corrigenda.

© OECD 2008 Photo credit: © David Wasserman/Brand X/Corbis You can copy, download or print OECD content for your own use, and you can include excerpts from OECD publications, databases and multimedia products in your own documents, presentations, blogs, websites and teaching materials, provided that suitable acknowledgment of OECD as source and copyright owner is given. All requests for public or commercial use and translation rights should be submitted to [email protected]. Requests for permission to photocopy portions of this material for public or commercial use shall be addressed directly to the Copyright Clearance Center (CCC) at [email protected] or the Centre français d'exploitation du droit de copie (CFC) [email protected].

FOREWORD

Foreword

U

nderstanding the role that technological change can play in achieving environmental objectives is important since innovations can allow for improved environmental quality at lower cost. However, the relationship between environmental policy and technological innovation remains an area in which empirical evidence is scant. In an attempt to bridge this gap, the OECD has examined these issues. Three different case studies have been undertaken: abatement technologies for wastewater effluent from pulp production; abatement of motor vehicle emissions; and development of renewable energy technologies. While particular focus has been placed on the role of environmental policy in bringing about the innovation documented, it is recognised that other factors play a key role in inducing innovation which has positive environmental implications. The work was overseen by the delegates to the Working Party on National Environmental Policies of the OECD, who provided valuable comments and inputs at all stages of the project. In addition, the work has been presented at a number of international conferences, and comments received have served to improve the study significantly. The authors would also like to thank Dominique Guellec and Hélène Dernis of the OECD Directorate for Science, Technology and Industry for their foresight and hard work in developing the OECD’s patent database upon which much of this study is based. And finally, Claire-Line Martin provided excellent support in the preparation of the manuscript, which was very much appreciated.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3

TABLE OF CONTENTS

Table of Contents List of Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

13

Chapter 1. Environmental Policy, Technological InnovatIon and Patent Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

17

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

18

2. The economics of innovation and eco-innovation . . . . . . . . . . . . . . . .

20

3. Measures of innovation and eco-innovation . . . . . . . . . . . . . . . . . . . . .

24

4. The policy determinants of eco-innovation . . . . . . . . . . . . . . . . . . . . . .

38

5. The case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

44

Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

45

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

47

Annex 1.A1. Sources of Patent Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

52

Annex 1.A2. Patent Classification Systems . . . . . . . . . . . . . . . . . . . . . . . . . . .

57

Annex 1.A3. Number of EPO Applications in Different Environmental Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

59

Chapter 2. Environmental Regulation and International Innovation in Automotive Emissions Control Technologies. . . . . . . . . . . . . .

63

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

64

2. Environmental regulation in the automobile sector . . . . . . . . . . . . . . .

65

3. Innovation in automotive emissions control technologies . . . . . . . . .

75

4. Empirical analysis and results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

90

5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

93

Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

94

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

95

Annex 2.A1. Overview of Technologies and Corresponding Patent Classes . .

97

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5

TABLE OF CONTENTS

Chapter 3. Policy Versus Consumer Pressure: Innovation and Diffusion of Alternative Bleaching Technologies in the Pulp Industry . . 107 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 2. The pulp and paper industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3. Pollution and the pulp and paper industry . . . . . . . . . . . . . . . . . . . . . . 114 4. Regulatory responses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 5. Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 6. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Annex 3.A1. Relevant Patent Classes for Pulp Bleaching Technologies. . . . 138 Chapter 4. Renewable Energy Policies and Technological Innovation: Energy Source and Instrument Choice . . . . . . . . . . . . . . . . . . . . 139 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 2. The renewable energy sector: trends, technologies and policies . . . . 140 3. Patent applications for renewable energy . . . . . . . . . . . . . . . . . . . . . . . 144 4. Empirical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Annex 4.A1. First Page of Sample Patent Application . . . . . . . . . . . . . . . . . . 164 Chapter 5. Policy Conclusions and Further Work . . . . . . . . . . . . . . . . . . . . . 167 1. Policy conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 2. Further work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Annex 5.A1. Glossary of Relevant Patent and Related Terms . . . . . . . . . . . . 173 List of tables 1.1. Comparison of the patent systems of the United States, Japan and Europe (circa 2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30

1.2. IPC patent classification system for solar concentrating devices

6

used for the generation of mechanical power . . . . . . . . . . . . . . . . . .

36

1.3. Characteristics of the case studies. . . . . . . . . . . . . . . . . . . . . . . . . . . .

45

2.1. Technologies covered. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

77

2.2. Fixed-effects model estimates for engine re-design technologies . . .

92

2.3. Fixed-effects model estimates for post-combustion devices. . . . . .

92

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

TABLE OF CONTENTS

3.1. Pulp producers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 3.2. Percentage of exports to each country: paper and paperboard . . . . 111 3.3. Summary of Ecolabel programs related to pulp and paper manufacturing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 3.4. Summary of key regulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 3.5. Number of domestic chlorine and non-chlorine patents, selected years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 3.6. Top domestic patent assignees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 4.1. Share of electricity production from renewable sources (excluding hydro) (%) by country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 4.2. Examples of policies aimed at supporting renewable energy . . . . . 145 4.3. IPC classifications for renewable energy. . . . . . . . . . . . . . . . . . . . . . . 147 4.4. Number of EPO patent filings in renewable energy technologies (Annual average 1978-2003, by inventor country) . . . . . . . . . . . . . . . 149 4.5. Number of EPO patent filings in renewable energy technologies (Annual average 1978-2003, per unit of GDP, by inventor country) . . . 150 4.6. Number of EPO patent applications in renewable energy technologies, normalised by overall patenting activity (1978-2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 4.7. Descriptive statistics of explanatory variables (1978-2003) . . . . . . . 155 4.8. Estimated coefficients of the negative binomial fixed effects models with individual policy variables . . . . . . . . . . . . . . . . . . . . . . . 156 4.9. Correlation coefficients between policy variables . . . . . . . . . . . . . . . 157 4.10. Estimated coefficients of the negative binomial fixed effects models with a composite policy variable . . . . . . . . . . . . . . . . . . . . . . 158 4.11. Estimated coefficients of the negative binomial fixed effects models with clusters of policy variables . . . . . . . . . . . . . . . . . . . . . . . 160 List of figures 1.1. Share of environmental R&D in total government R&D, 1981-2005 . .

26

1.2. Proportion of facilities by country with budgets for environment-related R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

26

1.3. Proportion of facilities by employee size class with budgets for environment-related R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27

1.4. Share of new products in turnover . . . . . . . . . . . . . . . . . . . . . . . . . . .

29

1.5. Number of TPF patent applications by inventor country . . . . . . . . .

34

1.6. Number of EPO “Environmental” patent applications and total EPO patent applications . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36

1.7. Number of EPO “Environmental” patent applications . . . . . . . . . . .

37

1.A3.1a. Number of EPO Applications in Different Environmental Areas . .

60

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

7

TABLE OF CONTENTS

1.A3.1b. Number of EPO Applications in Different Environmental Areas (suite)

61

2.1. Evolution of US HC and NOX standards for passenger cars (gasoline) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

66

2.2. Evolution of US CO Standards for passenger cars (gasoline) . . . . . .

67

2.3. Evolution of Japanese CO, HC and NOX standards for passenger cars (gasoline) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

68

2.4. Evolution of Japanese CO, HC, NOX and PM standards for passenger cars (diesel) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

68

2.5. Evolution of European CO standards for passenger cars (gasoline and diesel) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69

2.6. Evolution of European HC and HC + NOX standards for passenger cars (gasoline) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

70

2.7. Evolution of European HC + NOX standards for passenger cars (gasoline and diesel) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

70

2.8. Evolution of European HC and NOX standards for passenger cars (gasoline) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

71

2.9. Evolution of European PM standards for passenger cars (diesel) . .

71

2.10. Evolution of patent applications at the USPTO, 1975-2001. . . . . . . .

78

2.11. Evolution of patent applications at the GPTO, 1975-2001 . . . . . . . . .

79

2.12. Evolution of patent applications at the JPO, 1975-2001) . . . . . . . . . .

80

2.13. Evolution of patent applications at the EPO, 1975-2001 . . . . . . . . . .

82

2.14. Technology shares within engine re-design group at different patent offices (1975-2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

82

2.15. Source countries for patents (1975-2001) . . . . . . . . . . . . . . . . . . . . . .

83

2.16. Average patent family size by country and year . . . . . . . . . . . . . . . .

84

2.17. Source countries of USPTO engine re-design and post-combustion patents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

86

2.18. Source countries of German engine re-design and post-combustion patents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

87

2.19. Source countries of JPO engine re-design and post-combustion patents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

88

2.20. JPO engine re-design and post-combustion patents by domestic inventors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

89

3.1. Domestic ECF and TCF patents by country . . . . . . . . . . . . . . . . . . . . . 123 3.2. Average patent family size by country and year . . . . . . . . . . . . . . . . 127 3.3. ECF and TCF patent trends, selected countries . . . . . . . . . . . . . . . . . 128 3.4. Diffusion of ECF bleaching technologies . . . . . . . . . . . . . . . . . . . . . . . 129 3.5. Diffusion of ECF and TCF bleaching technologies . . . . . . . . . . . . . . . 130 4.1. Annual growth rates for renewable energy in the world and the OECD (1990-2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 4.2. Renewable energy sources in the OECD in 2004 . . . . . . . . . . . . . . . . 142

8

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

TABLE OF CONTENTS

4.3. Percentage of energy to be provided by renewable energy . . . . . . . 144 4.4. Introduction of policies by type for renewable energy in OECD countries (1973-2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 4.5. Number of EPO patent applications by type of renewable . . . . . . . . 148 4.6. Number of EPO patent technologies for renewables by country . . . 148 4.7. EPO patent filings in renewable energy technologies (annual mean, per unit of GDP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 4.8. Relationship between point of introduction of policies and patent counts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 4.9. Dendogram of policy variable clustering . . . . . . . . . . . . . . . . . . . . . . 159

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

9

LIST OF ACRONYMS

List of Acronyms ADt AOX BAT BOD CAAA CAFÉ CARB CAT CIS ClO2 CO COD CVCC DSTI ECF ECLA EGR EPO GBAORD GERD GPTO H2O2 HC IPR JPO NORSCAN NOX NSF O3 OBD PAH Pb PCT PM

Air-dried Tonne Adsorbable Organic Halide Best Available Technology Biological Oxygen Demand Clean Air Act Amendments Corporate Average Fuel Economy California Air Resources Board Catalytic Converters and Catalytic Regeneration Technology Community Innovation Survey Chlorine Dioxide Carbon Monoxide Chemical Oxygen Demand Compound Vortex Controlled Combustion Directorate for Science, Technology and Industry (OECD) Elemental Chlorine-Free European Patent Classification System Exhaust Gas Recirculation European Patent Office Government Budget Appropriations and Outlays for R&D Gross Domestic Expenditures on R&D German Patent and Trademark Office Hydrogen Peroxide Hydrocarbons Intellectual Property Rights Japanese Patent Office US, Canada, Sweden, Finland and Norway Nitrogen Oxides National Science Foundation Ozone On-Board Diagnostics Polycyclic Aromatic Hydrocarbons Lead Patent Cooperation Treaty Particulate Matter

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

11

LIST OF ACRONYMS

R&D SO2 TCF TPES TPF TRIPS TSS USPTO WIPO

12

Research and Development Sulphur Dioxide Totally Chlorine-Free Total Primary Electricity Supply Triadic Patent Family Trade-Related Aspects of Intellectual Property Rights Total Suspended Solids United States Patent and Trademark Office World Intellectual Property Organization

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Executive Summary

T

echnological innovation can help realise environmental objectives in a less costly manner than would otherwise be the case. Thus, understanding the role that technological innovation can play in achieving environmental objectives is important for policy debates. However, the relationship between environmental policy and technological innovation is an area in which empirical evidence remains limited. In an attempt to bridge this gap, the OECD has examined these issues, using patent activity as a proxy for technological innovation. This work has assessed the role that environmental policies have played in inducing “eco-innovation”. Three different case studies have been undertaken: abatement technologies for wastewater effluent from pulp production; abatement of motor vehicle emissions; and development of renewable energy technologies. These cases were selected in order to ensure variation in the issues addressed (product vs. process innovation; integrated abatement technologies vs. end-of-pipe technologies; degree of tradability of the technologies). While there is no ideal measure of innovation, patent data have been widely used to assess the effects of policy and other factors on technological innovation in general. As such, in this report patent data is used to assess the nature, extent, and causes of innovation specifically in the “environmental” context. While particular focus has been placed on the role of environmental policy in bringing about the innovation documented, it is recognised that other factors play a key role in inducing innovation which has positive environmental implications. The factors which drive innovation in general are also likely to encourage eco-innovation specifically. For instance, factors such as macroeconomic stability, functioning of capital markets, the degree of economic “openness”, and the quality of education systems affect innovation rates in general, but also the specific case of eco-innovation by extension. There are, however, some distinct concerns which arise with respect to eco-innovation. Most importantly, there are two externalities involved in the eco-innovation case: the positive externalities associated with information spillovers resulting from the innovation process; and, the negative externality associated with the environmental impacts. In the case of knowledge spillovers those who are responsible for innovation bear the full costs, without receiving all the benefits. In the absence of policy interventions, the rate of

13

EXECUTIVE SUMMARY

innovation will therefore be sub-optimal, and thus an economy will be less competitive and productive over time. In the case of environmental externalities, those who are responsible for emitting pollution receive the full benefits from doing so, but without paying the full costs. In effect the “price” of polluting is too low. And as with the general theory of induced innovation, this will provide incentives for innovation which “uses” the under-priced factor input intensively. Therefore, in the absence of policy interventions to internalise the externality, innovation will bend in a direction which is relatively more pollution-intensive than would otherwise be the case. Environmental and innovation externalities are usually the responsibilities of different Government Ministries. Policy coordination is, therefore, key. However, innovation and environmental policy have different objectives. While the former is largely concerned with internalising knowledge spillovers (and thus increasing competitiveness and productivity), the latter is concerned with addressing negative environmental externalities. While no single instrument is likely to be able to address both market failures, co-ordination between the two is vital, if both the rate and the direction of innovation are to be optimal. As such, this volume focuses on the role of environmental policy in “bending” innovation in a more environmentally-friendly direction. Several interesting conclusions have emerged from the three case studies that were undertaken:

14

●

Environmental policy does have an effect on technological innovation. For instance, in the study on renewable energy, the implementation of different policy measures had a measurable impact on innovation, with tax measures and quota obligations being statistically significant determinants of patent activity. However, the effect of the different policies varied by the type of renewable energy involved.

●

General scientific capacity matters. Again in the case study on renewable energy innovation, the variable reflecting expenditures on targeted R&D was statistically significant in every model estimated.

●

Relative prices induce particular kinds of innovation. In the case of motor vehicle emissions abatement, fuel prices encouraged investment in “integrated” innovation (in which fuel efficiency gains also arose), but not in “post-combustion” technologies. In the case of renewable energy, the role of electricity prices was rarely significant, except for solar energy. However, as fossil fuel prices rise (and renewables become more competitive), the price substitution effect is likely to become more important.

●

Other market factors can also be important spurs to innovation. In the case of bleaching technologies in the pulping process, public concerns about the

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

EXECUTIVE SUMMARY

environment appeared to spur the development of ECF and TCF technologies, pre-dating the introduction of regulatory standards. Interestingly, eco-labelling did not appear to have an influence on innovation in this case. ●

The type of innovation changes through time. In the renewables sector, different energy sources have reached maturity at different points, and there have been different “generations” of innovation within particular renewable energy sources. In the case of motor vehicle emissions abatement, there has been a shift from post-combustion technologies to integrated technologies.

●

International diffusion of environmental innovation is common. In the case of both bleaching technologies and motor vehicle emissions, abatement patent families (for some countries) are large, reflecting significant technology transfer. In the case of motor vehicle emissions abatement, the transfer of Japanese technologies to the US is striking.

●

In other areas, there is some evidence of a “first-mover” advantage. In the pulp and paper sector, the early policy interventions introduced by Finland and Sweden resulted in a strong comparative advantage in TCF technologies.

While not addressed directly in this project, two more general conclusions emerge from the literature. First, investing in R&D is risky. As such it is important that the environmental policy framework not add to this risk, but rather provide investors with a stable horizon in which to undertake research investments. If markets have difficulty efficiently dealing with commercial risk associated with innovative activity, they will be even less likely to deal efficiently with the uncertainties associated with unstable policy conditions. Second, the policy framework should allow for a variety of technological options to be adopted. Governments have limited resources at their disposal, as well as limited information about optimal technological trajectories. Moreover, with the potential for “lock in”, it is important to develop policies which minimise the downside risks of “picking losers”. In general, this means targeting environmental policies as close as possible to the environmental objective itself, not some proxy. The volume concludes that patent data is a helpful means by which to examine eco-innovation and suggestions for future policy research are proposed: a) development of robust indicators of eco-innovation across a wider spectrum of environmental areas (e.g.“green” chemistry, abatement of air pollution, carbon capture and mitigation, building energy efficiency, waste prevention, etc.); b) assessment of the environmental and economic “returns” on patented eco-innovation inventions in selected areas; and c) examination of the links between environmental policy, economic globalisation and eco-innovation, with a focus on international technology transfer.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

15

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Chapter 1

Environmental Policy, Technological InnovatIon and Patent Activity by Nick Johnstone and Ivan Hascic (OECD Environment Directorate) and Katrin Ostertag (Fraunhofer Institute)*

Technological innovation can allow for the realisation of environmental objectives in a manner which is less costly than would otherwise be the case. As such, understanding the role that technological innovation can play in achieving environmental objectives is important for policy debates. This chapter provides a review of the theory and evidence with respect to the role that environmental policies can play in inducing innovation, and provides an introduction to case studies undertaken. On the basis of patent data, the nature, extent, and causes of innovation with respect to renewable energy, wastewater treatment and motor vehicle emissions control were explored. While particular focus has been placed on the role of public policy in bringing about the innovation documented, it is recognised that other factors play a key role in inducing innovation which has positive environmental implications.

* Comments from Davis Popp and Frans de Vries gratefully acknowledged.

17

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

1. Introduction Technological innovation can help realise environmental objectives in a less costly manner than would otherwise be the case. For example, some innovations may drive down the cost of pollution abatement. This results in welfare gains. On the one hand, the same environmental quality may be achieved with fewer factor inputs devoted to abatement. On the other hand, for the same amount of factor inputs, environmental quality may be improved. Thus, an improved understanding the role that technological innovation can play in achieving environmental objectives is important for policy makers. However, the relationship between environmental policy and technological innovation remains an area in which empirical evidence is scant.1 Moreover, there are almost no empirical studies with international coverage. Part of the reason for this is the absence of appropriate comparative data to reflect the innovation process. In this report patent activity is used as a proxy for technological innovation. Three different case studies have been undertaken: abatement technologies for wastewater effluent from pulp production; abatement of motor vehicle emissions; and development of renewable energy technologies. These cases were selected in order to ensure variation in the issues addressed and the characteristics of the innovation. Specifically: ●

autos are a traded commodity, and as such international influences matter. In addition, autos are a consumer good and thus the objectives of regulations are to reduce end-use pollution, not pollution from production;

●

in the case of pulp and paper, unlike autos, we are focusing on pollution at the source of production. However, the focus is on a process technology, rather than end-of-pipe solutions. The analysis of innovation with respect to a process technology represents a new contribution to the literature; and,

●

renewable energy sources are both a process (generating electricity) and a product (e.g. the turbines themselves). Electricity is not generally a traded commodity, which contrasts with autos. However, the renewable equipment may be traded.

On the basis of patent data, the nature, extent, and causes of innovation in each of these areas were explored. Particular focus has been placed on the role of public policy in bringing about the innovation documented. This includes both policy stringency and the nature of the environmental policy

18

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

instruments applied. However, it is recognised that other factors play a key role in inducing innovation which has positive environmental implications. Factors such as fuel and energy prices, sectoral growth rates, and general scientific capacity have been controlled for in the more formal analyses included in the case studies. Several interesting conclusions have emerged from the three case studies that were undertaken: ●

Environmental policy does have an effect on technological innovation. For instance, in the study on renewable energy, the implementation of different policy measures had a measurable impact on innovation, with tax measures and quota obligations being statistically significant determinants of patent activity. However, the effect of the different policies varied by the type of renewable energy involved.

●

General scientific capacity matters. Again in the case study on renewable energy innovation, the variable reflecting expenditures on targeted R&D was statistically significant in every model estimated.

●

Relative prices induce particular kinds of innovation. In the case of motor vehicle emissions abatement, fuel prices encouraged investment in “integrated” innovation (in which fuel efficiency gains also arose), but not in “post-combustion” technologies. In the case of renewable energy, the role of electricity prices was rarely significant, except for solar energy. However, as fossil fuel prices rise (and renewables become more competitive), the price substitution effect is likely to become more important.

●

Other market factors can also be important spurs to innovation. In the case of bleaching technologies in the pulping process, public concerns about the environment appeared to spur the development of ECF and TCF technologies, pre-dating the introduction of regulatory standards. Interestingly, eco-labelling did not appear to have an influence on innovation in this case.

●

The type of innovation changes through time. In the renewables sector, different energy sources have reached maturity at different points, and there have been different “generations” of innovation within particular renewable energy sources. In the case of motor vehicle emissions abatement, there has been a shift from post-combustion technologies to integrated technologies.

●

International diffusion of environmental innovation is common. In the case of both bleaching technologies and motor vehicle emissions, abatement patent families (for some countries) are large, reflecting significant technology transfer. In the case of motor vehicle emissions abatement, the transfer of Japanese technologies to the US is striking.2

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

19

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

●

In other areas, there is some evidence of a “first-mover” advantage. In the pulp and paper sector, the early policy interventions introduced by Finland and Sweden resulted in a strong comparative advantage in TCF technologies.

This introductory Chapter begins with a review of the literature related to the key determinants of innovation, with specific focus on “environmental” innovation. It also provides a discussion of alternative measures of innovation and eco-innovation, before proceeding to a discussion of the impact of environmental policy on environmental innovation. The Chapter concludes with a summary of the characteristics of the case studies that were undertaken. The following Chapters then present the case studies, followed by a concluding Chapter which summarises the main policy conclusions and proposes areas for further research.

2. The economics of innovation and eco-innovation When assessing the determinants of innovation which results in reduced environmental impacts (hereafter eco-innovation), it is important to first understand the determinants of innovation in general. The role that general market and policy factors play in encouraging innovative activities is therefore crucial to an understanding of the factors that facilitate innovation which is more specifically environmental in nature. Since the primary interest of this report is the analysis of the influence of environmental policy instruments on ecoinnovation, the following review of the more general determinants of overall innovation will be very brief.

2.1. The determinants of innovation in general There is considerable variation in innovative activity – both across OECD countries and within OECD countries between sectors of the economy (see OECD, 2007a). Researchers have sought to identify the reasons for this variation, and a number of factors appear to be significant. These can be distinguished as i) market and firm-level factors; and ii) policy factors.

2.1.1. Market and firm-level factors Over six decades ago, Schumpeter (1942) argued that there was a positive correlation between market concentration and innovation. In theory, this is because a monopoly is in a better position to prevent imitation, and because it will have more resources with which to finance R&D activities. A monopoly may therefore be better able to bear the risk (and reap the benefits) associated with R&D investments. However, in a seminal paper, Arrow (1962) argued that the efficiency incentives associated with perfect competition are actually more conducive to innovative activities.

20

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Given these conflicting theoretical views concerning the potential role of market structure on incentives to invest in R&D, it is not surprising to find that the empirical evidence is also mixed. While Gerosky (1990) found support for the positive effects of competition, Kraft (1987) and Acs and Audretsch (1987) found support for the Schumpeterian hypothesis.3 Therefore, theoretical and empirical support has been found for both a negative and a positive relationship between the level of innovation and the degree of market competition. The spatial scope of the market is also thought to be an important driver of innovation. In effect, the more global the market in which a firm operates, the more likely the firm will be to innovate. Criscuolo et al. (2005) found that markets serve as a conduit of information; as such, the pool of information available to global firms is greater than that which is available to national and local firms. For similar reasons, it is sometimes argued that foreign direct investment can be an important conduit, expanding the potential knowledge pool upon which the firm can draw. Some recent empirical evidence (Jaumotte and Pain 2005a) indicates that “openness” is among the most important determinants of investment in R&D. Another important feature of industrial R&D arises out of the frequent need for a firm to self-finance its R&D investments. The inherent riskiness of these investments, the skewed nature of potential returns, and the potential for asymmetric information between prospective borrowers and lenders make external sources of financing more difficult to secure (Jaumotte and Pain, 2005a). Under such conditions, capital markets may have difficulty assessing the optimality of potential investments in R&D (Scherer and Harhoff 2000). Consequently, it is argued that firms with greater internal financial resources are more likely to invest in R&D. Two explanations have been advanced for this (see for example, Kamien and Schwartz, 1978): ●

Outside lending may be difficult to obtain, because a failed R&D project leaves few valuable tangible assets. Given the risk associated with R&D projects, external lenders may be reluctant to finance such projects without substantial collateral.

●

Firms may be unwilling to reveal private information that could make the project interesting to outside lenders, because they fear that this information will then become available to rivals.

Such financing problems are not thought to be as important for firms quoted on the stock market – because these firms have easier access to capital than other firms (Syrneonidis, 1996). Moreover, the recent deepening of venture capital markets in some countries/regions (North America, Netherlands, and UK), has tended to reduce some of the need to rely upon internal finance (OECD, 2006a).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

21

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Small firms may face particular difficulties in financing their R&D projects. As noted by Jaffe et al. (2003), these firms “have less internally generated cash and/ or less access to financial markets”. Firm size may also be important if there are significant economies of scale associated with R&D. The evidence in this area is mixed as well (Syrneonidis, 1996). However, it would appear that the effect of firm size is non-linear, becoming less important once a minimum threshold is reached. Empirical evidence indicates that this threshold may be as low as approximately 100 employees (Syrneonidis, 1996).

2.1.2. Public policy factors According to work by the OECD a number of general “policy framework conditions” are also likely to be important drivers of innovation. First, stable macroeconomic conditions are positively related to innovation. Low and stable interest rates are particularly important for risky investments, such as those associated with R&D (OECD, 2006a). Second, firms which are exposed to international market competition are more likely to be innovative. As such, more open international trade and foreign investment policies will tend to result in more innovation. Third, and in a similar vein, when a firm’s products operate in markets that are less heavily regulated, innovation tends to be stronger (Jaumotte and Pain, 2005c). The effects of more targeted “innovation” and “science” policies have also been assessed. One intrinsic characteristic of the knowledge produced through R&D is that it is very difficult to exclude others from the benefits that arise from that R&D. This makes knowledge a public good – as such, it will be underprovided by the market, because the private returns to R&D investments are much lower than the social returns [for an empirical analysis, see for example Mansfield et al. (1997)]. In addition, and as noted earlier, even in the absence of the public good elements of R&D, there can be particular difficulties in financing R&D at optimal levels. There is, therefore, a need for policy intervention, whether to increase the returns on innovation or to reduce the cost of its generation. [See OECD (2004) for a review of current practice in OECD countries.] OECD governments have provided financial support for R&D for many years. In recent years, the form in which this support is provided has changed, with a transition being made from grants to various forms of tax credits (OECD, 2007b). However, the evidence on the returns to public sector support for R&D is mixed. While Hall and van Reenen (2000) find positive evidence that tax incentives encourage investment in R&D, a meta-analysis undertaken by Garcia-Quevado (2004) did not find systematic evidence in support of the more general contention that subsidies encourage greater private sector R&D. The effectiveness of subsidies depends upon the degree of “additionality” associated with the subsidies being provided – i.e. the extent to which they

22

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

lead to a higher overall level of R&D expenditure. On the one hand, “crowding out” can occur if the funds provided displace private sector expenditures that would otherwise have been made. Falk (2004) found evidence that government subsidies have a significant positive effect on business R&D, but only when previous R&D is controlled for – suggesting that the beneficiaries of subsidies may be mainly firms that have previously undertaken R&D. On the other hand, “crowding in” may also occur: if there are capital constraints and economies of scale in innovation, provision of public support may lead to greater privatelyfinanced R&D. Interestingly, Lach (2002) found that subsidies to Israeli SMEs had a significant (and positive) impact on company-financed R&D, but this was not the case for larger firms. In order to address the public good nature of the knowledge produced by R&D, legal protection can be provided to help innovators capture potential rents. Thus, the strength of a country’s intellectual property rights (IPR) regime is often thought to be a key driver of R&D. However, recent cross-sectional empirical work [e.g. Cohen et al. (2000)] generally found that the IPR regime is a weak predictor of R&D investment in OECD countries. This may be due to the fact that there is little difference in the stringency of IPR regimes across countries, so isolating their effect is complicated. Nonetheless, in work undertaken at the OECD, it was found that a 1% increase in the standard deviation of the “Park” index of IPRs resulted in an 8% increase in the number of patents and 1-1.5% in R&D expenditures (Jaumotte and Pain, 2005a). In addition, many governments have put in place programmes which facilitate cooperation between public research institutions and industry. [See Jaumotte and Pain (2005b) for a discussion.] While many of these programmes are motivated by a wish to realise greater economic benefits from publiclyfunded R&D, they may also serve as a spur to private sector R&D. If well-designed, these programmes can encourage the “internalisation” of knowledge externalities between public and private bodies and (most importantly) between different private bodies.

2.2. The specific case of eco-innovation The lessons learned concerning innovation in general, and the role that public policy can play in inducing this innovation are, of course, directly relevant to a more specific discussion of eco-innovation. For instance, market structure may be important because some of the sectors where environmental impacts are potentially most important (e.g. chemicals, pulp and paper, energy) also have concentrated markets. There are, however, some distinct concerns which arise with respect to eco-innovation. Most importantly, there are two externalities involved in the eco-innovation case: the positive externalities associated with information

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

23

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

spillovers resulting from the innovation process and the negative externality associated with the environmental impacts (see Jaffe et al., 2005). In the case of knowledge spillovers those who are responsible for innovation bear the full costs, without receiving all the benefits. In the absence of policy interventions, the rate of innovation will therefore be sub-optimal, and thus an economy will be less competitive and productive over time. In the case of environmental externalities, those who are responsible for emitting pollution receive the full benefits from doing so, but without paying the full costs. In effect the “price” of polluting is too low. And as with the general theory of induced innovation,4 this will provide incentives for innovation which “uses” the under-priced factor input intensively. Therefore, in the absence of policy interventions to internalise the externality, innovation will bend in a direction which is relative more environment-intensive than would otherwise be the case. In general, distinct policies need to be implemented to resolve both of these problems. Such policies have generally been the responsibilities of different Ministries. This is appropriate, because of the differing underlying policy objectives. However, in some cases, it can lead to policy incoherence. For instance, the objectives of environmental policy measures with respect to eco-innovation may be undermined by innovation policy measures which support more polluting technologies (see Goel and Hsieh, 2006). Many OECD governments have made significant efforts to co-ordinate these two sets of policies (e.g. the European Union’s Environmental Technology Action Plan). On the one hand, the “innovation” effects of environmental policies have become an increasingly important criterion in policy assessment for Ministries of the Environment. On the other hand, the “environmental” effects of innovation policies have become an important criterion for policy assessment in Ministries of Science, Technology and Industry (see Kivimaa and Mickwitz, 2006 and Kivimaa and Mickwitz, 2004). Indeed, in the OECD’s Science, Technology and Industry Outlook (2004), a majority of countries cite environment-related concerns in their science and technology priorities, including: Australia (environmentally sustainable Australia); Austria (environment, energy and sustainability); France (development of renewable energy); Germany (clean processes and production technologies); Hungary (environmental protection); Norway (energy and environment); UK (sustainable energy) and the US (climate, water and hydrogen).

3. Measures of innovation and eco-innovation Given the importance of technological innovation in modern economies, identifying reliable measures of technological innovation has long preoccupied economists (and continues to do so). The task is further complicated when

24

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

measures of specific “types” of innovation are sought – such as “eco-innovation” – even if a reliable measure of innovation can be identified, it may be difficult to determine whether it is specifically “environmental” in nature.

3.1. Input and output measures of innovation Due to the complexity of the innovation process, all potential measures of innovation (e.g. budgets for research and development, number of scientific personnel) are at best imperfect indicators of the “innovativeness” of an economy. [For annual data on research and development and other relevant indicators, see OECD Research and Development Statistics (2006c), and Main Science and Technology Indicators (2006b).]5 Following the different phases of the R&D process [i.e. basic and applied research and (experimental) development] through to the different stages of innovation (from invention to diffusion), a range of different indicators for measuring technological change can be found in the literature. Three basic groups can be distinguished: resource (or input) indicators; indicators of R&D results (i.e. output indicators of the R&D process); and progress indicators (i.e. output indicators focussing on the economic impacts of innovations) (Grupp, 1998). Input indicators follow the rationale that, for technological change to take place, the necessary resources have to be invested in knowledge acquisition. Among the most commonly applied input indicators are: i) R&D expenditures and ii) R&D personnel. Generally, data on R&D personnel is not as widely available as data on R&D expenditure. However, input indicators may also refer to other ways of knowledge acquisition, such as investment in R&D intensive goods, or expenditure for licenses. The OECD has also sought to disaggregate public R&D data by “socioeconomic” objective (OECD 2002). (No effort has been made to collect private sector R&D data disaggregated by socio-economic objective.) In principle, the allocation of expenditures to specific objectives is determined on the basis of managerial intentions at the time of commitment of the funds. However, given the uncertainty associated with general R&D, further disaggregating public R&D data by “socio-economic” objective may be even more difficult to establish with confidence, particularly for more basic forms of research. Moreover, even if this could be done, drawing boundaries between the different objectives6 is by no means straightforward. Bearing these caveats in mind, data on public sector R&D expenditures for “control and care of the environment” is presented in Figure 1.1. Based on this figure, many OECD countries have clearly been increasing their investment in environmental research and development (R&D), in an effort to boost technological developments that improve environmental quality.7

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

25

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Figure 1.1. Share of environmental R&D in total government R&D, 1981-2005 % 6

Germany

Korea

France

United Kingdom

Japan

United States

5 4 3 2 1 0 1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

Source: OECD (2005).

At the micro level, a survey of 4 000 manufacturing facilities in seven countries undertaken by the OECD Environment Directorate in 2003 provided data on those facilities which reported having undertaken “environmentrelated” R&D. 8 In the sample, 58.7% of facilities reported that they had incurred R&D expenditures, and 9.3% of facilities reported having invested in “environment-related R&D”. Figure 1.2 shows the proportion of facilities reporting that they had environment-related R&D expenditures (by country). Figure 1.2. Proportion of facilities by country with budgets for environment-related R&D Mean +/– 1 std. error 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00

Canada

France

Germany

Hungary

Japan

Norway

United States

Total

Source: Johnstone (2007).

26

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Norway had the highest percentage, with just under 15% reporting having do ne so, while for Germany, it is only 3.6%. For four of the seven countries (France, Canada, Japan and the US), the proportion was approximately 10%.” Figure 1.3 shows the proportion of facilities with environment-related R&D budgets (according to facility size). As can be seen clearly, larger facilities (more than 500 employees) are more likely to have undertaken such investments (with almost 20% reporting having done so), while this figure is less than 10% for facilities with less than 250 employees. In the sample as a whole, the average size of facilities responding affirmatively had over 720 employees; for those who responded negatively, the relevant figure was less than 300. The data appears to be consistent with results obtained from other surveys. For instance, according to statistics on research and development from the US National Science Foundation,9 55.3% of US companies with more than 5 employees in the manufacturing sector reported R&D expenditures for 2001. In the OECD sample, 51% of US manufacturing facilities (with more than 50 employees) are engaged in R&D. Among those, 16% reported having a R&D budget specifically for environmental matters. In the OECD survey, the sectors with the highest percentage of facilities reporting having undertaken investments in environment-related R&D were: coke, petroleum and refining (13.6%); chemicals and chemical products (12.8%); paper and paper products (12.5%); and motor vehicles (12.4%). Figure 1.3. Proportion of facilities by employee size class with budgets for environment-related R&D Mean +/– 1 std. error 0.25

0.20

0.15

0.10

0.05

0.00

< 100

100-249

250-499

> 500 Number of employees

Source: Johnstone (2007).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

27

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Although input indicators (and R&D expenditures in particular) reflect an important element of the innovation system, there are a number of disadvantages associated with their use as an indicator of innovation. For example, and as noted earlier with respect to private R&D expenditures, the data are incomplete. Further, the data are only available at an aggregate level and cannot be broken down by technology group. In addition, efforts to “unbundle” specific “environmental” R&D from other types of R&D can be difficult, particularly as efforts to improve environmental performance become more integrated with general business strategies (see Johnstone and Labonne, 2006). However, the greatest shortcoming associated with the use of R&D data is that it measures inputs to the innovation process, rather than successful outputs. Given the uncertainty associated with the innovation process, the economic value of equivalent R&D expenditures can also vary widely. As such, an “output” measure of innovation is broadly preferable. There are two principal candidates for indicators of the output of R&D: bibliometric data (scientific publications) and technometric data (patent publications). The use of bibliometric data as a measure of innovation has been given renewed impetus with the growth of the internet, combined with increasingly efficient search engines. Using keywords and indexing codes, searches of relevant databases (e.g. the Science Citation Expanded Index) are typically undertaken. Data on author, affiliation, date of publication, etc. can be extracted, and counts developed to assess relative innovative capacity (see Meyer, 2002). This kind of indicator is particularly useful as for analysing the diffusion of knowledge among inventors (and between countries), based on co-publications and citations. However, there are some shortcomings associated with the use of bibliometric data. In particular, while such data is indeed an output indicator of innovation, it is only an indicator of an “intermediate” output. Publication in a peer-reviewed journal reflects a scientific advance, but not necessarily one which has commercial applications. It is difficult to use citations as an index of quality, let alone economic importance. For these reasons, patent data are more commonly used as output indicators of innovation. Compared to bibliometric data, patents are more closely linked to applied research and experimental development. They are therefore closer to the commercial stage of innovation. In addition, for reasons explained below, patents lend themselves to cross-country comparisons of innovation performance in specific technology fields. Downstream output-based measures of innovation necessitate the identification of specific innovations which have been developed and commercialised. Data of this kind is usually collected through surveys of firms; and self-reporting of the introduction of novel goods/technologies or successful inventions. However, since the answers provided in these surveys are highly

28

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

subjective, they are often difficult to interpret. Most commonly, a company’s turnover with new products is used. For instance, the European Community Innovation Survey (CIS) requests that respondents indicate the value share of “novel” products in total turnover.10 While it is difficult to assess the reliability of responses to such questions, they do give an indication of self-perceived innovation. Unfortunately, the CIS did not previously ask respondents to provide information of specific relevance to environmental concerns. Figure 1.4 provides the aggregate data from the third CIS, which reported on responses from the 1998-2000 survey. Figure 1.4. Share of new products in turnover Share in total turnover (%) – 1998-2000 25 20 15 10

y an

d lan

rm Ge

Fin

ain Sp

ly Ita

l

m

ga rtu

iu lg Be

Po

k ar nm

ria

De

st Au

ce

ce

an Fr

ee

g ur bo m

Gr

ay rw Lu

xe

No

Ic

ela

0

nd

5

Source: Eurostat, Third Community Innovation Survey.

3.2. Patents as a measure of innovation For this report, patents have been used as a key measure of innovation.11 With the exception of the European Patent Office, patents are granted by national patent offices (usually specialised agencies) in individual countries. They give the holder the right to exclude others from the production of a specific good (or from the use of a specific process) for a defined number of years. This period may vary, depending on the nature of the innovation. In order to be eligible for a patent, the innovation must be novel, involve a nonobvious inventive step, and be useful (see Dernis and Guellec, 2001). Definitions of “novelty” differed in the past among countries. Today generally, the concept of universal novelty is implemented in national patent legislation. It requires that “no publication of any sort (…) may occur anywhere in the world prior to the filing date of the invention” (Adams, 2006).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

29

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Patent protection is only valid in the country that grants the patent. Inventors who desire patent protection in other nations must file separate applications in those nations, either directly or indirectly [e.g. through a regional patent office such as the European Patent Office (EPO) or the World Intellectual Property Organisation (WIPO)]. Although the broad principles are the same, there are differences in administrative procedures and nature of protection granted between patent offices. Some of the main characteristics are presented in Table 1.1.12

Table 1.1. Comparison of the patent systems of the United States, Japan and Europe (circa 2000) Prosecution Features

United States

Japan

Europe

Basis of deciding patent priority

First to invent

First to file

First to file

Native language filing permitted?

Any language (English translation for record)

Japanese or English

Any EPO member language, with English, French or German translation; translations required for each country designated

Patent term

20 years from filing

20 years from filing

20 years from filing

Publication of patent

Yes, 18 months from filing

Yes, 18 months from filing

Yes, 18 months from filing

Examination deferral

No

Yes, 7 years from filing

Yes, 6 months from publication

Third-party contestation of patent

Post-grant opposition

Post-grant oppositions

Post-grant oppositions

Patentability standards

Least unfavourable to applicants

Strict standards-claims need working examples

Moderate standards; some variation by nationality, limited grace period

Breadth of claim awarded

Very Broad

Narrow

Broad

Overall comparable costs (1993)

$13.000

$30.000

$120.000

Strength of enforcement

Very strong

Weak

Medium (varies) – enforcement by national patent offices

Interpretation of claims

Literal interpretation, pro patentee court systems, equivalents applied

Narrow interpretation linking back to specifications, equivalents not applied

Interpreted less literally; equivalents applied, but under different principles

Judicial style

Adversarial court trials, jury trials often seen

Brief intermittent hearing with judge, written testimony is widely used

Largely by written testimony, some oral hearings, court experts used

Uncertainty in outcomes

High, due to lay judges (in District Courts) and juries. No in UK

High, especially due to delays, and pressure to settle out of court

Relatively low in all jurisdictions

Time taken (1993)

18-24 months

–36 months + opposition (3-5 years)

–30 months + opposition (expeditious)

Source: Adapted from Somaya (2000) and Gallini (2002).

30

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

While differences remain, there has been considerable convergence in the characteristics of intellectual property rights regimes over time. Indeed, applying the “Park Index”13 of the strength of IPRs from 1980 to 2000, Jaumotte and Pain (2005c) found that there is relatively little variation across OECD countries. Recent data suggest that the coefficient of variation in the index for OECD countries fell from 0.32 in 1990, to 0.07 in 2005 (Park and Lippoldt 2007). EPO patent applicants may designate as many of the 32 EPO member-states for protection as desired. The application is first examined and – if successful – granted by the EPO. The patent is then transferred to the individual national patent offices designated for protection. Because EPO applications are more expensive, European inventors typically first file a patent application in their home country, and then apply to the EPO if they desire protection in other European countries. However, by filing with EPO, the total costs will be much less than if individual applications are made to each country (Popp, 2005). The procedure through the WIPO is based on the Patent Co-operation Treaty (PCT). In this case, any of the more than 140 signatory nations can be designated for patent protection. The WIPO issues a preliminary examination report, but does not actually grant the patent. This is the responsibility of the national patent office, to which the application is subsequently transferred. Applications abroad must be filed within one year of the priority date, which is the year in which the initial application was filed. If the inventor does file abroad within one year, the inventor will have priority over any similar patent applications received in those countries since the priority date. This legal concept of “priority” was first introduced in the Paris Convention in 1883. Additional fees apply for each application. Because of these features of patent law, only the more valuable inventions are filed in several countries. Moreover, filing a patent application in a given country is a signal that the inventor expects the invention to be profitable in that country. Patents are typically sorted by the priority year. If a patent is granted, protection begins from the priority date. It corresponds quite closely to the date when the inventive activity took place, as patent applications are usually filed early in the innovation process. Initially, patents were only published when granted. However, due to long delays of such publications and the resulting disadvantages (e.g. duplication of R&D), starting in the 1960s, most patent offices adopted the “deferred examination process”. This includes the requirement that an application be published while it is still pending – typically around 18 months after the earliest filing date (Adams, 2006). In empirical research, it is therefore important to distinguish between data on patent applications and data on patents granted.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

31

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Patent data provide a rich data source for empirical studies. In particular, they have the following strengths: ●

There are very few examples of major inventions which have not been patented (Dernis and Guellec, 2001).

●

Compared to R&D expenditure, patents are an output indicator (i.e. they measure productive innovation activity).

●

Further, the data classified into detailed technologies. A distinction can, therefore, be made between, for example, air pollution control devices designed to reduce NOX emissions and devices designed to control SO2 emissions (see e.g. Popp, 2005).

●

Patent data is generally readily available (although not always in a convenient format). They are based on objective standards that change slowly on the basis of invention novelty and utility (e.g. Griliches, 1990).

●

Patent data are also discrete, and thus easily subject to statistical analysis. Moreover, given the objective standard of assessing patents, these statistics provide some uniformity in comparing innovation across countries.

●

And finally, patent data have the potential to be linked to sectoral data (see e.g. EPO and OECD work on concordance between ISIC and IPC codes) and to firm level data.14

It is important to recognise that patents cannot be used to develop a measure of all innovations. First, they are designed only to protect technological inventions. Other IPR regimes exist to protect innovations in other fields – for example, copyrights for literature, trademarks for words or graphic devices which distinguish a product and the registration of designs, and where protection is focussed on the appearance of a product (Adams, 2006). Less formal ways (than intellectual property rights) to protect technological inventions also exist – notably industrial secrecy, or purposefully complex technical specifications (Frietsch and Schmoch, 2006).15 In return for receiving the monopoly rights inferred by a patent, the inventor is required to publicly disclose the invention. Rather than make this disclosure, inventors may prefer to keep their invention secret. Surveys of inventors indicate that the rate at which new innovations are patented varies across industries (Cohen et al., 2000 and Blind et al., 2006). For meaningful empirical analyses, it is therefore important to control statistically for these differences in the “propensity to patent”. A further critique of patent data relates to the fact that not everything that is patented is eventually commercialised and adopted. There are, however, significant fees attached to the examination of a patent application (and to renewal fees, once the patent has been granted). So it is safe to assume that, at least in the expectations of the applicant or patent holder, the prospects for

32

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

commercialisation and adoption are good. Nonetheless, the economic value of patents varies (Popp, 2005). The OECD’s Triadic Patent Family only includes patents which have been deposited at all of the European Patent Office (EPO), the Japanese Patent Office (JPO), and the US Patent and Trademark Office (USPTO). This ensures that the patents included in the TPF database are of wide commercial value. In addition, different patent offices apply different rules to define the “scope” of a patent. In particular, the JPO often requires multiple patents for a single patent at the USPTO or the EPO. When comparing patent counts across countries, this can result in a significant bias. Most patent databases apply a “consolidation” filter, in which all patents which are derived from the same “original invention” are included in a single patent family. The latter is thus defined as the set of patent applications in multiple countries pertaining to the same invention (and thus, sharing the same priority details 16 ). The inventor country of a patent family is the country of residence of the inventor, as stated in the priority filing. Drawing on the feature that additional fees apply for each patent filing abroad, Lanjouw and Schankerman (2004) have used data on patent families as proxies for the quality of individual patents. However, when comparing patent families across different countries of origin, the general patterns of patent filings abroad have to be kept in mind. Since these are motivated by commercial strategies, they depend on regional specificities, size of the domestic market, and the export orientation of an economy (Grupp 1998). As a result, the propensity for patent applicants to file internationally varies significantly. These factors explain why, for European countries, the percentage of patent families that are first filed in one European country, and then followed by subsequent patent applications for the same invention abroad, is relatively high (e.g. 56% for Germany in the years 2000-2005). In contrast, the same indicator for the US (with its very large domestic market) and for Japan (as a more “closed” economy) is 42% and 19%, respectively (WIPO 2007).17 In terms of host countries, the patent offices in the US, China and the EPO receive the most non-resident patent filings (WIPO, 2007). On the basis of the TPF database, the OECD has developed a wide variety of indicators of the innovation capacity of different OECD economies. This includes data on the total number of patents filed by country at both the European Patent Office and the USPTO, as well as in these two offices and the Japanese Patent Office. Figure 1.5 provides data on the number of TPF applications by inventor country for selected OECD countries. Further data is also being made available (e.g. citation data, inventor data) with on-going development of the PATSTAT database; it is anticipated that this additional work will also be exploitable for future empirical work related to eco-innovation.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

33

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Figure 1.5. Number of TPF patent applications by inventor country Germany 20 000

Japan

United States

A. Germany, Japan and United States

18 000 16 000 14 000 12 000 10 000 8 000 6 000 4 000 2 000

19 83 19 84 19 85 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

0

1 200

Australia

Austria

Belgium

Canada

Denmark

Finland

France

Hungary

Ireland

Italy

Korea

Netherlands

Norway

Spain

Sweden

Switzerland

United Kingdom

B. Other OECD Countries

1 000

800

600

400

200

02

01

20

00

20

99

20

98

19

97

19

96

19

95

19

94

19

93

19

92

19

91

19

90

19

89

19

88

19

87

19

86

19

85

19

84

19

19

19

83

0

While the TPF database eliminates certain biases in patent data, the geographic and consolidation filters applied in its development may not be sufficient for examining very specific areas of technological innovation. Since a very large number of applications and grants do not pass through the relevant filters, the resulting patent counts may be very small. The more “specific” the area under analysis, the more important this problem will be. Perhaps more

34

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

significantly, these filters may not be appropriate for assessing the effects of domestic environmental policies on innovation (see Popp, 2005). As such, the TPF database may not always be the most appropriate source of data for very specific research questions. In the work undertaken, patent data was therefore also obtained from a commercial database (Delphion) for two of the three case studies. (See Annex 1.A1 for a discussion of different sources of patent data). Delphion applies a “consolidation” filter similar to that used in the TPF and as such, differences in the “breadth” of patents between different countries have been corrected for.

3.3. Patents as a measure of eco-innovation Arguably, the identification of those innovations which are specifically “environmental” in nature is facilitated through the use of patent data – due to the nature of the classification systems used for IPRs. Specifically, relevant patents can be identified using the International Patent Classification (IPC), developed at the World Intellectual Property Organisation (www.wipo.org). While there are other classification systems, most researchers use the IPC classification system when searching for patents. This classification system involves a hierarchy of codes, structured into different levels. When patents are granted, they are given technology classifications and sub-classifications by various patent offices. These classifications can then be used to identify all relevant patents for a given type of technology.18 The IPC uses a technology-oriented approach. (Annex 1.A2 provides an overview of the IPC system.) Since very detailed engineering-based descriptions are given for each class, this allows for the identification of technologies which have important environmental implications. Table 1.2 gives an idea of the hierarchical structure, taking the example of solar concentrating devices used for the generation of mechanical power. This example, taken from the case study on renewable energy, is relatively straightforward. All patent applications within class F03G 6/08 are clearly related to renewable energy, and there are likely to be very few patents involving solar energy concentrating technologies which do not list this class, particularly since each application can list multiple classes. A broader effort to identify “environmental” patents was also undertaken by the OECD, building upon a search algorithm developed at the Directorate for Science, Technology and Industry. Using the DSTI definition of “environmental” patents, counts of patents deposited by different countries in a number of different environmental areas have been derived. For most policy areas, Germany, the US, and Japan dominate (see Figure 1.6). There has apparently been continuous growth over recent years (particularly in air and water pollution innovations), except for solid waste and recycling, which peaked in the

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

35

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Table 1.2. IPC patent classification system for solar concentrating devices used for the generation of mechanical power Example of an IPC Code

Number of Subdivisions

Symbol

Title

Section

8

F

Mechanical engineering; lighting; heating; weapons; blasting

Subsection

21

F01-F04

Engines or pumps

Class

120

F03

Machines or engines for liquids; wind, spring, or weight motors; producing mechanical power or a reactive propulsive thrust, not otherwise provided for

Subclass

628

F03G

Spring, weight, inertia, or like motors; mechanicalpower-producing devices or mechanisms, not otherwise provided for or using energy sources not otherwise provided for

Main group

Ca. 6.900

F03G 6

Devices for producing mechanical power from solar energy.

Subgroup

Ca. 62.100

F03G 6/08

With solar energy concentrating means

Subdivision

Figure 1.6. Number of EPO “Environmental” patent applications and total EPO patent applications Waste disposal Recycling Air pollution abatement Water pollution abatement Noise control Monitoring Total (right axis) 600

120 000

500

100 000

400

80 000

300

60 000

200

40 000

100

20 000

0 1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

0

early 1990s. However, the rate of growth is generally lower than for overall TPF patent activity. Interestingly, there also appears to be some degree of specialisation that is occurring – with the US being particularly innovative with respect to water pollution; and Japan and Germany with respect to air pollution. In recent years, there have been relatively few patents in the area of waste disposal (Figure 1.7).

36

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Figure 1.7. Number of EPO “Environmental” patent applications DE-Waste US-Air

JP-Waste DE-Water

US-Waste JP-Water

DE-Air US-Water

JP-Air

160 140 120 100 80 60 40 20 0 1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

Source: Data drawn from the OECD Project on Environmental Policy and Technological Innovation (www.oecd.org/env/cpe/firms/innov).

Annex 1.A3 provides data for three further sub-areas of environmental technology and several additional countries. While such figures are interesting at a very aggregate level, they should be interpreted with some care. It may be relatively straightforward to identify relevant classes in areas in which innovations take the form of “end-of-pipe” pollution abatement – such as electrostatic precipitators or catalytic converters to remove air pollutants or membrane technology to remove water pollution. However, this is more difficult when the innovation takes the form of changes in production processes (“clean production”) or product characteristics (“clean products”).19 In this latter case, only a subset of applications within each class will be relevant, and it will be necessary to use keywords to identify both the relevant classes and the relevant applications within each class. Thus, a fourstage procedure was adopted in this report for the two cases in which process technologies are likely to be most important (pulp and paper bleaching technologies and motor vehicle emissions abatement): ●

compile a list of relevant keywords, based on a thorough literature review;

●

search patent database by keywords, in order to identify relevant patent classes;

●

test the “keyword yield” – both individually and in combinations with the patent classes identified;20

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

37

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

●

develop a final search strategy, on the basis of iterative assessment of the testing exercise above.

Since changes in production processes and product characteristics are increasingly prevalent as a means of addressing environmental concerns, the problem of defining an appropriate patent search strategy is likely to become more acute over time. Another problem related to this trend is that, generally, product innovations are patented more often than process innovations [for the latter, more informal ways of protecting knowledge often exist (Popp, 2005)]. With respect to environmental patents, this creates a potential bias towards inclusion of “end-of-pipe” technologies in patent counts, so efforts were made to correct for this in the search strategy adopted.

4. The policy determinants of eco-innovation To some extent, the policy factors which drive eco-innovation will be the same as those which drive the “rate” of innovation in general – i.e. stable macroeconomic conditions, economic openness, protection of intellectual property, etc. However, steering innovation in a “direction” which is less environmentally-damaging can also be done by using policy measures which target environmental externalities directly. Whether explicitly or implicitly, environmental policies result in either a change in the cost of different factor inputs or a change in the relative prices of goods and services that are eventually produced. As such, there are likely to be increased returns to the identification of production processes and product characteristics which are “environmentsaving”. In this section, the implications of different measures are discussed, along with relevant empirical evidence.

4.1. The theory of environmental policy choice and innovation Different policy instruments will affect the incentives for firms and households to develop and adopt environmentally-beneficial technologies in different ways. An ideal policy provides incentives to search for both the optimal rate of innovation, and the optimal direction of that innovation. Any policy measure which affects relative prices or constrains production choices will induce innovation. It is therefore not sufficient to assess the effects of a given policy on the rate of innovation, however those effects may be measured. Whether this innovation is welfare-improving will also depend on the direction of innovation induced (Johnstone, 2005). Efficiency with respect to the direction of innovation has two aspects. First, innovation in one area can “crowd out” innovation in other areas. If policy support for “environmental” R&D reduces R&D in other areas (i.e. health technologies), this needs to be taken into account in policy assessment. This issue has not been generally addressed in previous empirical work. Second,

38

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

within the “environmental” sphere, the direction of innovation should be costminimising with respect to the ultimate environmental objective. In this regard, the point of incidence of the policy is key, and it is widely argued that a policy which is technology-neutral and which targets the environmental objective directly, is more likely to result in a cost-minimising technological trajectory. A strong case has been made for the use of market-based instruments (e.g. taxes, tradable permits), rather than direct regulation (e.g. technologybased controls, performance standards) in order to induce innovation (see Jaffe et al. 2002 for a review). In particular, it is argued that the rate of innovation under market-based instruments is more likely to be optimal since a greater proportion of benefits of technological innovation and adoption will be realised by the firm itself than is the case for many direct forms of regulation. Moreover, since market-based instruments are not “prescriptive”, they are more likely than many types of direct regulation to ensure that the direction of technological change is cost-minimising with respect to the avoidance of damages (see Downing and White, 1986; Milliman and Prince, 1989; Nentjes and Wiersma, 1987; and, Jung et al., 1996). The incitation given to innovation in the US with respect to SO2 abatement – which followed the replacement of technologybased standards for a tradable permit regime for power plants – is an oft-cited example. However, different forms of market-based instrument may have different dynamic effects on the technological trajectory of the economy. For equivalent environmental targets, auctioned tradable permits and taxes are likely to have comparable effects. However, if policy targets are not adjusted in light of increased information on abatement costs or environmental impacts, the effects of the two measures may differ markedly – because one is a price-based measure and the other is a quantity-based measure. Thus, Jung et al. (1996) found that when governments pre-commit to a given tax rate (or to a given number of permits), the effect under the two regimes will differ since in the case of taxes the “price” of emissions remains constant, even as innovation reduces abatement costs. Moreover, with a grandfathered permit system, the innovator will face adverse financial effects from reduced permit prices. If the innovating firm is a seller of permits, it will have less incentive to allow for the diffusion to other firms (unless the innovation is patented, in which case, the incentives will depend on relative rates of return for permit and technology sales (see Milliman and Prince, 1989). Under these (rather restricted) conditions, grandfathered permits may perform worse than direct controls (e.g. mandated emission reductions) in terms of incentives to induce technological diffusion. With direct controls, the only costs will be those associated with abatement; under grandfathering, permit sellers will also lose from increased diffusion21 (Albrecht, 2001).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

39

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

However, the stark juxtaposition of the innovation effects of marketbased instruments and more direct forms of regulation is a caricature. It is important to compare measures not just in terms of the form they take (tax, standard, permit, etc.), but also their point of incidence in terms of proximity to the environmental externality. An “environmental” tax which only targets the environmental objective in a proximate sense is actually no more likely to lead to an optimal direction of innovation with respect to this externality than a technology-based standard is. In this case, innovation will be misdirected toward those technologies which save on the tax, but do not reduce environmental damages per se. Since there can be high administrative costs to target environmental externalities directly with environmental taxes and other market-based instruments, this is a significant concern. Analogously, a performance-based standard which targets the environmental objective directly will provide similar incentives for the direction of innovation to those of a tax which has the same point of incidence. For example, whether a tax is imposed on the lead content of fuels, or a performance standard is imposed in terms of lead content, innovators will seek to identify leadminimising technologies. However, for a given environmental objective, incentives with respect to the rate of innovation will still be less under a regulatory approach than under most market-based instruments. This is because, under performance standards, savings will only develop up to the point at which the regulatory standard is met. In general, technology-based standards suffer from the same shortcomings as performance-based standards with respect to the rate of innovation. Innovators (and adopters) do not have any incentives to go beyond the standard, unless they can induce a further regulatory response. However, by definition, technology-based standards are unable to target the externality directly. In order to induce an optimal direction of innovation, the regulator must also be able to identify the cost-minimising technological trajectory, and to adjust standards in line with this trajectory. This demands a great deal of the regulator. Indeed, one of the most important criteria by which to assess the innovation effects of different policy measures is the information requirements they demand of regulators. While the theoretical literature focuses mainly on the distinction between market-based instruments and direct regulations, there are, of course, a variety of other policy measures which are frequently applied. These each have different impacts on innovation. In particular, and as noted earlier, there are a number of policies which are specifically targeted at the innovation process. For instance, the use of subsidies in support of “environmental” R&D is common, whether in the form of grants or tax credits. There is some positive evidence of their impacts in the area of energy efficiency (Jaffe et al., 2002 and Jaffe et al., 2005). However, there are three main problems associated with these subsidies.

40

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

First, in the presence of asymmetric information between the regulator and potentially-innovating firms, it may be difficult for regulators to identify either recipients or promising technologies in an efficient manner – i.e. there may be problems of selection. Second, depending on the design of the subsidy scheme, there may be significant administrative costs for both the regulator and potentially innovating firms. And third, even if they provide incentives for the substitution of “dirty” technologies for “clean” technologies, they have positive scale impacts, increasing the output of the activities supported. On the demand side, some governments have introduced various types of “technology prizes” to encourage environmentally beneficial technological change. In some cases, a market is guaranteed for the successful invention, while in other cases, a financial prize is awarded directly. Perhaps the most famous example is the US “Super-Efficient Refrigerator Program” (SERP), which encouraged the development of a more energy-efficient refrigerator which was CFC-free (Newell and Wilson, 2005). The winning firm (Whirlpool) received some financial support, but the refrigerator did not actually perform very well in the market.22 Also on the demand side, public procurement can be used to induce environmental innovations. Starting in 1988, the Swedish Government established a major energy conservation programme which included about 30 technology procurement initiatives.23 Key elements in the programme were the bundling of demand between different consumers, the harmonisation of specifications for the product to be procured, a competitive tendering procedure based on these specifications, and a range of measures securing a certain initial market (Westling, 1996; Energimyndigheten, 1998). It thus combines demandpull and technology-push forces for innovation (see Ostertag and Dreher, 2002).24 An evaluation of the market transformation effects of five procurement projects showed that, on average, the procured technologies were 30% more efficient than prevailing products, their sales volumes were increasing, and (in some cases) unit-costs of technologies decreased (Lund, 1997). The use of information-based measures can also serve as a spur to innovation, by addressing information problems – either at the invention stage, or with respect to diffusion. For instance, demonstration projects are very common and have been used to spur investment in wind turbines in Germany (Walz, 2007). Targeting consumers directly through eco-labels can also have upstream impacts on invention. In a study of the determinants of the energy efficiency of air conditioners offered for sale, Newell et al. (1998) found that the energy labelling of appliances had a significant (and positive) effect on the efficiency of models offered for sale, thereby complementing energy price incentives.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

41

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

In Section II the benefits which might arise through government initiatives to support innovation partnerships or clusters were discussed. By bringing together innovating firms, university laboratories, and downstream users, some of the positive externalities which might otherwise be lost may be internalised. The government can serve an important “matchmaking” role in this context, by ensuring that positive spillovers are realised, without firms jeopardising competitive advantages. In the environmental sphere, two examples are the Danish “Partnerships in Innovation” (Danish EPA, 2007) and the Finnish “Environmental Cluster Research Programme” (Honkasalo and Alasaarela, 2003).

4.2. Empirical evidence with respect to patents There is increasing empirical evidence to support the contention that environmental policies do lead to technological innovation. Jaffe, Newell and Stavins (2002) and Vollebergh (2007) both provide recent reviews of the empirical literature on this theme. The following paragraphs draw on their results. An early paper was that of Lanjouw and Mody (1996), who examined the relationship between the number of patents granted and environmental policy stringency, measured in terms of pollution abatement expenditures at the macroeconomic level, for Japan, the US, and Germany. For the period 1971-1988, they found that pollution abatement cost affects the number of patents successfully granted, but with a 1-2 year lag. However, their study was not entirely satisfactory, because other factors that are likely to affect technical innovation were not controlled for in the analysis. Using US industry-level data, Jaffe and Palmer (1997) extended Lanjouw and Mody’s study, by incorporating various factors that potentially affects environmental innovation. They examined the relationship between stringency and innovation more broadly (not only environmental patents) for a set of US manufacturing industries in the period 1977-1989, where innovation was captured in terms of both R&D expenditures and patents. They found that increased environmental stringency (higher level of PACE) does increase R&D expenditures. But the study did not support the hypothesis that the number of patents increased in response to environmental regulation. They also stressed the necessity to assess the relative strength of the effects of flexible (versus prescriptive) environmental policy regulation regarding environmental innovation. Brunnermeier and Cohen (2003) built on Jaffe and Palmer’s work, by narrowing innovation down to purely “environmental” patents. They used US manufacturing industry data and empirically analyzed factors that determined environmental technological innovation – paying close attention to the fact that emission reduction pressures come not only from domestic regulatory

42

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

authorities, but also evolve from international competition. For indicators of emission reduction pressures, they used pollution abatement costs and the number of inspections undertaken by the direct regulatory institutions. Contrary to Jaffe and Palmer, they found that the PACE variable has a statistically significant (and positive) effect on environmental innovation, whereas subsequent monitoring does not. Moreover, they found that international competition stimulates environmental innovation; although once again, the effect of inspections was not confirmed. Taylor et al. (2003) studied the time path of innovation in sulphur dioxide (SO2) control, especially activities related to flue gas desulfurization. Analyzing a 100 year time span (1887-1995), they found that consistently more patent applications were placed after SO2 regulation was introduced in the 1970s. In addition to SO2 regulation, Popp (2006) also examined NOX regulation in the US, as well as the German and Japanese electricity sectors – to explore whether these regulations affected (inter)national innovation and diffusion. One of Popp’s main findings was that it is mainly domestic regulation that fosters innovative activities in the home country. But he also found an important role being played by foreign innovation in the development of these patents. Very few studies have examined the role of policy instrument choice. Using patent data, Popp (2003) examined the effects of the introduction of the tradable permit system for SO 2 emissions as part of US Clean Air Act Amendments on the technological efficiency of flue-gas desulphurisation. Comparing patent applications after the introduction of the tradable permit scheme with those submitted under the previous technology-based regulatory system, he found evidence of the improved removal efficiency of scrubbers. In one of the few available cross-country studies, de Vries and Withagen (2005) investigated the relationship between environmental policy regarding SO2 and patent applications in relevant patent classifications. Applying three different models which vary according to the manner in which policy stringency was modelled, they found some evidence that strict environmental policies lead to more innovation. However, they recognised that the modelling of environmental policy in their analysis required further refinement. Moreover, they expressed concerns about their ability to identify all relevant patent classes. Crabb and Johnson (2007) assessed the effects of fuel prices (and thus taxes) on innovation in automotive energy-efficient technologies in the US in the period 1980-1999. Using USPTO patent classes, they found that applications for patents on relevant automotive products and processes were induced by increases in domestic “wellhead” extraction costs (but not by increases in the import price of oil or the price of gasoline). This is consistent with the induced innovation hypothesis, if it is assumed that domestic sources can substitute for imported oil, at least temporarily.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

43

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

4.3. Summary It has generally been argued that environmental policy stringency should induce innovation in environment-related technologies. However, the nature of the policy instrument that is used seems to matter. Generally, market-based instruments are thought to provide stronger incentives for innovation than direct forms of regulation. Since there is an opportunity cost associated with any level of emissions, market instruments provide innovators with the potential to earn much greater returns. Moreover, incentives based on market approaches tend to be continuous, and not dependent upon continually “ratcheting up” the regulatory stringency. It is also frequently argued that market-based approaches are less prescriptive, resulting in a greater “space” across which innovations can be applied. This leads, in principle, to a cost-minimising technological trajectory. However, whether market-based approaches are, in fact, less prescriptive than regulatory ones depends in part on the policy design. The point of incidence of a policy is key, and there is no one-to-one relationship between instrument choice and point of incidence. This aspect is often neglected in the literature dealing with policy choice. However, the empirical evidence remains limited, both with respect to the overall effects of environmental policy on technological innovation, as well as the more specific question of the extent to which this is reflected in patent activity. The empirical evidence with respect to the use of other policy measures (subsidies for environmental R&D, technology prizes, communication tools, support for networks) is even more limited.

5. The case studies As noted earlier, case studies were undertaken for this report in three areas: i) renewable energy; ii) motor vehicle emissions abatement; and iii) bleaching technologies in pulp and paper. In all cases, an effort was made to assess the role of public policy (and other factors) on innovation. However, due to differences in the scope of the studies, different approaches were actually applied in each case (Table 1.3). Issues that have been addressed in the case studies include:

44

●

the effect of environmental policy stringency in inducing technological innovations which reduce environmental impacts;

●

the relative importance of different types of environmental policy instruments in encouraging these innovations;

●

the potential international market opportunities for new technologies which arise for those countries which introduce environmental policies;

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Table 1.3. Characteristics of the case studies Renewable Energy Generation

Wastewater from Pulp and Paper

Motor Vehicle Emissions Abatement

Technologies

Generation of renewable energy from wind, solar, geothermal, wave-tide, biomass, and waste-to-energy sources

Elementally-chlorine free and totally-chlorine free bleaching technologies

Post-combustion (e.g. catalytic converters) and integrated (e.g. on-board diagnostics) technologies

Major issues

Policy instrument choice; Type of innovation

Government policy vs. consumer pressure as market determinants; First-mover advantage

Type of innovation; International technology transfer

Countries

Austria, Australia, Belgium, Canada, United States, Brazil, Canada, Switzerland, Germany, Finland, Sweden, Germany, Denmark, Spain, Japan Finland, France, United Kingdom, Greece, Hungary, Ireland, Italy, Japan, Korea, Netherlands, Norway, New Zealand, Poland, Portugal, Sweden, Taiwan, United States

Germany, Japan, United States

Time period

1978-2003

1975-2003

1978-2001

Data source

OECD patent database

Delphion

Delphion

●

the roles of foreign policy and market factors on the generation of domestic environmental innovations;

●

the extent to which environmental innovations take the form of integrated product or process changes (rather than end-of-pipe solutions); and

●

the innovation impacts associated with regulating multiple pollutants from a single source in a coherent and integrated manner (rather than one-byone).

The main results from the work undertaken in these three areas are reviewed in the Chapters which follow.

Notes 1. See Vollebergh (2007) and Jaffe et al. (2002) for recent reviews. 2. Danish exports of wind power technologies to the US are also important, but patent family data was not collected in this case. 3. For a more detailed review of the empirical work in this area, see Syrneonidis (1996). 4. See Binswanger (1974), Binswanger and Ruttan (1978) and Hicks (1932). 5. www.oecd.org/department/0,2688,en_2649_34451_1_1_1_1_1,00.html.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

45

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

6. Exploration and exploitation of the earth, infrastructure and planning of land use, control and care of the environment, protection and improvement of human health, production, distribution and rational use of energy, agricultural production and technology, industrial production and technology, social structures and relationships, exploration and exploitation of space, defence. 7. Government budget appropriations or outlays for R&D provided by the Government Budget Outlays or Appropriations of R&D (GBOARD) database were used (OECD, 2005). 8. For further information on the survey, see Johnstone (2007) and www.oecd.org/env/ cpe/firms. 9. www.nsf.gov/statistics/nsf05305/htmstart.html. 10. Differences in sampling across countries mean that the data is not strictly comparable across countries. For the most recent version of the CIS, see Eurostat (2004). 11. For overviews of the use of patents as measures of innovation, see Adams (2006), Schmoch and Hinze (2004), Grupp (1998), and Grilliches (1990). 12. The table is based on Table 9.1 and 9.3 in Somaya (2000) and Gallini (2002). 13. See Park and Wagh (2002) and Park (2002). The index includes variables which reflect patent coverage, duration, and enforcement, as well as membership in international treaties, and protection from restrictions on patent rights. 14. Work is underway on the harmonisation of patent applicants’ names, so that patents by the same applicant have a single identifier. 15. For instance, the German Machinery and Industrial Equipment Manufacturers’ Association (VDMA) has noted that patents are an ineffective means of capturing rents when enforcement of IPRs is imperfect. See www.vdma.org/original. 16. Priority details include the Priority Application Number(s), the Priority Application Date and the Country in which priority application was filed (see Adams 2006, p. 21). 17. Despite this low figure in 2005, Japan – together with the US – was still the largest filer of patent applications in other countries, accounting for 23% of non-resident patent filings worldwide (WIPO, 2007). Underlying these figures is Japan’s generally very high number of patent applications. 18. However, care must be taken when choosing patent classifications, to ensure that they are valid for the entire time-frame being searched. Fortunately, IPC manuals include a record of when each classification was added. (The IPC is currently on its 8th edition.) Revisions are typically made every five years. Definitions of classifications that have been added since the 1st edition include numbers next to their definition, stating when the class was added. 19. See Popp (2005). 20. special algorithms can be used to calculate weights, based on the frequency of selected keywords in titles and abstracts. 21. It must be emphasised that this distinction is not important if innovations are generated by specialist firms which are external to the sector, and thus not themselves involved in the permit market. 22. A variant on “technology prizes” arises when the government agrees to acquire the right to a patentable invention and then to compensate the winner directly. In theory, this should spur both invention and diffusion, since the firm will not

46

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

capture monopoly rents once the invention has been commercialised. While this has been used in other areas (i.e. medicines), it has not been applied in the environmental sphere (see Mandel, 2005). 23. For an overview and evaluation of results, see Neij (2001), Suvilehto and Öfverholm (1998) and Lund (1997). 24. See BERR/DIUS/DEFRA (2007) for a discussion of the merits of such an approach in the UK context.

References Adams, S.R. (2006), Information Sources in Patents, Muenchen: K. G. Saur Verlag. Albrecht, J. (2001), “Tradable CO2 Permits for Cars and Trucks” in Journal of Cleaner Production, Vol. 9, pp. 179-189. Arrow, K.J. (1962), “Economic Welfare and the Allocation of Resources to Invention”, in R. Nelson (ed.) The Rate and Direction of Inventive Activity, Princeton University Press. Binswanger, Hans (1974), “A Microeconomic Approach to Induced Innovation” in The Economic Journal, Vol. 84, pp. 940-958. Binswanger, H.P. and V.W. Ruttan (1978), Induced Innovation: Technology, Institutions, and Development, John Hopkins University Press, Baltimore and London. Blind, Knut et al. (2006), “Motives to Patent: Empirical Evidence from Germany” in Research Policy, Vol. 35, pp. 655-672. Brunnermeier, S. B. and M.A. Cohen (2003), “Determinants of environmental innovation in US manufacturing industries” in Journal of Environmental Economics and Management, Vol. 45, pp. 278-293. Cohen, W.M. et al. (2000), “Protecting Their Intellectual Assets: Appropriability Conditions and Why US Manufacturing Firms Patent (or Not)”, NBER Working Paper No. 7552. Crabb, Joseph M. and D.K.N. Johnson (2007), “Fueling the Innovation Process: Oil Prices and Induced Innovation in Automotive Energy-Efficient Technology”, Working Paper, Colorado Department of Economics and Business, May 2007. Criscuolo, Chiara, Jonathan E. HASKEL and Matthew J. Slaughter (2005), “Global Engagement and the Innovation Activities of Firms”, NBER Working Paper No. 11479. De Vries, Frans P. and Cees Withagen (2005), “Innovation and Environmental Stringency: The Case of Sulfur Dioxide Abatement”, Center Discussion Paper #2005-18, Tilburg University. Denmark Environmental Protection Agency (2007), “Danish Solutions to Global Environmental Challenges” (DEPA). Dernis, Hélène and Dominique Guellec (2001), “Using Patent Counts for Cross-Country Comparisons of Technology Output” in STI Review 27 (www.oecd.org/LongAbstract/ 0,3425,en_2649_33703_21682516_1_1_1_1,00.html). Dernis, Hélèhne and Mosahid Kahn (2004), “Triadic Patent Families Methodology,”STI Working Paper 2004/2, Organisation for Economic Co-operation and Development, Paris. Downing, Paul B. and Lawrence J. White (1986), “Innovation in Pollution Control” in Journal of Environmental Economics and Management, Vol. 13, pp. 18-29.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

47

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Energimyndigheten (ed.) (1998), Procurement for Market Transformation for Energy-efficient Products (Eskilstuna: Energimyndigheten). Eurostat (2004), “Innovation in Europe: Results for the EU, Iceland and Norway”, Luxembourg, Eurostat. Falk, M. (2004), “What Drives Business R&D Intensity Across OECD Countries?”, WIFO Working Paper No. 236/2004. Frietsch, R. and U. Schmoch (2006), “Technological structures and performance as reflected by patent indicators”, in U. Schmoch, C. Rammer, H. Legler (eds.) National Systems of Innovation in Comparison. Structure and Performance Indicators for Knowledge Societies (Dordrecht: Springer). Gallini, Nancy T. (2002), “The Economics Of Patents: Lessons From Recent US Patent Reform”, in Journal of Economic Perspectives, Vol. 16, No. 2, pp. 131-154. Quevado, G.-Garcia (2004), “Do Public Subsidies Complement Business R&D? A MetaAnalysis of the Econometric Evidence”, Kyklos, Vol. 57, pp. 87-102. Goel, Rajeev K. and Edward W.T. HSIEH (2006), “On Coordination Environmental and Technology Policy”, in Journal of Policy Modelling, Vol. 28, pp. 897-908. Griliches, Zvi (1990), “Patent Statistics as Economic Indicators: A Survey”, in Journal of Economic Literature, Vol. 28, No. 4, pp. 1661-1707. Grupp, H. (1998), Foundations of the Economics of Innovation – Theory, Measurement and Practice, Cheltenham, Edward Elgar. Hall, B.H. and J. Van Reenan (2000), “How Effective are Fiscal Incentives for R&D? A Review of the Evidence”, Research Policy, Vol. 29, pp. 449-469. Hassan, E. (2003), “Mapping the Knowledge Base for Fuel Cells: A Bibliometric Approach”, IEA Conference on Innovation in Energy Technologies, Washington, Sept. 2003. Hemmelskamp J (1999), “The Influence of Environmental Policy on Innovative Behaviour: An Econometric Study”, IPTS Working Paper, Seville. Hicks, J.R. (1932), The Theory of Wages (Reprinted in 1948 by Macmillan, London). Honkasalo, Antero and Erkki Alasaarela (2003), “On the Cluster Approach to Environmental Research and Development”, Finnish Ministry of the Environment, Working Paper 653. Jaffe A., R. Newell and R.N. Stavins (2002), “Technological Change and the Environment”Environmental and Resources Economics, Vol. 22, pp. 41-69. Jaffe A.B. and R.N. Stavins. (1995), “Dynamic Incentives of Environmental Regulations: The Effects of Alternative Policy Instruments on Technology Diffusion”, in Journal of Environmental Economics and Management, Vol. 29, pp. 43-63. Jaffe, Adam B. and Karen Palmer (1997), “Environmental Regulation and Innovation: A Panel Data Study”, in The Review of Economics and Statistics, Vol. 79, No. 4, pp. 610-619. Jaffe, A.B. et al., (2005), “A Tale of Two Market Failures”, in Ecological Economics, Vol. 54, pp. 164-174. Jaumotte, F. and N. Pain. (2005a), “From Ideas to Development: The Determinants of R&D and Patenting”, OECD Economics Department Working Paper No. 457 [ECO/ WKP(2005)44]. Jaumotte, F. and N. Pain, (2005b), “An Overview of Public Policies to Support Innovation”, OECD Economics Department Working Paper No. 456 [ECO/WKP(2005)43].

48

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Jaumotte, F. and N. Pain, (2005c), “Innovation in the Business Sector”, OECD Economics Department Working Paper No. 459 [ECO/WKP(2005)46]. Johnstone, N. (2005), “The Innovation Effects of Environmental Policy Instruments”, in Jens Horbach (ed.) Indicator Systems for Sustainable Innovation, Berlin, SpringerVerlag, pp. 21-42. Johnstone, N. (2007), Environmental Policy and Corporate Behaviour, Cheltenham, Edward Elgar. Johnstone, N. and J. Labonne (2006), “Environmental Policy, Management and R&D”, in OECD Economic Studies, No. 43, 2006/2. Jung, C. et al. (1996), “Incentives for Advanced Pollution Abatement Technology at the Industry Level”, in Journal of Environmental Economics and Management, Vol. 30, pp. 95-111. Kivimaa, Paula and Per MICKWITZAKIVIMA (2006), “The Challenge of Greening Technologies – Environmental Policy Integration in Finnish Technology Policies”, in Research Policy, Vol. 35, pp. 729-744. Kivimaa, Paula and Per Mickwitzamakivi (2004), “Driving forces for environmentally sounder innovations”, in K. Jacob et al. (eds), Governance for Industrial Transformation, Berlin, Environmental Policy Research Centre. Kneese A. and C. Schultze (1975), Pollution, Prices and Public Policy, Washington, John Wiley. Lach, S. (2002), “Do R&D Subsidies Stimulate or Displace Private R&D? Evidence from Israel”, Journal of Industrial Economics, Vol. L, pp. 369-390. Lanjouw, J.O. and A. MODY (1996), “Innovation and the International Diffusion of Environmentally Responsive Technology,” in Research Policy, Vol. 25, pp. 549-571. Lanjouw, J.O. and M. Schankerman (2004), “Patent Quality and Research Productivity: Measuring Innovation with Multiple Indicators”, in Economic Journal, Vol. 114, No. 495, pp. 441-465. Lund, P.D. (1997), Evaluation of the Swedish Programme for Energy Efficiency: Successful Examples of Market Transformation Through Technology Procurement, Proceedings of ECEEE 97 Summer Study, Czech Republic, June 1997, 6 pages. MandeL, G.N. (2005), “Promoting Environmental Innovation with Intellectual Property Innovation: A New Basis for Patent Rewards”, Available at SSRN: http://ssrn.com/ abstract=756844. Meyer, Martin (2002), “Tracing Knowledge Flows in Innovation Systems”, in Scientometrics, Vol. 54, No. 2, pp. 193-212. Milliman S.R. and R. Prince (1989), “Firm Incentives to Promote Technological Change in Pollution Control”, in Journal of Environmental Economics and Management, Vol. 17, pp. 247-265. Neij, Lena (2001), “Methods to Evaluate Market Transformation Programs – Experience of the Swedish Market Transformation Program”, in Energy Policy, Vol. 29, pp. 67-79. Nentjes A, and D. Wiersma (1987), “Innovation and Pollution Control”, in International Journal of Social Economics, Vol. 15, pp. 51-71. Newell R., G.A. Jaffe and R. Stavins (1998), “The Induced Innovation Hypothesis and Energy-Saving Technological Change”, in Quarterly Journal of Economics, Vol. 114, No. 3, pp. 941-975.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

49

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Newell R.G. and N.E. Wilson (2005), “Technology Prizes for Climate Change Mitigation”, RFF Discussion Paper, June 2005. OECD (2002), Frascati: Manual Proposed Standard Practice for Surveys on Experimental Research and Development, OECD, Paris. OECD (2004), Science, Technology and Industry Outlook, OECD, Paris. OECD (2005), Research and Development Statistics, OECD, Paris. OECD (2006a), Economic Policy Reforms: Going for Growth, OECD, Paris. OECD (2006b), Main Science and Technology Indicators, OECD, Paris. OECD (2006c), Research and Development Statistics, OECD, Paris. OECD (2007a), Compendium of Patent Statistics (www.oecd.org/dataoecd/5/19/37569377.pdf) OECD (2007b), Science, Technology and Industry: Scoreboard 2007, OECD, Paris. Ostertag, K. and C. Dreher (2002), “Cooperative procurement: Market transformation for energy-efficient products” in P. Clinch et al. (eds.), Greening the Budget. Budgetary Policies for Environmental Improvement, Cheltenham, Northampton: Edward Elgar, pp. 314-332. Park, W. and D. Lippoldt (2007), “Technology Transfer and the Economic Implications of the Strengthening of Intellectual Property Rights in Developing Countries”, Trade Policy Working Paper, OECD, Paris, underlying data. Park, G. Walter (2002), “Patent Rights and Economic Freedom: Friend or Foe?”, in Journal of Private Enterprise, Vol. 18, No. 1, pp. 84-121. Park, G. Walter and Smita Wagh (2002), “Index of Patent Rights”, in James Gwartrey et al. (eds) Economic Freedom of the World: 2002 Annual Report (Cato Institute). Popp, David (2003), “Pollution Control Innovations and the Clean Air Act of 1990”, in Journal of Policy Analysis and Management, Vol. 22, No. 4, pp. 641-660. Popp, David (2005), “Using the Triadic Patent Family Database to Study Environmental Innovation”, OECD ENV/EPOC/WPNEP/RD(2005)2). Popp, David (2006), “International Innovation and Diffusion of Air Pollution Control Technologies: The Effects of NOX and SO2 Regulation in the US, Japan, and Germany,” in Journal of Environmental Economics and Management, Vol. 51, Issue 1, pp. 46-71. Rollinson, J. and D. Lingua (2007), “The Development of PATSTAT”, Presentation at the EPO and OECD Conference Patent Statistics for Policy Decision Making, Venice, Italy 2-3 October 2007. Rollinson, J. and R. Heijna (2006), “EPO Worldwide Patent Statistical Database (PATSTAT)”, Presentation at the EPO and OECD Conference on Patent Statistics for Policy Decision Making, Vienna, Austria, 23-24 October 2006. Schmooch, U. and S. Hinze (2004), “Opening the Black Box”, in H.F. Moed, W. Glänzel, U. Schmoch (eds.) Handbook of Quantitative Science and Technology Research. The Use of Publication and Patent Statistics in Studies of S&T Systems (Dordrecht: Kluwer Academic Publishers), pp. 215-235. Somaya, Deepak (2000), “Obtaining and Protecting Patents in the United States, Europe and Japan”, in Kagan, Robert A. and Lee Axelrad (eds) Regulatory Encounters (Berkeley: University of California Press).

50

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

M.-Suvilehto, H. and E. Öfverholm (1998), “Swedish Procurement and Market Activities – Different Design Solutions on Different Markets”, Proceedings of the 1998 ACEEE Summer Study on Energy Efficiency in Buildings, Berkeley, California and Washington D.C. UK BERR/DIUS/DEFRA (2007), Commission on Environmental Markets and Economic Performance: Report (November 2007), www.defra.gov.uk/environment/business/ commission/index.htm Vollebergh, Herman (2007), “Impacts of Environmental Policy Instruments on Technological Change”, OECD COM/ENV/EPOC/CTPA/CFA(2006)36/FINAL. Walz, Rainer (2007), “The Role of Regulation for Sustainable Infrastructure Innovations: The Case of Wind Energy”, in International Journal of Public Policy, Vol. 2, Nos. 1/2, pp. 57-88. Westling, Hans (1996), “Cooperative Procurement. Market Acceptance for Innovative Energy-Efficient Technologies” (Eskilstuna: Energimyndigheten).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

51

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

ANNEX 1.A1

Sources of Patent Data1 The TPF Database In the “Triadic Patent Family (TPF) Database”, patents are included if they have been deposited in the EPO, the Japanese Patent Office (JPO), and the US Patent and Trademark Office (USPTO). Because of the additional costs of filing abroad, along with the one-year waiting period that gives inventors additional time to gauge the value of their invention, only the most valuable inventions are filed in several countries. This ensures that the patents included in the TPF database are of wide commercial value. Moreover, filing a patent application in a given country is a signal that the inventor expects the invention to be profitable in that country. In addition to the geographic filter, a “consolidation” filter is applied (Dernis and Khan, 2004). This is necessary because different patent offices apply different rules for the “scope” of a patent. It is difficult to define the boundaries of an invention in a precise manner and different offices tend to apply different criteria. In particular the JPO often requires multiple patents for a single patent at the USPTO or the EPO. When comparing patent counts across countries this difference can result in a significant bias. For example, three separate priorities, representing three separate Japanese patents, might be shared by just one or two USPTO and EPO patents. In the TPF database, this problem is addressed by including in a single “family” all patents which share at least one common priority right. Thus, all later patents which are derived from this “original invention” are included in a single patent family. Any patents that are directly or indirectly linked to other patents are counted as a single patent family. For example, if patent A shares priorities 1 and 2, and patent B shares priorities 2 and 3, priorities 1 and 2 are directly linked, as are priorities 2 and 3. As they both share a direct link with priority 2, priorities 1 and 3 are indirectly linked. All patents sharing one of these three priorities are assembled in a single patent family in the OECD TPF database (Popp, 2005).

52

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

As a result of the application of these two filters, counts of environmental innovations using the OECD patent families will result in smaller counts than found when data from single patent offices is used. On the one hand, all patents which do not find wide commercial markets internationally will not be included. On the other, patent applications with a common “priority” will be bundled into a single family. As Popp (2005) makes clear, this has some benefits; however, it also has costs. Perhaps more significantly, the geographic and consolidation filters applied in the development of the TPF database may not be appropriate for the assessment of the effects of domestic environmental policies on innovation. For instance, there may be a greater tendency for innovators from “laggard” countries to deposit patents in all three offices, since their potential market will be greater than for innovators from “leader” countries. Popp (2005) found that the TPF data is downwardly biased with respect to German NOx and SO2 abatement innovation, because German environmental policy predated that of other countries, and innovators had little incentive to deposit patents in countries with “lax” environmental policies.

Esp@cenet An alternative data set is espacenet. This is a searchable on-line database provided by the European Patent Office (EPO) that can be accessed at http:// ep.espacenet.com/. While the website includes access to four patent databases (EP, WIPO, Japan, and worldwide), it is the worldwide database that is of most use.2 The worldwide database currently includes information on patents from 78 patent offices.3 The range of data availability varies by country, so researchers should check the availability before beginning research to determine the time frame suitable for analysis. The most relevant data limitations are that only recent data is available for some developing countries, and European Patent Classifications (ECLA) are not readily available for many non-EPO countries (with the US being an exception). This classification scheme is based on the often-used International Patent Classification system (IPC), but includes classifications at a greater level of detail. More importantly, as new classifications are added, the EPO updates the ECLA of older patents in its database. Researchers can use the esp@cenet search pages to generate a list of patents meeting a specific criterion (e.g. all patents from a specific country and classification in a given year). These searches can be done using the advanced search page. To specify a patent country, simply enter the two letter country abbreviation in the publication number field. This will yield all patents published in a specific country. Unfortunately, it is not possible to limit searches to all patents with inventors from a specific country. The closest

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

53

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

approximation is to use the priority number. Since most applicants first file in their home country, entering the same country in both the publication number and priority number will approximate the number of domestic patents. However, this will not yield a perfect count, as some patents have multiple priority numbers. For example, a search of US publications published in 2000 in ECLA class B01D 53/56 yields 35 patents. Limiting this to US priority numbers results in 14 patents. These 14, however, include one patent from Japanese inventors with an earlier Japanese priority number, as well as a patent from Canadian inventors with a US priority number. One other limitation to espacenet searches arises if the technology being searched includes several possible ECLA codes. Only four search terms can be combined at a given time. If more than 4 classifications are being searched, multiple searches will need to be run. However, these searches may yield duplicate results. For example, searches for ECLA B01D 53/34 and for ECLA B01D 53/56 may each yield a patent that includes both ECLA. Thus, simply adding the search results will lead to double counting. This can be avoided by saving the individual patent numbers generated from the search, and then manually removing the duplicates. Unfortunately, one drawback of the espacenet site is that data on individual patents cannot be downloaded. Thus, this would be done by first manually cutting and pasting the search results into a separate programme.

Delphion An alternative source that does allow users to download descriptive data is the Delphion on-line database.4 This database, available only by subscription, allows the user to download multiple patent records in machinereadable formats. Descriptive data available include the priority, application and issue dates for each patent, the home country of the inventor, and data on patent families. Unfortunately, updated ECLA classifications are not available from Delphion. Popp (2006) matched this data to lists of patents from relevant ECLA, by first getting lists of relevant patent numbers from espcenet. He then downloaded patents from Delphion that fell into related IPC classifications, and merged these data with the lists of patent numbers to restrict the analysis to those falling into the relevant ECLA classes. Various counts can be obtained from the raw data using standard data management software. Finally, like the espacenet data, the Delphion database does not include detailed descriptive data for Japanese patents. Some additional data on these patents can be obtained from the Japanese Patent Office (JPO) website.5 The JPO does not use the ECLA. However, it uses its own system: the F-term, which, like the ECLA, provides greater detail than the IPC. F-terms consist of a 5-digit theme code, a 2-digit viewpoint symbol, and a 2-digit number. For example,

54

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

consider technologies classified as 4D002 AA12. The 5-digit theme code 4D002 pertains to “processing of waste gases”. The two letter viewpoint symbol AA pertains to “compounds to be processed”. Finally, the 2-digit number specifies that NOX is the component processed. By searching by year, researchers can get lists of patents for a given application year. However, the priority date of these patents is not available, nor is it possible to separately identify foreign and domestic Japanese patents using this database.

The EPO/OECD World Patent Statistical Database (PATSTAT) Over the last several years, the OECD’s Directorate for Science, Technology and Industry, jointly with other members of the OECD Patent Statistics Taskforce,6 have developed a patent database that is suitable for statistical analysis – the OECD Patent Statistics Database.7 Recently, further work has been undertaken by the Taskforce members towards developing a worldwide patent database – the EPO/OECD World Patent Statistical Database (PATSTAT). The European Patent Office (EPO) took over responsibility for development and management of the database, which is drawn directly from the EPO’s master database (Rollinson and Lingua, 2007). The PATSTAT database has been developed specifically for use by (inter)governmental organisations and academic institutions, and is optimised for use in the statistical analysis of patent data (Rollinson and Heijnar, 2006). The PATSTAT database has a worldwide coverage (over 80 patent offices) over a time period stretching back to 1880 for some countries, and contains over 70 million patent documents. It is updated on a regular basis biannually. Patent documents are categorised using the international patent classification (IPC), European classification (ECLA),8 and national classification systems. In addition to the basic bibliometric and legal data, the database also includes patent descriptions (abstracts) and harmonized citation data.

Notes 1. This Annex is taken from Popp (2005) (www.oecd.org/dataoecd/35/33/38283097.pdf). 2. Note that the worldwide database also includes data on patents from the EPO, WIPO, and Japan. One difference is that the worldwide database does not include translations of patent abstracts from Japan, whereas the Japanese database does. 3. An updated list of available data can be found at http://ep.espacenet.com/ep/en/ helpV3/detailedcoverage.html. 4. Available at www.delphion.com. 5. This database can be found at www.ipdl.jpo.go.jp/homepg_e.ipdl.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

55

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

6. Other Taskforce members include the European Patent Office (EPO), Japan Patent Office (JPO), United States Patent and Trademark Office (USPTO), World Intellectual Property Organisation (WIPO), National Science Foundation (NSF), Eurostat, and DG Research of the European Commission. 7. The TPF Database is one of the significant outputs of this work. 8. The ECLA classification system is an extension of the IPC and contains 132 200 subdivisions (i.e. about 62 000 more than the IPC).

56

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

ANNEX 1.A2

Patent Classification Systems* The TPF database contains three potential classifications for each patent. First, the database includes US Patent Classifications (USPC) placed on US patents by the USPTO. Second, it contains International Patent Classifications (IPC) placed on the patent by both the EPO and the USPTO. Note that the IPC classifications assigned to patents in the same family may differ by patent office. The EPO uses the IPC as its main classification system. However, the USPTO relies on the USPC system. It assigns IPC classifications to patents based on a concordance between the USPC and IPC. However, these two systems differ in fundamental ways. The USPC uses a “function-oriented” classification system, whereas the IPC uses an “application-oriented” approach. As such, most researchers use the IPC classification system when searching for patents. In addition, since the mid-1990s, the EPO has made use of the European Classification System (ECLA). This scheme is based on the IPC, but includes classifications at a greater level of detail. This additional detail allows for separate identification of technologies, based on the pollutants they control. For example, IPC classification B01D 53/86 includes catalytic processes for pollution control. ECLA class B01D 53/86F2 specifies catalytic processes for reduction of NOX, and B01D 53/86B4 specifies catalytic processes for reduction of SO2. These classes can also help rule out irrelevant technologies. Using the example of catalytic converters from above, ECLA class B01D 53/94 and F01N 3/08B2 relate to catalytic converters for autos, and were thus not included in the Popp (2005) study. Moreover, as new classifications are added, the EPO updates the ECLA of older patents in its database. As EPO patents in the late 1990s included these extra letters at the end of the IPC record, the TPF database appears to use the ECLA for these more recent patents. However, it

* This Annex is taken from Popp (2005) (www.oecd.org/dataoecd/35/33/38283097.pdf).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

57

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

does not appear to include ECLA assignments given to older patents after their initial publication. Because the TPF database includes original patent classifications, care must be taken when choosing patent classifications to ensure that they are valid for the entire timeframe being searched. Fortunately, IPC manuals include a record of when each classification was added. The IPC is currently on its eighth edition. Revisions are typically made every five years. Definitions of classifications that have been added since the first edition include numbers next to their definition stating when the class was added.

58

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

ANNEX 1.A3

Number of EPO Applications in Different Environmental Areas

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

59

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Figure 1.A3.1a. AT IT

CH JP

DE NL

FR SE

GB US

Patent count 70 A. Waste disposal 60 50 40 30 20 10 0

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000 2002

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000 2002

1986

1988

1990

1992

1994

1996

1998

2000 2002

Patent count 100 B. Recycling 80 60 40 20 0

1978

1980

Patent count 120

C. Air pollution abatement

100 80 60 40 20 0

60

1978

1980

1982

1984

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENT ACTIVITY

Figure 1.A3.1b. (suite) AT IT

CH JP

DE NL

FR SE

GB US

Patent count 140 D. Water pollution abatement 120 100 80 60 40 20 0

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000 2002

1984

1986

1988

1990

1992

1994

1996

1998

2000 2002

1986

1988

1990

1992

1994

1996

1998

2000 2002

Patent count 40

E. Noise protection

30

20

10

0

1978

1980

1982

Patent count 60 F. Environmental monitoring 50 40 30 20 10 0

1978

1980

1982

1984

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

61

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Chapter 2

Environmental Regulation and International Innovation in Automotive Emissions Control Technologies by Frans de Vries (Department of Economics, Stirling University) and Neelakshi Medhi (Maxwell School, Syracuse University)*

This chapter focuses on innovative activities undertaken by the automobile industry in the US, Europe (with an emphasis on Germany), and Japan, focusing on the extent to which differential environmental policy measures have affected technological innovation in automotive emissions control technologies, within and across these regions. The chapter distinguishes between: a) different types of policy measures; and b) patents which relate to post-combustion technologies (such as catalytic converters) and more fundamental changes (such as engine re-design). Significant evidence of technology transfer is found. In addition, the increasing role of integrated abatement strategies is documented, with evidence that this is affected by both fuel prices and environmental policy design.

* Assistance comments and suggestions from David Popp Nick Johnstone and Ivan Hascic are gratefully acknowledged.

63

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

1. Introduction As noted in the Introduction, regulatory policies which provide strong incentives to undertake innovative activities are crucial for encouraging costeffective technological development in pollution control (e.g. Jaffe et al., 2003; Vollebergh, 2007 and references cited therein). This Chapter focuses on innovative activities undertaken by the automobile industry in the US, Europe (with an emphasis on Germany), and Japan, focusing on the extent to which differential environmental policy measures have affected technological innovation in automotive emissions control technologies, within and across these regions. Automotive emissions are among the most important sources of local air pollution. The major pollutants emitted by motor vehicles include carbon monoxide (CO) and nitrogen oxides (NOX). In OECD countries, car emissions account for 55% of CO and 36% of the ozone-causing NOX emissions.1 Other automobile pollutants include hydrocarbons (HC) and particulate matter (PM), contributing 21% and 12% respectively to air emissions in OECD countries (OECD, 2007).2 Given the relatively large contribution of automotive emissions to overall air pollution, reducing the amount of emissions generated by motor vehicles can contribute significantly to improving local air quality. The objective of this Chapter is to examine the links between environmental regulation and technological innovation in automotive emissions control. The following questions are addressed. ●

How did environmental regulation in the field of automotive emissions control develop over time (1970-2000) in North America, Europe and the Asia Pacific region?

●

What effect did environmental regulations in these regions have on the incentives to innovate in automotive emissions abatement technologies?

●

Is there significant evidence of international technology transfer with respect to different types of pollution control technologies?

●

Are there important differences in the factors which encouraged the development of “integrated” pollution control technologies and end-of-pipe solutions?

Patent applications and grants are used as a proxy for innovative activity. The use of patent statistics as an indicative measure of innovation is particularly interesting here, because these statistics are readily available, they are directly

64

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

tied to inventiveness (as a measure of R&D output rather than input), and they are based on an objective standard that changes slowly on the basis on novelty and utility (e.g., Grilliches, 1990). The availability of patent statistics allows the analyst to trace the patent classes that specifically contain technologies that are tailored towards controlling vehicle emissions, hence providing detail and coverage on the relevant technologies under investigation. Moreover, given the objective standard of assessing patents, these statistics provide some uniformity in comparing innovation across countries. By addressing innovation in emissions control technologies, this report complements previous reports (e.g. Taylor et al., 2003; Jaffe and Palmer, 1997; Brunnermeier and Cohen, 2003; Lanjouw and Mody, 1996) reviewed in the Introduction. In this case, however, the technologies relate to a final commodity, i.e. cars.3 As a consequence, car manufacturers in every country need to be concerned with regulations in both “home” and “overseas” markets. The structure of the Chapter is as follows. Section II gives an overview of the relevant regulations of the automobile sector in the US, Europe and Japan. In Section III, the different types of automotive emissions control technologies are discussed (both qualitatively and quantitatively), in terms of patent counts. Section IV presents preliminary results of formal estimations. Section V concludes and discusses policy implications.

2. Environmental regulation in the automobile sector There has been significant evolution in the key regulatory measures affecting the automobile sector in the US, Japan and the EU in recent decades. Alongside the summary provided here, Annex 2.A1 includes all the relevant legislation and corresponding pollutants that are currently regulated. For a more extensive discussion of regulation and other strategies to cut back emissions from motor vehicles, see OECD (2004).

2.1. Emission standards 2.1.1. US emission standards In principle, compliance with performance standards, such as emission standards for different types of pollutants, is “technology-neutral” – in the sense that car manufacturers can use the technology of their choice to reduce emissions. While car manufacturers based their compliance technology strategies generally on the use of catalytic converters in the early 1970s, by the early 1980s, some car manufacturers were applying three-way catalysts in a closed-loop emissions control system, using sophisticated electronic devices for controlling engine functions. Others relied solely on the use of three-way catalyst without these electronic devices (e.g. Bresnahan and Yao, 1985).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

65

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

The US standards have generally had a “technology-forcing” character, with the introduction of performance standards that cannot be met with existing technology and as such have not been demonstrated in practice (e.g. Gerard and Lave, 2005). “Technology-following” standards are less strict, and can be met with existing technology. It has been argued that this is the type of standards that was mainly adopted in Europe (Faiz et al., 1996). Figure 2.1 depicts how the US level of emission standards for HC and NOX evolved in the period 1970-2004 for gasoline-driven passenger cars. Figure 2.2 does the same for CO.4 Relative to the initial levels applied in the early 1970s, standards are now extremely stringent. Compared to the level of 1970, the 2004 HC standard was reduced by almost 97%, whereas the 2004 standard for NOX has been reduced by about 94%, relative to the level in 1973. The current (stringent) levels for both HC and NOX emissions represent the “Tier II standards”, which were phased in from 2004, as a follow up of the “Tier I standards”, which had come into effect in 1994. The Tier II for light-duty vehicles (cars and light trucks) was fully phased in by 2007; heavy light-duty trucks will be fully phased in by 2009. Regarding CO, the 2004 standard is 1.7 g/mile, implying a 95% reduction, compared to the 1970 level. Since 1994, standards with respect to PM have also been implemented, which are specified at a level of 0.08 g/mile.5 Figure 2.1. Evolution of US HC and NOX standards for passenger cars (gasoline) HC

NO x

Gram/mille 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 1970

1974

1978

1982

1986

1990

1994

1998

2002 Year

66

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.2. Evolution of US CO standards for passenger cars (gasoline) CO Gram/mille 40 35 30 25 20 15 10 5 0 1970

1974

1978

1982

1986

1990

1994

1998

2002 Year

2.1.2. Japanese emission standards In 1968, Japan introduced the Air Pollution Control Law. The subsequent standards, set in 1972, required a 90% reduction of CO and HC emissions by 1975, and a 90% reduction of NOX emissions by 1976 (Zhu et al., 2006). Additional Japanese automobile regulations are embedded in the 1992 Motor Vehicle NOX Law, which specified performance standards for NOX emissions from in-use vehicles. In 1996, in joint cooperation with the Japan Automobile Manufacturers Association (JAMA) and the Petroleum Association of Japan (PAJ), the Government established the Japan Clean Air Program (JCAP). Figures 2.3 and 2.4 how the change in Japanese emission standards for the relevant pollutants for gasoline and diesel engines, respectively. After the mid-1970s, the standards for gasoline driven passenger cars remained stable until 2000. The biggest increase in stringency was then implemented for CO emissions.6 The standards for HC and NOX followed the same pattern and have converged since 1978. The CO standard for diesel engines also coincides with the standard for gasoline in the period 1986-1999. Compared to the increased stringency level for gasoline driven cars in 2000 (from 2.1 g/km to 0.67 g/km), the CO standard for diesel became more strict in 2002 (from 2.1 g/km to 0.63 g/km). The HC and NOX standards for diesel are generally less strict than the corresponding standards for gasoline, except for 2005 where the HC standard for diesel became more strict (0.024 g/km for diesel versus 0.05 g/km for gasoline). Furthermore, since 1994 PM standards have been set for diesel, and were gradually reduced from 0.23 g/km to 0.0135 g/km in 2005; an increase in stringency of about 94%.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

67

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.3. Evolution of Japanese CO, HC and NOX standards for passenger cars (gasoline) CO

HC

NO x

Gram/km 2.5

2.0

1.5

1.0

0.5

0 1976

1980

1984

1988

1992

1996

2000

2004 Year

Figure 2.4. Evolution of Japanese CO, HC, NOX and PM standards for passenger cars (diesel) CO

HC

NO x

PM

Gram/km 2.5

2.0

1.5

1.0

0.5

0 1986

1989

1992

1995

1998

2001

2004 Year

2.1.3. EU emission standards The basis for EU-wide standards on automobile emissions is laid down in Directive 70/220/EEC. This (1970) Directive specified the maximum limits for CO and HC vehicle emissions. In subsequent decades, the stringency levels have been increased, through a series of amending Directives. Directive 83/351/EEC imposes further stringency levels on CO emissions and introduces

68

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

limits on the combined HC + NOX emissions; Directive 88/76/EEC also regulates these pollutants, as well as imposing restrictions on NOX emissions separately; Directive 91/441/EEC introduces standards for PM emissions. The most important amendments are Euro 1-Euro 5. For example, the 1992 Euro 1 performance standard forced manufacturers to install the three-way catalytic converters in gasoline vehicles. Since Euro 2 (1996), there have been different standards for diesel and gasoline vehicles, and the emission limits became more stringent with the introduction of Euro 2 to Euro 5. Figures 2.5 to 2.9 summarise the evolution of environmental stringency for different pollutants. CO standards for gasoline were already in place in 1971; for diesel, they were introduced in 1984. From that time on, the CO standards for both diesel and gasoline became more stringent, with the standards for diesel cars relatively more stringent than gasoline after 1995 (Figure 2.5). For both gasoline and diesel cars, the environmental stringency of CO emissions has involved a reduction of 97% in 2008, compared to their initial levels in 1971 and 1984, respectively. Figure 2.5. Evolution of European CO standards for passenger cars (gasoline and diesel) CO petrol

CO diesel

Gram/km 40 35 30 25 20 15 10 5 0 1971

1975

1979

1983

1987

1991

1995

1999

2003

2007 Year

Figure 2.6 shows the historical development of EU limits for HC and combined HC + NOX emissions. With respect to HC, the EU specified a limit for gasoline cars only for 1971-1983. In this particular period, stringency increased by almost 15% (from 2.49 g/km to 2.12 g/km). Since 1984, combined limits on HC + NOX emissions came into place, since they are synergistic precursors to ozone. Indeed, it is more efficient to regulate the sum of emissions (HC + NOX) than the two pollutants separately. For gasoline-driven cars, combined HC + NO X

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

69

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

standards were imposed until 1999; for diesel cars, they are specified until 2008. Figure 2.7 shows the development of environmental stringency of the combined HC+NOX standard for both gasoline and diesel passenger cars.7 The standard for HC + NOX emissions for gasoline cars was reduced by approximately 91% in the period 1984-1999 (from 5.43 g/km to 0.5 g/km). For diesel, the associated decrease in the HC + NOX standard in 1984-2008 was about 95% (from 5.43 g/ km to 0.25 g/km). Figure 2.6. Evolution of European HC and HC + NOX standards for passenger cars (gasoline) HC

HC + NO x

Gram/km 6 5 4 3 2 1 0 1971

1975

1979

1983

1987

1991

1995

1999 Year

Figure 2.7. Evolution of European HC + NOX standards for passenger cars (gasoline and diesel) HC + NO x petrol

HC + NO x diesel

Gram/km 6 5 4 3 2 1 0 1984

70

1988

1992

1996

2000

2004

2008 Year

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

As illustrated in Figure 2.8, since 2000, separate standards for HC and NOX emissions were again implemented for gasoline-driven cars. For these types of cars, the limits on NOX emissions are relatively stricter, as compared to HC emissions in 2000-2008. Figure 2.8. Evolution of European HC and NOX standards for passenger cars (gasoline) HC

NO x

Gram/km 0.25

0.20

0.15

0.10

0.05

0 2000

2002

2004

2006

2008 Year

The EU has also set very ambitious targets on PM emissions for diesel cars. The increase in environmental stringency is shown in Figure 2.9. The 1990 target of 0.27 g/km has been reduced by 98% to 0.005 g/km for 2008. As of 2008, gasoline-driven cars will also face the same PM standard. Figure 2.9. Evolution of European PM standards for passenger cars (diesel) PM Gram/km 0.30 0.25 0.20 0.15 0.10 0.05 0 1990

1992

1994

1996

1998

2000

2002

2004

2006

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2008 Year

71

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Comparing the policy history (from 1975 to today) between the three regions with respect to emission standards, one aspect immediately stands out. Whereas US and EU regulation show a more “gradual” adjustment towards higher stringency levels, the Japanese standards have been relatively constant for a longer period of time. However, the Japanese standards were initially set at a relatively more stringent level.8 For example, the Japanese CO, HC and NOx standards for gasoline vehicles did not change from the mid1970s until 1999. Only Japanese NOx and PM standards for diesel vehicles feature some gradual phase-in from the mid-1980s onward. The different temporal patterns of regulatory tightening may have some implications for the innovation process. In addition, the EU experience with “joint” standards for different pollutants (HC and NOx) may have implications for the nature of innovation. By encouraging an integrated strategy, such standards may have provided different incentives for innovation than under the “separate” standards which preceded (and followed) this particular policy strategy.

2.2. Other relevant regulations in the automobile sector 2.2.1. On-board diagnostic systems In addition to the use of emission standards, a growing field of technological innovation relates to the development and implementation of On-Board Diagnostic (OBD) systems. Two generations of OBD systems can be distinguished: OBD-I and OBD-II. The OBD-I system was the first generation of technology that makes use of electronic means to diagnose engine problems and to control engine functions, such as fuel and ignition. Sensors are also used to measure the performance of the engine, as well as the level of automotive emissions. In addition, the sensors are helpful in providing early diagnostic assistance (B&B Electronics, 2005). OBD-II systems are more sophisticated and ensure that vehicles remain as “clean” as possible over their entire life, by monitoring virtually every component that can affect vehicle emission performance. In case of a problem, the system automatically adjusts performance characteristics (EPA, 2004). In the US, the first implementation of OBD requirements occurred in California in 1988, and was implemented by the California Air Resources Board (CARB). This (OBD-I) system had to be installed on all 1988 and subsequent model-year vehicles. The requirements were specifically related to the control of fuel and ignition functions of the vehicle. Induced by this first generation of OBD systems in California, the 1990 US Clean Air Act Amendments (CAAA) then mandated the implementation of a more sophisticated and expanded OBD system for the US as a whole. This advanced (OBD-II) system, developed by the Society of Automotive Engineers (SAE), became the basis for the legal standard

72

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

set by the US EPA. The 1990 CAAA mandated that all light-duty vehicles and trucks built in 1996 had to be equipped with an OBD-II system. OBD requirements in Japan ran more or less in parallel with Japan’s pursuit of a more stringent environmental policy with respect to exhaust gas emissions, and which were introduced on 1 October 2000. Since then, OBD systems have also been explicitly mandated. However, by 1996 Japanese car manufacturers had already developed cars equipped with OBD-II systems similar to those mandated in the US. In the European Union the most stringent standards (Euro 3, Euro 4 and Euro 5 for the years 2000, 2005 and 2008, respectively) are included in Directive 1999/96/EC. Article 4 of this Directive states that “From 1 October 2005, new types of vehicles, and from 1 October 2006, all types of vehicles, should be equipped with an OBD system or an On-Board Measurement (OBM) system to monitor in-service exhaust emissions.” In addition to the monitoring of inservice exhaust emissions, monitoring of durability requirements and inservice control, and limits for non-regulated pollutants – such as Polycyclic Aromatic Hydrocarbons (PAHs), ultra-fine particulate and formaldehyde – that may become important as a result of the widespread introduction of new alternative fuels are also covered by this Directive. With respect to OBD requirements, Directive 98/69/EC (Euro 3 and Euro 4) already mandates the introduction of such systems for emission control (Euro 3). In summary, the forerunner in the development and implementation of both OBD legislation and OBD systems was the US, which introduced the first generation of OBD systems (OBD-I). At the federal level, the 1990 CAAA mandated the installation of the more advanced OBD-II system for 1996 model year vehicles. Although Japan formally mandated the implementation of OBD-II systems as of October 2000, Japanese automobile manufacturers had already installed this type of system by 1996. European legislation on OBD systems, included in Directive 1999/96/EC, lags behind the OBD policy implementation in both the US and Japan. The first European-wide installation of OBD systems was only required by 1 October 2005. Regulation tailored to OBD systems differs from regulation that is explicitly directed toward s reducing exhaust emissions from vehicles. The “environmental” motivation for OBD regulation seems to have been of a more “technology-following” character. For example, the principal motivation for developing OBD was related more to (general) improvements in engine performance and engine design, rather than on reducing pollution emissions per se. The latter effects were incidental benefits associated with the development of more advanced OBD systems. The policy implication is that the “public good” motivation for OBD regulation came into being once these systems revealed their potential for environmental benefits, but only after they had

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

73

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

already proven to be useful for “non-environmental” reasons (i.e. engine performance improvement). This makes the nature of OBD regulation fundamentally different from regulation mandating the development and installation of catalytic converters. In the former case, there is a mix of private and social benefits, while in the latter case, the benefits are exclusively social. Whereas post-combustion devices only generate environmental benefits at the “end-of-pipe”, the environmental benefits from OBD systems could be classified as being complementary with other financial benefits, such as increased fuel efficiency and reduced maintenance costs. Incentives for innovation in the two cases will be very different.

2.2.2. The lead phase-down Alongside CO, HCs, NOx and PM emission standards, lead (Pb) standards were also introduced over the course of the last couple of decades. The catalysts used to reduce other local air pollutants are rendered inactive by lead. This incompatibility necessitated the use of unleaded fuels in vehicles with catalytic converters, leading to a gradual phase-down of leaded gasoline in the US during the 1970s and 1980s (e.g. Kerr and Newell, 2003). The phase-out of lead in Japan began during the 1970s, and by the early 1980s, only 1-2% of gasoline contained lead (Michaelowa, 1997). The production and use of leaded gasoline has now been fully eliminated in Japan. In Europe, Germany was the first country to adopt standards to control the lead content of gasoline – ranging from 0.4 grams lead per liter in 1972 (benchmark of 0.6 grams per liter) to 0.15 grams of lead per liter in 1976. In 1985, Germany also passed a law to reduce total automobile emissions, and included the introduction of unleaded gasoline, because the largest reductions of NOx and CO could be achieved by catalytic converters that were incompatible with lead (Von Storch et al., 2002). As of 1981, the EU set a standard of 0.4 grams lead per liter (Council Directive 78/611/EEC), which lagged almost a decade behind the German law. From October 1989, all EU member States had to offer unleaded gasoline, with a maximum of 0.15 grams of lead per liter. The 1998 Aarhus Treaty mandated the use of only unleaded gasoline by 2005.

2.2.3. Fuel economy Emission standards as examined above represent an upper limit to the amount of a certain pollutant that can be emitted per vehicle kilometer (mile) driven. American Corporate Average Fuel Economy (CAFE) standards refer to the amount of fuel that is required for a vehicle to travel a certain distance (for example, in terms of litres/kilometer). Although fuel economy is obviously of primary importance to the automobile sector, this Chapter does not try to link

74

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

innovation in fuel economy of cars to CAFE regulations. Further OECD work looking at both fuel efficiency and emission of local air pollutants is currently envisioned.9

3. Innovation in automotive emissions control technologies 3.1. Automotive emissions control technologies Automobile pollution control technologies comprise all technologies that are used to reduce pollutants generated and released into the atmosphere by automobiles. These automotive-generated emissions fall broadly into two categories, based on the point of emission: a) tailpipe or exhaust emissions; and b) evaporative emissions. Tailpipe emissions are produced as a by-product of the (imperfect) combustion of fuels to power the vehicle.10 These comprise the pollutants that are released from the vehicle’s exhaust system. Evaporative emissions are produced as a result of the evaporation of fuel, due to heating of the vehicle or release of vapor while refueling. The gas tank heats up due to high day-time atmospheric temperatures and/or from engine heat (even after coming to a halt). Gasoline in the tank evaporates when the temperature of the tank increases, raising the pressure inside the tank, which has to be released and vented into the atmosphere. Any vacant space in a gasoline tank is filled with HC vapor, which is forced out of the tank and into the air during refueling. These particles are heavier, thus remaining closer to the ground level. As such, effective abatement of pollution from vehicles must target both tailpipe emissions and gas tank venting. Searches conducted for these technologies are primarily based on specific regulations that were imposed on the automobile sector like the US-Tier standards, as well as the EU’s Euro standards.11 Tailpipe emissions can be controlled by increasing engine efficiency, increasing vehicle efficiency, increasing driving efficiency, or cleaning up the emissions emitted. Of these four control methods, driving efficiency is not improved through technological innovations, but depends on nontechnological aspects, such as driving techniques and levels of congestion. These issues are not considered here. Aspects that increase vehicle efficiency – using light-weight materials or aerodynamic design of the external body of the vehicle – are not addressed. Which type of pollutant (and how much) is emitted from motor vehicles is largely determined by the type of vehicle engine installed and the type of fuel consumed. Two of the most common types of engines are spark-ignition engines and diesel engines. Emissions from spark-ignition engines can be reduced by re-designing the engine, changing conditions under which combustion takes place, and treatment of post-combustion of pollutants by

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

75

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

using catalysts before releasing into the atmosphere. Technologies that improve fuel efficiency also reduce exhaust emissions, while some emissions control requirements have also resulted in improved fuel quality and fuel efficiency. Advances in engine and vehicle technology continually reduce the amount of pollutants generated, but this is generally considered insufficient to meet environmental objectives. Technologies designed to react with, and thus clean up, the remaining emissions (such as catalytic converters, catalytic regeneration technology, and particulate filters) are therefore an important part of emissions control efforts. While tailpipe emissions result from the combustion process, evaporative emissions can arise even when the engine is idle. Technologies to control evaporative emissions require additional modifications which are not necessarily related to engine design.

3.2. Patent data: identification of classes and descriptive results 3.2.1. Identification of classes The World Intellectual Property Organization (WIPO)12 descriptions on the IPC classification (8th Edition) was used to identify IPC classes that matched the emissions control technologies described above. A total of 67 different IPC classes were identified that dealt with the purification of gases and emissions control (see Annex 2.A1). The 67 identified IPC classes are broadly categorised into the three major technology groups identified above: 1) those that relate to improvements in engine re-design, and therefore generate less emissions; 2) those that treat pollutants produced before they are released to the atmosphere; and 3) those that reduce evaporative emissions. Unfortunately, the latter category is somewhat opaque, because there is no IPC sub-classification that clearly defines improvements to nozzles and/or canisters. However, both modifications to fuel refilling nozzles as well as charcoal canisters show up under classes already grouped under some tailpipe emissions category. For example, charcoal canisters are included under IPC code F01N3/28. The three broad categories have been further subdivided into nine technology groups, reflecting the technologies discussed in the preceding Sections. Eight of these subgroups pertain to tailpipe emissions, and only one to evaporative emissions. Six subgroups are classified under “improved engine re-design” and two under “post-combustion devices”. Two-way, three-way, and lean-NOX catalyst/catalytic converters are grouped under one technology heading (“catalytic converters”). Similarly, electrical, electronic and plasmabased technologies are placed under one subgroup. Table 2.1 summarises the technologies covered in the analysis.

76

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Table 2.1. Technologies covered Improved engine re-design technologies

Post-combustion technologies

Air-fuel ratio devices

Catalytic converters and catalytic regeneration technology (CAT)

Exhaust gas recirculation (EGR) valves

Particulate filters and regeneration

Oxygen, NOX and temperature sensors Electronic control systems and plasma-based technologies Fuel injection systems Crankcase emissions and control

Finally, note that (except for the US) all patent data includes both granted patents and the number of patent applications; the US patents cover only granted patents. The period of focus is 1975-2001.13

3.3. Descriptive results This Section presents the main descriptive results of the patent searches. It shows the development of the number of patent filings over time in the different regions. Given differences in the data (i.e. only granted patents in the US), the results should be primarily interpreted in terms of trends, rather than absolute differences in patent counts between these regions. Details are also presented about who files where (i.e. which country is the home-country of the inventor – the “source country”) and in which country the inventors file for patent protection. The latter provides some insight into the role of regulation in market segmentation and technology transfer.

3.3.1. Trends in patent applications Figures 2.10-2.11 illustrate the number of patent applications that have been filed in the US and Germany. (Recall that the US patents only include those that have been granted, and not the total number of applications.) The Figures also show the evolution of the two main technology groups: engine redesign and post-combustion patents, as well as a group called “integrated”. This latter group includes both engine design and post-combustion treatment components. Therefore, they could not be assigned solely to one of the other two groups. Given the starting date of the examination period (1975), the first relevant implementation of standards in the US was the tightening of the upper CO limit in 1981. The second increase in the stringency of environmental policy concerned the HC and NOX standards in 1994. There was hardly any effect in the number of patent applications prior to the imposition of the 1981 CO limits. Between 1981 and 1994, however, a positive trend in all three types of patent applications can be discerned.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

77

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.10. Evolution of patent applications at the USPTO, 1975-2001 Post-combustion

Engine re-design Patent applications US 1 800 NO x HC 1 600

CO NO x

OBD-l

HC NO x

Integrated

OBD-ll

1 400 1 200 1 000 800 600 400 200 0 1975

1977 1979

1981

1983

1985

1987

1989

1991

1993

1995 1997

1999 2001 Priority year

A more prominent effect might be the anticipated impact of the OBD requirements. In the period prior to the implementation of OBD-I (1988), patents on engine re-design exhibited a moderate increasing trend. However, a significant increase in the trend rate of the number of patents related to engine re-design is evident, starting from 1992-1993. This increase makes sense, because OBD was specifically tailored towards diagnosing engine problems, as well as towards controlling engine functions. However, the beginning of the increase in the trend rate falls between the point of introduction of OBD-I and OBD-II, perhaps reflecting anticipation of the latter regulation. Even after its introduction, engine re-design patents exhibited strong growth. Engine re-design patents filed in Germany increased from 1992 onwards, preceded by a three-year decrease in patent activity (Figure 2.11). Inventors do not seem to have anticipated the introduction of stricter standards for CO, HC+NOX and PM emissions that were set in the EU in 1992. Rather, a higher degree of innovative activity took place after the standards were implemented. The increase in the number of patent applications in Germany continues after the implementation of stricter EU stringency levels for CO, HC + NOX and PM in 1996. The effects of these policy shocks on the innovative output in postcombustion technologies is less clear. There is a moderate positive trend, with less pronounced changes in the region of the policy shocks. This also holds for patenting activity related to “integrated” technologies: a steady increase can be seen in patent filings as of 1987. It is only after 1996 that a sharp increase in integrated technology patents is observed. In 1985, Germany was the first European country to pass a law to reduce total automobile emissions, including the introduction of unleaded gasoline.

78

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.11. Evolution of patent applications at the GPTO, 1975-2001 Engine re-design

Integrated

Post-combustion

Patent applications Germany EU: CO diesel/gasoline 1 400 EU: HC + NOx 1 200 Introduction unleaded gasoline in Germany 1 000

EU: PM diesel

EU: CO, PM, HC + NO x

EU: HC, PM, NO x , CO diesel OBD

800 EU: member states offer unleaded gasoline

600 400 200 0 1975

1977 1979

1981

1983

1985

1987

1989

1991

1993

1995 1997

1999 2001 Priority year

The latter was initiated because the largest reductions of NOX and other pollutants could be realised with catalytic converters (catalysts that were incompatible with the use of leaded fuel). Catalytic converters that could deal with lead were already in use in the US, because the US automobile industry was faced with relevant standards (Von Storch et al., 2002). From 1982-1985, there was a 236% increase in “post-combustion” patent filings. It is likely that this effect came about from inventors anticipating this new law, particularly since Germany had already drafted the law in 1984. Patents for engine redesign technologies also increased in the years prior to 1985, but not as much as catalytic converters: 22% in 1983-1985. Prior to the introduction of a stricter policy for NOX in Japan in 1978, there was actually a decrease in post-combustion patents. Japanese car manufacturers successfully developed the first three-way catalyst in 1977, perhaps partly in anticipation of US regulations. The first US-invented three-way catalyst was not commercialised until 1981 (Gerard and Lave, 2005). Following the US in their pursuit of a more stringent environmental policy, in 1978 and 1979, Japan implemented the most stringent limit on NO X emissions world-wide for gasoline and diesel engines. Domestic regulation also directly triggered the innovative anticipation of Japanese car producers that resulted in the development of the three-way catalytic converter (Zhu et al., 2006). In addition to the NOX regulation, Zhu et al. (2006) argued that other aspects played a role in the timing of the development of the new three-way catalyst in 1977. For instance, the three-way catalytic converter was preceded by the development of the “Compound Vortex Controlled Combustion” (CVCC)

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

79

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

engine. This passed US EPA certification in December 1972 (Zhu et al., 2006). This is a clear example of an innovation that was developed at least in part in response to foreign regulation. Despite its potential, the CVCC engine ultimately did not become the dominant technology because it could not effectively cope with fuel economy and engine power problems. Moreover, the engine encountered difficulties in meeting the tight 1976 US standards for HC, CO and NOX. The relative failure of CVCC technology also triggered Japanese competitors to come up with a substitute technology that would be able to meet existing regulatory requirements. Surprisingly, this particular technological achievement is not reflected in the number of patent applications, either in Japan or in the US. In fact, catalytic converter patents filed for protection in Japan actually decreased in the year of its introduction (around 1977), and continue to exhibit a decreasing trend over the entire period 1975-1984. It was not until 1985/1987 that Japanese patent applications started to grow again, probably because of the introduction of CO and HC standards for diesel-driven passenger cars. In the period 1975-1982, there was a decreasing trend for US patent applications related to catalytic converters. However, in 1979 and 1980, there was a slight increase in the number of patent applications related to catalytic converters, increasing from a total of 72 applications in 1978, to 78 in 1979 and 84 filings in 1980. Only in the EU patents was there an increasing trend in the number of patent filings over the whole period, starting in 1977. Figure 2.12. Evolution of patent applications at the JPO, 1975-2001 Engine re-design

Integrated

Post-combustion

Patent applications Japan 3 500 Introduction CO and HC diesel standards 3 000 NO x gasoline NO x diesel 2 500

US: OBD-ll

CO, HC, NO x gasoline

PM diesel

2 000 1 500 1 000 500 0 1975

80

1977 1979

1981

1983

1985

1987

1989

1991

1993

1995 1997

1999 2001 Priority year

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

With respect to trends in overall Japanese patent applications (Figure 2.12), it can be seen that after the implementation of the tighter NOX standard in 1978 “engine re-design” patents increased significantly and started to level off in 19831984. Around 1985, when maximum CO and HC emissions for diesel vehicles are introduced, there was a very sharp decrease in engine re-design patents, followed by a very steep increase in 1986, and a strong decrease in corresponding patents in 1987. Thus, inventors filing in Japan appeared to respond quite dramatically in terms of innovative output in the mid-1980s. The next Japanese policy shock occurred almost ten years later, with the introduction of PM standards in 1994. In the period before 1994, both postcombustion patents and integrated technology patents showed an increase, which started around 1987. In contrast, after 1990 engine re-design patents decreased until 1994. In 2000, Japan also imposed stricter CO, HC and NOX limits for gasoline cars. As of 1996, both engine re-design and integrated patents increased towards 2000; patents related to post-combustion devices exhibited a more stable and constant development over this period. Recall that although Japan mandated the implementation of OBD-II systems as of October 2000, this system was already installed by Japanese automobile manufacturers in their 1996 vehicles. The Japanese patent data shown in Figure 2.12 illustrates that, in the period 1995-2001, engine re-design patents and integrated technology patents increased by 44% and 208%, respectively. Finally, patents deposited at the EPO (Figure 2.13) can also be examined.14 Following Germany’s lead, the EU mandated that all Member States had to offer unleaded gasoline as of 1989. In 1989, the number of patents filed for engine re-design technologies (as well as post-combustion technologies) increased. However, immediately after 1989, there was a decrease in engine re-design patent filings, which lasted until 1992, when CO, combined HC + NOX and PM standards became more stringent. After 1992, engine redesign patents showed a very sharp increase, particularly after 1996, the year in which the stringency of mandated levels for CO, HC + NOX and PM was increased. Since 1997, integrated patents and post-combustion patents also tended to increase relatively more quickly, in comparison with the period before 1996. Figures 2.10-2.13 reveal some common developments. First, engine redesign patents were more common than patents for post-combustion devices. Second, all three technology classes had a positive trend, but the trend was most pronounced for engine re-design patents. Third, patents related to integrated approaches started to became much more common in the late 1980s and early 1990s. Fourth, for US, German, and EPO patents, the growth in innovative output was strongest in the period 1992-2000. For Japanese patents, the growth was less pronounced and started a bit later (around 1994).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

81

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.13. Evolution of patent applications at the EPO, 1975-2001 Engine re-design CO diesel/gasoline

Patent applications EU 1 000 HC + NO x 900

Integrated

Post-combustion PM diesel

CO, PM, HC + NO x

HC, PM, NO x CO diesel

800 Member states have to offer unleaded gasoline

700 600 500 400 300 200 100 0 1975

1977 1979

1981

1983

1985

1987

1989

1991

1993

1995 1997

1999 2001 Priority year

Within the engine re-design group, patenting is mainly concentrated on electronic control systems and plasma-based technologies, and fuel injection systems. However, as can be seen in Figure 2.14 there are differences across countries. Compared to the US, Germany, and the EU, there was a greater concentration in Japan on the development of electronic control systems, and less on fuel injection. Within the overall “post-combustion technology” group (not contained in Figure 2.15), innovation was largely focused on the development of Figure 2.14. Technology shares within engine re-design group at different patent offices (1975-2001) United States

%

Germany

Japan

EU

70 60 50 40 30 20 10

s

.

or ns Se

je Fu

el

in

on .c El

82

ct

tro

l

R ec

kc an Cr

EG

as

e

c. le re Ai

Ai

rf

ue

l

0

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

catalytic converters (about 95% in all four regions); the rest of the innovation was in the field of particulate filters.15

3.3.2. Source countries Figure 2.15 provides information on the source countries of USPTO, GPTO, JPO and EPO patents by type of technology. The first four columns on the lefthand-side of the first panel give the share of domestic (successful) US inventors that filed in the US, disaggregated according to the three different technological classifications, as well as in total. Surprisingly, more of the patents granted in the US came from Japanese inventors than from US inventors. German inventors, and “others” followed. For the different types of technology, domestic US Figure 2.15. Source countries for patents (1975-2001) Engine re-design % 60

Post-combustion

Source countries Unites States patents

% 70

Integrated

Total

Source countries German patents

60

50

50

40

40 30 30 20

20

10

10

0

0 USA

% 100

JPN

DEU

USA

Rest

Source countries Japanese patents

% 50

80

40

60

30

40

20

20

10

0

JPN

DEU

Rest

Source countries EPO patents

0 USA

JPN

DEU

Rest

USA

JPN

DEU

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

Rest

83

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

inventors took the lead with respect to post-combustion technologies, while Japanese inventors were granted more US patents in the fields of engine redesign and integrated technology approaches. Concentrating on Germany, the upper-right panel of Figure 2.16 illustrates that most of the patented innovations (almost 60%) in all three technology domains (as well as in aggregate) came from domestic inventors. Japanese inventors filed the next most frequently, followed by the “other” category. On average, US inventors filed least often for patent protection in Germany. The domestic bias was even more pronounced for Japanese patents (lower-left panel). More than 80% of patents filed in Japan come from Japanese inventors filing in their home country in all three technology domains. Figure 2.16. Average patent family size by country and year United States

Germany

Japan

Average family size 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999 2001 Priority year

European patents exhibited a somewhat different picture in this respect. The number of patent applications that came from different countries is more uniform. Overall, EPO patents consisted mainly of German inventions, followed by Japanese. However, compared to Germany, Japanese inventors filed more frequently for patents for integrated technologies. The US did the least patenting in this technology domain in Europe, but stood more-or-less on an equal footing with Japan regarding post-combustion patents. Patent data also provides some additional insight into the pattern of inventive knowledge flows across countries. International patent families are based on the original invention, which are filed in other countries, hence creating protection of the original invention in these countries. After the initial application, inventors can file for patent protection within one year.

84

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Since it is costly to file in multiple countries, international patent families can serve as an indicator of the relative international market potential of the invention. Figure 2.16 shows the evolution of the average international patent family size of domestic patents across both engine re-design and postcombustion technologies during the period 1975-2001. Japanese patents had the smallest patent family size. This implies that Japanese inventions were essentially only filed in Japan. This confirms the result that is reflected in the lower-left panel of Figure 2.15, namely that most of the patents filed in Japan came from domestic inventors across all technology domains. US and German inventors applied for patent protection in overseas markets to a greater extent. In addition, the variability in the size of the average Japanese patent family has been relatively less over time, with a small increase in the average family size after 1977. This might be due to the Japanese development of the three-way catalytic converter at that particular time. The average family size of US and German patents fluctuated more over time, but has remained largely in the 3-4 range, following more-or-less the same pattern. Both the US and German patents showed a decreasing trend in family size as of 1997.

3.3.3. Technology transfer Although Figure 2.16 provides information on the average size of the international patent family – thereby illustrating the propensity for domestic inventors to file abroad – it does not show which country files where for patent protection, and when. The latter information is summarised in Figures 2.17-2.20, which shows the change in patent filings over time, grouped by inventor country. The Figures provide a specific indication of the knowledge flows across the different countries for both engine re-design patents and postcombustion patents. The top panel of Figure 2.17 provides data on the number of US patents for engine re-design disaggregated by source country – domestic, German, and Japanese inventors and a “rest of world” category referred to as “other”. The lower panel presents the same information for post-combustion patents. The main policy shocks are also indicated in the different graphs. The top-panel reveals that, over time, most of the US patents in engine re-design came from Japanese inventors. Focusing on the OBD-I and OBD-II regulations in 1988 and 1996 shows that prior to 1988 the number of Japanese inventors filing in the US grew only marginally. One year before the introduction of OBD-I German inventors started to apply for patent protection more frequently. There also appears to have been some anticipation of the introduction of OBD-II on the part of

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

85

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.17. Source countries of USPTO engine re-design and post-combustion patents Germany

United States

Japan

Other

United States – Engine (re)design

Patent applications 700 NO x 600

HC

CO NO x

HC NO x

OBD-l

OBD-ll

500 400 300 200 100 0 1975

1977

1979

1981

1983

NO x

1987

1989

1991

1993

1995

1997

1999 2001 Priority year

United States – Post-combustion

Patent applications 160 140

1985

HC

CO NO x

HC NO x

OBD-l

OBD-ll

120 100 80 60 40 20 0 1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999 2001 Priority year

Japanese and German inventors. However, Japanese patenting in the US declined after the actual introduction of OBD-II regulation. Surprisingly, two years before OBD-II, the number of US patents that came from domestic inventors actually decreased, but started to increase significantly following the implementation of OBD-II. As can be seen in the lower panel, until 1986 the level of innovative output with respect to post-combustions technologies was relatively stable. However, it started to increase significantly after 1986. With respect to engine re-design, most of the patents that were filed in Germany came from domestic inventors (Figure 2.18). Until 1992, there was a slight increasing trend, but the rate of growth in the number of patent

86

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

applications became significantly higher after 1992. Although the first EU-wide installation of OBD systems was only required on 1 October 2005, these provisions are included in Directive 1999/96/EC, which is indicated by the vertical line in the top-panel of Figure 2.18. There was probably some anticipation on the part of German inventors with respect to EU policy, but it is difficult to trace the length of the anticipation period. Foreign patenting in Germany remained largely constant over time, perhaps with a slightly increasing trend.

Figure 2.18. Source countries of German engine re-design and post-combustion patents Germany

United States

Japan

Other

Germany – Engine (re)design

Patent applications 1 000 900

EU: OBD

800 700 600 500 400 300 200 100 0 1975

1977

1979

1981

1983

1985

1987

1989

1991

Germany – Post-combustion

Patent applications 250

1993

1995

1997

1999 2001 Priority year

1997

1999 2001 Priority year

EU: CO, PM, HC + NO x

EU: PM diesel Introduction unleaded EU: CO diesel/ gasoline in Germany gasoline

200

150

100

EU member states have to offer unleaded gasoline

50

0 1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

87

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

German post-combustion patents, as highlighted in the lower panel of Figure 2.18, followed a similar pattern. That is, domestic inventors were the prominent source of the patent filings in Germany. The degree of patent output increases over time, particularly after 1982. However, from 2000 there was a sharp decline. There appears to have been some anticipation in the development of post-combustion patents (mainly catalytic converters) prior to the German introduction of unleaded gasoline in 1985. Thus, German inventors mainly developed catalytic converters that could deal with lead by themselves without importing them from the US, which already had developed that specific Figure 2.19. Source countries of JPO Engine re-design and post-combustion patents Germany

United States

Other

Japan – Engine (re)design

Patent applications 600 500 400 300 200 100 0 1975

1977

1979

1981

1983

1987

1989

1991

1993

1995

Japan – Post-combustion

Patent applications 80 70

1985

Development Japanese three-way catalytic converter

1997

1999 2001 Priority year

CO, HC, NO x gasoline

Introduction CO and HC diesel standards

PM diesel

60 50 40 30 20 10 0 1975

88

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999 2001 Priority year

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

technology earlier (see e.g. Von Storch et al., 2002). As Figure 2.18 illustrates, this German landmark law brought the level of inventive activity in the field of post-combustion technologies to a higher level. Since then, the number of patent applications in Germany has increased more steadily, probably (partly) induced by more stringent EU regulations regarding CO, PM and combined HC+NOX standards. The effects of EU policy measures on foreign patenting are more difficult to disentangle. Figures 2.19 and 2.20 provide the same information for Japan. Foreign applicants are included in a separate graph (Figure 2.19) because of scaling, since the number of Japanese patent filings (Figure 2.20) is considerably greater. German inventors filed more frequently in Japan than US and “other” inventors. Moreover, the number of German patent applications seeking protection in Japan increased considerably after 1996. This difference is mainly due to engine re-design patents. While German inventors have filed significantly more over time in Japan in the field of engine re-design technologies compared to the US and “other”, there is no difference in post-combustion technologies. All three sources grew rapidly from the mid-1980s, the year in which Japan introduced CO and HC standards for diesel passenger cars. The variation in post-combustion patents filed by US, German and “other” inventors is likely to have been largely driven by the adjustment of the corresponding stringency levels of the standards for diesel engines, since the standards of CO, HC and NOX emissions for gasoline-driven passenger cars were not adjusted during 1978-1999. Figure 2.20. JPO engine re-design and post-combustion patents by domestic inventors Japan: Engine (re)design

Japan – Engine (re)design and post-combustion

Patent applications 4 000

Development Japanese three-way catalytic converter

3 500

Japan: Post-combustion

Introduction CO and HC diesel standards

CO, HC, NO x gasoline

PM diesel

3 000 2 500 2 000 1 500 1 000 500 0 1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

1999 2001 Priority year

89

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Figure 2.20 provides information on patent filings in Japan by domestic inventors. By the end of the 1970s, the degree of domestic patent applications in engine re-design technologies started to diverge significantly from the level of domestic post-combustion patents. Further, where the latter form of innovation showed a rather gradual increase over time in the number of patent applications, the pattern of engine re-design patents featured more variation, including a sharp peak in 1986. It is not certain if this sharp fluctuation arose from the introduction of the aforementioned CO and HC standards for diesel engines. This seems unlikely given, that the implementation of emission standards is more oriented towards the technological advancement of postcombustion techniques.

4. Empirical analysis and results The limited size of the sample and the difficulty in modeling policy measures in a commensurable manner across countries restricts the issues that can be addressed empirically. However, based on the data collected, some preliminary analysis was undertaken on the relative importance of environmental policies and other factors on both “post-combustion” and “engine re-design” patents. Given that the latter are likely to result in private nonenvironmental benefits, it is expected that factors other than environmental policy (i.e. fuel prices) are likely to play a more important role than in the former case. An econometric model was, therefore, estimated to evaluate the impact of environmental policy and market drivers as determinants of patenting activity in automotive emission control technologies. A dataset consisting of a panel of three countries (US, Japan, and Germany) over a period of 24 years (1978-2001) was used to estimate the regression models. The full model is specified as:

PATENTSit = f (EnvPolicyit , ENPRICEit ,VAit , PAT_TOTALit , α i ) + ε it where i = 1, 2, 3 indexes the country (US, Japan and Germany) and t = 1978, …, 2001 indexes the year. The dependent variable, a patent count (PATENTS), is the total number of (successful or unsuccessful) patent filings deposited, distinguished by country (US, Japan, and Germany), year (1978-2001), and technology group (postcombustion devices vs. engine re-design technologies). This distinction is made because it is expected that environmental policy and market drivers have different impacts on innovation in the respective technological groups. The proxy variable for environmental policy is based on the introduction of OBD requirements in the US. As noted above, the introduction of

90

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

regulations mandating OBD systems in the US appears to have had important cross-border implications for Japanese and European manufacturers. The early introduction of such regulations in the US, coupled with the size and importance of the US car market, was an important driver of innovations overseas. In this paper, the environmental policy shock is thus represented by the introduction of OBD-I and OBD-II requirements in the US (DUM1_USOBD, DUM2_USOBD) – two dummy variables that have the value “zero” for all years until a given OBD system was mandated in the US, and the value of “one” thereafter. The fuel price (ENPRICE) variable is constructed as the after-tax price of unleaded gasoline, and represents the market driver of innovations in the automobile industry.16 For instance, a recent study by Crabb and Johnson (2007) found that higher energy (oil) prices stimulate innovation in the field of energy-efficient automotive technology.17 Value added by the automobile industry (VA) serves as an indicator of the scope of technological opportunities, as well as the potential financing of expenditures for research and development in the industry.18 Total patents (PAT_TOTAL), measured as the total number of patent filings deposited with the three national patent offices (the US, Japan, and European patent office) in any technological area, serves as a proxy for the propensity to patent which varies across countries and over time. Fixed effects (i) are introduced to capture unobservable country-specific heterogeneity. All the residual variation is captured by the error term (it), which is assumed to be normally distributed. Empirical results for the two technology groups are presented below. The results for re-design technologies (Table 2.2) indicate that fuel price (ENPRICE) has a positive and statistically significant (at the 1% level) effect on patenting activity in engine re-design technologies. Conversely, neither of the coefficients of the two policy dummies (DUM1_USOBD, DUM2_USOBD) are statistically significant. The results also suggest that higher patenting activity in general (PAT_TOTAL), as well as higher industry value added (VA) both yield more innovations in the specific field of motor vehicle emission control. Contrary to the above, the estimates given in Table 2.3 suggest that innovation for post-combustion technologies is primarily driven by environmental policy shocks. The coefficients of the two dummy variables (DUM1_USOBD, DUM2_USOBD) are positive and statistically significant at the 1% and 5% levels, respectively. Conversely, fuel prices (ENPRICE) have a statistically insignificant effect on patenting activity with respect to postcombustion devices. The results also suggest that, while higher industry value added (VA) is associated with higher levels of patenting activity in automotive emission control, this is not the case for patenting activity in general

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

91

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Table 2.2. Fixed-effects model estimates for engine re-design technologies Variable

Coefficient

Std. error

t-statistic

Probability

C

5.103

0.219

23.279

0.000

VA

0.006

0.003

2.374

0.021

ENPRICE

1.287

0.249

5.163

0.000

PAT_TOTAL/1000

0.061

0.014

4.309

0.000

DUM1_USOBD

0.067

0.078

0.864

0.391

DUM2_USOBD

0.056

0.077

0.731

0.468

Fixed Effects _US–C

–0.096

_DE–C

–0.476

_JP–C

0.573

Observations

72

R-squared

0.920

Adjusted R-squared

0.911

Durbin-Watson

1.239

Notes: Dependent variable is a log of patent count for engine re-design and integrated technologies.

(PAT_TOTAL). This is also consistent with expectations, because regulationinduced patenting activity may be unrelated to general determinants of innovation. In terms of environmental policy design, the results of the two models would suggest that fuel taxes would be more likely to induce innovation with respect to engine re-design, while regulatory measures induce post-combustion innovation. These results are consistent with expectations. In effect, the Table 2.3. Fixed-effects model estimates for post-combustion devices Variable

Coefficient

Std. error

t-statistic

Probability

C

2.812

0.381

7.385

0.000

VA

0.019

0.005

3.937

0.000

ENPRICE

0.346

0.466

0.743

0.460

PAT_TOTAL/1000

0.093

0.021

4.490

0.000

Fixed Effects _US–C

–0.484

_DE–C

–0.209

_JP–C

0.693

Observations

72

R-squared

0.905

Adjusted R-squared

0.895

Durbin-Watson

1.246

Notes: Dependent variable is a log of patent count for post-combustion devices.

92

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

development of post-combustion technologies appears to be an activity which is separate from general innovative activity, while for engine re-design technologies which reduce pollution, this is not the case. Given the small size of the sample, as well as the simplicity of the model being estimated, these results should be interpreted with caution. For instance, with the potential for lagged impacts of these policies on innovation, it is difficult to be certain of the precise effect of a specific policy on patent activity.

5. Conclusions This report has examined the links between environmental policy and technological innovation, drawing on patent data. The report distinguishes between: a) different types of policy measures; and b) patents which relate to post-combustion technologies (such as catalytic converters) and more fundamental changes (such as engine re-design). While the evidence is preliminary, some interesting conclusions emerge: a) Environmental policy “shocks” do appear to generate innovations. However, the continuous increase in both policy stringency and patent activity makes it difficult to disentangle the precise effects of individual policies with such a small sample; b) The share of “engine re-design” patents has been growing much faster than patents for “post-combustion abatement” technologies. Whether this is attributable to environmental policy design warrants further study; c) Foreign regulatory pressures appear to have an influence on domestic innovation. For instance, Japanese inventors have played a lead role in the development of catalytic converter patents, even though the regulatory “shock” initially came from the US; d) There seems to be a distinct difference in the propensity for foreign inventions to be patented in the three markets. In the US, many of the patents came from foreign inventors, whereas in Japan, the vast majority of patents came from Japanese inventors. The role of policy design in explaining this could also be assessed in further work; e) Within the “engine re-design” category, Japanese inventors have been particularly strong with respect to electronic controls, whereas in Germany (and in Europe more generally), fuel injection patents were more common. The US had a strong comparative advantage in air-fuel ratio devices. f) It appears that fuel prices and general trends in innovation have played a stronger role in the development of “engine re-design” technologies than in “post-combustion” technologies, while regulatory pressures have been more important for the latter category.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

93

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

These latter results indicate that more flexible policies which encourage “integrated” solutions to environmental problems are more likely to allow for the realisation of economies of scope between general R&D expenditures on product design, and those which relate specifically to abatement control.

Notes 1. In the US, Europe and Japan, the contribution of road transport to CO emissions is 61%, 42% and 36%, respectively; the corresponding figures for NOX emissions are 38%, 40% and 28% (OECD, 2007). 2. In the US, Europe and Japan, the contribution of road transport to HC emissions is 25%, 20% and 6%, respectively; the corresponding figures for PM emissions in the US and Europe are 3% and 17% (OECD, 2007). 3. Crabb and Johnson (2007) examined the effects of motor fuel prices on patent activity in automotive energy-efficient technologies. While this is related to this Chapter the effects of environmental policy on technologies which are explicitly emission-reducing are not addressed. 4. Carbon monoxide standards have been included in a separate chart, due to differences in scale. 5. The NOX standards for diesel engines were the same as the standards that applied to gasoline until 1994. From that time on, diesel standards became a bit more lenient (1.0 g/mile for diesel versus 0.4 g/mile for gasoline). As of 1994, PM standards for gasoline have been set at 0.08 g/mile. 6. Note that for 2005, the value of the CO standard (1.15 g/km) is not included in Figure 2.3, because it is higher than the 0.67 g/km that holds for the period 2000-2004, implying a “break” in the time-series. This is due to changes in the test cycle (JAMA, 2005, p. 48). 7. Thus, the HC + NOX standard contained in Figure 2.6 is also included in Figure 2.9. 8. On the other hand, prior to reaching the standards that were in place as of 1975, Japan gradually set more stringent CO, HC and NOX standards for gasoline vehicles in the period 1965-1975. 9. This would bring together the insights from this Chapter and that of Crabb and Johnson (2007), which focuses on fuel efficiency. Such an approach would allow for a richer understanding of optimal policy design in the presence of pollutants which are emitted jointly in combustion. 10. Under perfect combustion, all carbon is converted into CO2, leaving no CO behind. 11. An important source of information about these technologies is the various publications and websites of automobile and parts manufacturers. Searches on websites of major environmental research institutes, as well as government (environment agency) websites were also used to develop this section. Faiz et al. (1996) contains much of the description of technologies that were developed prior to the first half of the 1990s. 12. www.wipo.int/classifications/ipc/ipc8/?lang=en. 13. Until the late 1980s, each Japanese patent application could only include one claim. Thus, a patent with four claims in the US would be four separate patents in Japan, for example. Even today Japanese patents tend to have fewer claims, as some applicants are still in the habit of having one (or at least just a few) claims on a single patent.

94

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

14. Note that in the figure patents deposited at the EPO do not include patents filed in Germany. 15. Many air-fuel ratio devices use electronic control, and as such a specific category (AIR ELEC) was created, including both air fuel ratio devices in general, as well as electronic controls. 16. Data was obtained from the IEA Energy Prices and Taxes Database (www.iea.org/ Textbase/stats/index.asp). 17. See also Popp (2002) for a general empirical exploration of the induced innovation hypothesis in energy-efficient technologies and energy prices. 18. Data was obtained from the OECD Structural Analysis (STAN) database (www.oecd.org/document/15/0,3343,en_2649_201185_1895503_1_1_1_1,00.html).

References B&B Electronics (2005), “OBD-II Background Information”, The OBD-II Home Page (www.obdii.com/background.html ). Bresnahan, T.F. and D.A. YAO (1985), “The Nonpecuniary Costs of Automobile Emissions Standards”, RAND Journal of Economics 16, pp. 437-455. Brunnermeier, S.B. and M.A. Cohen (2003), “Determinants of Environmental Innovation in the US Manufacturing Industries”, Journal of Environmental Economics and Management 45, pp. 278-293. Center for Automotive Research (CAR) Publication, April 2004 (www.cargroup.org/pdfs/ AICEFinalReport.PDF ). Concawe (Conservation of Clean Air and Water in Europe) (1994), Motor Vehicle Emission Regulations and Fuel Inspections – 1994 Update, Report 4/94, Brussels. Corning, website of components manufacturers. www.corning.com/environmental technologies/products-applications/. Crabb, Joseph M. and D.K.N. Johnson (2007), “Fueling the Innovation Process: Oil Prices and Induced Innovation in Automotive Energy-Efficient Technology”, Working Paper, Colorado Department of Economics and Business, May 2007. Environmental Protection Agency (2002), “OBD: Frequently Asked Questions”, Office of Transportation and Air Quality (www.epa.gov/otaq/regs/im/obd/f02014.pdf). EnvironmentaL Protection Agency (2004), “On-Board Diagnostics – Basic Information” (www.epa.gov/orcdizux/regs/im/obd/basic.htm). Environmental Protection Agency (2007), “Particulate Matter” (www.epa.gov/oar/ particlepollution/). European Union (1999), Directive 1999/96/EC of the European Parliament and of the Council, Article 4. Faiz, A., C.S. Weaver and M.P. Walsh (1996), “Air Pollution from Motor Vehicles: Standards and Technologies for Controlling Emissions”, The World Bank, Washington, D.C. Gérard, D. and L.B. Lave (2005), “Implementing Technology-Forcing Policies: The 1970 Clean Air Act Amendments and the Introduction of Advanced Automotive Emissions Controls in the United States”, Technological Forecasting and Social Change 72, pp. 761-778.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

95

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Griliches, Z. (1990), “Patent Statistics as Economic Indicators: A Survey”, Journal of Economic Literature, Vol. 28, No. 4, pp. 1661-1707. Grösslinger, E., K. Radomsky and M. Ritter (1996), “Corinair 1990 Summary Report”, under supervision of the European Environment Agency, Copenhagen. Jaffe, A.B. and K. Palmer (1997), “Environmental Regulation and Innovation: A Panel Data Study”, Review of Economics and Statistics 79, pp. 610-619. Jaffe, A.B., R.G. Newell and R.N. Stavins (2003), “Technological Change and the Environment”, pp. 461-516 in K.-G. Mäler and J. R. Vincent (eds.), Handbook of Environmental Economics, Vol. I, Elsevier, Amsterdam. Jama (Japan Automobile Manufacturers Association), Annual Reports 2000-2005 (www.jama.org/library/brochures.htm). Kerr, S and R.G. Newell (2003), “Policy-Induced Technology Adoption: Evidence from the US Lead Phasedown”, Journal of Industrial Economics, Vol. 51, No. 3, pp. 317-343. Lanjouw, J.O. and A. Mody (1996), “Innovation and the International Diffusion of Environmentally Responsive Technology”, Research Policy, Vol. 25, pp. 549-571. Manufacturers of Emission Controls Association (MECA): Publication #: EPA 400-F-92-007, Fact Sheet OMS-5 (August 1994): US EPA Office of Mobile Source (www.meca.org/ page.ww?name=Who+We+Are§ion=Organization+Info ) Matthey, J., Website of components manufacturers (http://ect.jmcatalysts.com/ technologies). Michaelowa, A. (1997), “Phasing out lead in gasoline – How developing countries can learn from the experiences of the industrialized world”, in: A. Faircloiugh (ed.), World Development Aid and Joint Venture Finance 1997/98, pp. 268-272. OECD (2004), Can Cars Come Clean? Strategies for Low-Emission Vehicles, OECD, Paris. Popp, D. (2002), “Induced Innovation and Energy Prices”, American Economic Review, Vol. 92, No. 1, pp. 160-180. Popp, D. (2006), “International Innovation and Diffusion of Air Pollution Control Technologies: The Effects of NOX and SO2 Regulation in the US, Japan and Germany”, Journal of Environmental Economics and Management, Vol. 51, pp. 46-71. Von Storch, H., C. Hagner, F. Feser, M. Costa-Cabral, J. Pacyna and S. Kolb (2002), “Curbing the Omnipresence of Lead in the European Environment since the 1970s – A successful example of efficient environmental policy, Working Paper of The GKSS Lead Project” (http://w3g.gkss.de/staff/blei/). Schwartz, J. (2003), “No Way Back: Why Air Pollution Will Continue to Decline”, AEI Press, Washington, D.C. Taylor, M.R., E.S. Rubin and D.D. Hounshell (2003), “Effect of Government Actions on Technological Innovations for SO2 Control”, Environmental Science and Technology, Vol. 37, pp. 4527-4534. Vollebergh, H. (2006), “Impacts of Environmental Policy Instruments on Technological Change”, OECD COM/ENV/EPOC/CTPA/CFA(2006)36/FINAL. Zhu, Y., A. Takeishi and S. Yonekura (2006), “The Timing of Technological Innovation: The Case of Automotive Emission Control in the 1970s”, IIR Working Paper WP No. 076-05, Institute of Innovation Research, Hitotsubashi University, Tokyo, Japan.

96

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

ANNEX 2.A1

Overview of Technologies and Corresponding Patent Classes A systematic overview is given below of the IPC classifications related to automotive emission control technologies. For example, class F015/ 00 pertains to patents that deal with “exhaust or silencing apparatuses”. These are followed by various sub-classifications (N01N5/02 etc.), listed in descending order of precedence. Before each of the IPC classifications, a short outline of the main technology is provided.

Exhaust emission control technologies: Improved engine design or engine redesign Air-fuel ratio devices Air-fuel ratio has an important effect on engine power, efficiency, and emissions. The ratio of air to fuel in the combustion mixture is a crucial design parameter for spark-ignition engines. A very early emissions control system – the Air Injection Reactor (AIR) – reduces the products of incomplete combustion (HC and CO), by injecting fresh air into the exhaust manifolds of the engine. In the presence of this oxygen-laden air, further combustion occurs in the manifold and exhaust pipe. Generally, the air is driven through an engine-driven smog-pipe and air tubing into the manifolds. This technology was introduced in 1966 in California, and was in use for several decades. With cleaner burning engines and better catalytic converters, this technology is no longer as prevalent. However, having an air-fuel mixture that has the exact amount of air to burn the fuel without either air or fuel leftover is “stoichiometric” and has a normalized ratio of 1. Mixtures with more air than fuel are therefore “lean”, while those with more fuel are not. Lean mixtures are more efficient than stoichiometric ones, and are termed “leanburn” engines.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

97

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

F01N3/05

Mechanical engineering; lighting; heating; blasting engines or pumps. Machines or engines in general. Gas-flow silencers or exhaust system apparatus for machines or engines in general; gas-flow silencers or exhaust apparatus for internal combustion engines. Exhaust or silencing apparatus having means for purifying, rendering innocuous, or otherwise treating exhaust. [N: by means of air e.g. by mixing exhaust with air, in tailpipes.]

F02M67/00

Apparatus in which fuel-injection is effected by means of highpressure gas, the gas carrying the fuel into working cylinders of the engine, e.g. air-injection type [N: fuel-gas mixture being compressed in a pump for subsequent injection into the engine].

F02M 23/00

Supplying combustion engines in general, with combustible mixtures or constituents thereof: engine pertinent apparatus for feeding, or treating before their admission to engine, combustion-air, fuel, or fuel-air mixture. Apparatus for adding secondary fuel-air mixture.

F02M 25/00

Supplying combustion engines in general, with combustible mixtures or constituents thereof (charging such engines F02B). Engine-pertinent apparatus for feeding, or treating before their admission to engine, combustion-air, fuel, or fuel-air mixture [N: treatment by admission of activating fluids]; Engine-pertinent apparatus for adding non-fuel substances or small quantities of secondary fuel to combustion-air, main fuel, or fuel-air mixture F02M43/00 takes precedence; adding secondary air to fuel-air mixture F02M23/00.

F02 M3/00

Mechanical engineering: combustion engines, carburetors combined with low-pressure fuel-injection apparatus. Idling devices, with means for facilitating idling below operational temperatures.

F02 M3/02

Preventing flow of idling fuel.

F02 M3/04

Under conditions where engine is driven instead of driving, e.g. driven by vehicle running downhill.

Exhaust Gas Recirculation (EGR) valves Engines produced after the 1973 model year in the US were fitted with exhaust gas recirculation valves on the intake manifold. The main purpose of these valves was to reduce NOx emissions, by re-introducing exhaust gases into the fuel mixture and lowering exhaust temperatures.

98

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

F01N5/00

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Exhaust or silencing apparatus combined or associated with devices profiting by exhaust energy (using kinetic or wave energy of exhaust gases in exhaust systems for charging F02B; predominant aspects of such devices, see the relevant classes for the devices).

F01N5/02

The devices using heat.

F01N5/04

The devices using kinetic energy.

Oxygen, NOX and temperature sensors Various sensors have been developed to provide feedback and increase competence of computerized controls. For instance, temperature sensors were initially used to gauge coolant temperature to send signals of radiator fluid temperature, to warn and thus prevent overheating. However, modernday sensors for temperatures monitors can determine critical factors like airfuel mixtures and timing advances. These sensors also need to be more durable and reliable, as they essentially send signals to a computerized system of controls, rather than just acting as a warning for the driver. Durable and precision sensors are critical to on-board diagnostic systems. Similarly, precision sensors are installed in order to detect levels of oxygen and NOx. Therefore, these sensors are tailor-made for different makes and models, to maximize the sensitivity and accuracy of readings. F02D41/14

Mechanical engineering; lighting; heating; weapons; blasting engines or pumps. Combustion engines (cyclically operating valves thereof, lubricating, exhausting, or silencing engines). Electrical control of supply of combustible mixture or its constituents. [N: characterised by the type of sensor used.]

Electronic control systems and plasma-based technologies For easier functioning of stoichiometric engines, electronic control systems (also called On-board Diagnostic (OBD) systems) were designed to control air-fuel mixtures. These systems can both measure the air-fuel ratio in the exhaust, and adjust the air-fuel mixture going into the engine. Japan has incorporated such computers for such control systems since 1978 and they have been in use in US engines since 1981. European cars manufactured in the late 1980s are also fitted with computerized control systems (Faiz et al., 1996). These computerized versions are also capable of handling the control of other features like spark timing, exhaust gas recirculation, idle speed, air injection systems and purging of evaporative canisters. These systems simultaneously inform the driver of malfunctions, and are more resistant to tampering and maladjustments than mechanical controls.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

99

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

OBD systems are being installed in all modern vehicles, to monitor the malfunctioning of various critical processes. These devices are sensitive to sulphur content in fuels; plasma-based technologies are being introduced to overcome this drawback. Advances in engine and vehicle technology continually reduce the amount of pollutants generated, but this is generally considered to be insufficient to meet emissions goals. Therefore, technologies that react with (and clean up) the remaining emissions are an important part of emissions control.

100

F02D41/00

Electrical control of combustion engines; electrical control of supply of combustible mixture or its constituents.

F02D41/02

By changing the composition of the exhaust gas, e.g. for exothermic reaction on exhaust gas treating apparatus.

F02D41/04

Electrical control of supply of combustible mixture or its constituents; introducing corrections for particular operating conditions (F02D41/14 take precedence).

F02D41/06

For engine starting or warming-up.

F02D41/10

For acceleration.

F02D41/12

For deceleration.

F02D41/14

Introducing closed-loop corrections.

F02D41/16

For idling.

F02D41/18

By measuring intake air flow measuring flow.

F02D41/20

Output circuits, e.g. for controlling currents in command coils (current control in inductive loads.

F02D41/22

Safety or indicating devices for abnormal conditions.

F02D41/24

Characterised by the use of digital means.

F02D41/26

Using computer, e.g. microprocessor.

F02D41/28

Interface circuits.

F02D41/30

Controlling fuel injection.

F02D41/32

Of the low pressure type.

F02D41/34

With means for controlling injection timing or duration (ignition timing F02P 5/00).

F02D41/36

With means for controlling distribution (arrangement of ignition F02P 7/00).

F02D41/38

Of the high pressure type.

F02D41/40

With means for controlling injection timing or duration.

F02D43/00

Conjoint electrical control of two or more functions, e.g. ignition, fuel-air mixture, recirculation, supercharging, exhaust-gas treatment (electrical control of exhaust gas treating apparatus per se F01N 9/00).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

F02D43/02

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Using only analogue means.

F02D43/04

Using only digital means.

F02D45/00

Electrical control not provided for in groups F02D41/00 to F02D43/00 (electrical control of exhaust gas treating apparatus F01N9/00; electrical control of one of the functions; ignition, lubricating, cooling, starting, intake-heating, see relevant subclasses for such functions).

F02M51/00

Mechanical engineering; lighting; heating; weapons; blasting engines or pumps. Combustion engines. Supplying combustion engines in general, with combustible mixtures or constituents thereof. Fuel injection apparatus characterised by being operated electrically.

F01N9/00

Mechanical engineering; lighting; heating; weapons; blasting engines or pumps. Machines or engines in general; gas flow silencers or exhaust apparatus for internal combustion engines [evacuation of the fumes from the area where they are produced B08B15/00; arrangements in connection with gas exhaust of propulsion units in vehicles B60K13/00. Electric control of exhaust gas treating apparatus; [N: electrical control of supply of combustible mixture or its constituents in relation with the state of exhaust treating apparatus F02D41/02C4].

Fuel injection systems Fuel injection systems were first widely marketed by Robert Bosch AG. These systems injected fuel continuously through nozzles at each intake port. The rate of injection in turn was controlled by varying the pressure supplied to the nozzles by an electric fuel pump. These fuel injection systems are now fully computerized. There are two types of injection systems – those with one or two centrally located fuel injectors; and those with multi-port fuel-injection systems. Electronic multi-port fuel injection systems either fire all the fuel injectors at once, or each injector is fired sequentially at the optimal time during engine rotation. Sequential systems allow for better air-fuel mixtures and therefore better performance and fewer emissions. F02M39/00

Mechanical Engineering; Lighting; Heating; Weapons; Blasting Engines or Pumps. Combustion Engines. Supplying combustion engines in general, with combustible mixtures or constituents thereof. Fuel injection apparatus with respect to engines; pump drives adapted top such arrangements.

F02M39/02

Arrangements of fuel-injection apparatus to facilitate the driving of pumps.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

101

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

F02M41/00

Fuel-injection apparatus with two or more injectors fed from a common pressure-source sequentially by means of a distributor.

F02M43/00

Fuel-injection apparatus operating simultaneously on two or more fuels or on a liquid fuel and another liquid, e.g. the other liquid being an anti-knock additive.

F02M43/02

Fuel-injection apparatus operating simultaneously on two or more fuels or on a liquid fuel and another liquid, e.g. the other liquid being an anti-knock additive: Pumps peculiar thereto.

F02M43/04

Fuel-injection apparatus operating simultaneously on two or more fuels or on a liquid fuel and another liquid, e.g. the other liquid being an anti-knock additive: Injectors peculiar thereto.

F02M45/00

Fuel-injection apparatus characterised by having a cyclic delivery of specific time/pressure or time/quantity relationship.

F02M47/00

Fuel-injection apparatus operated cyclically with fuel-injection valves actuated by fluid pressure.

F02M49/00

Fuel-injection apparatus in which injection pumps are driven, or injectors are actuated, by the pressure in engine working cylinders, or by impact of engine working piston.

F02M53/00

Fuel-injection apparatus characterised by having heating, cooling, or thermally-insulating means.

F02M55/00

Fuel-injection apparatus characterised by their fuel conduits or their venting means.

F02M57/00

Fuel injectors combined or associated with other devices.

F02M59/00

Pumps specially adapted for fuel-injection and not provided for in groups F02M39/00 to F02M57/00.

F02M61/00

Fuel injection not provided for in groups F02M39/00 to F02M57/00.

F02M69/00

Low-pressure fuel-injection apparatus.

F02M71/00

Combinations of carburetors and low-pressure fuel-injection apparatus.

Crankcase emissions and control The compressed gases that blow-by the piston rings in the crankcase mostly consist of unburned or partly burned hydrocarbons. Prior to regulations, such blow-by gases were vented into the atmosphere. To control these emissions, the crankcase vent port requires closing the vent port and venting the crankcase emissions back into the air-intake system. This is done by using a check valve.

102

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

F01M13/04

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Mechanical engineering; lighting; heating; weapons; blasting engines or pumps. Machines or engines in general; engine plants in general; steam engines. Lubricating of machines or engines in general. Lubricating internal combustion engines; crankcase ventilating or breathing [N: having means of purifying air before leaving crankcase, e.g. removing oil].

Post-combustion devices Catalytic converters, lean NOx catalysts, NOx adsorbers, Catalytic Regeneration Technology (CAT) A catalytic converter is a device that is placed in the exhaust pipe and is capable of converting various types of emissions into less harmful ones, depending on the catalyst used. The catalytic converter comprises a ceramic support with a wash-coat of (usually) aluminum-oxide, providing a large surface area. This comprises the substrate, which is then layered with the catalyst. Substrates are sometimes honeycomb-like structures with thousands of parallel channels. The walls of these channels provide the surface for precious-metal catalysts. The catalysts used are made of “noble” metals, like platinum, palladium, and rhodium, either singly or in combination. They convert noxious emissions into carbon dioxide, nitrogen and water vapor. Catalytic converters have greatly improved over time, and can be easily retrofitted onto exhaust systems, in order to reduce tailpipe emissions. An additional beneficial effect on pollution levels was that these catalysts are rendered inactive by lead. This incompatibility further required use of lead-free fuel in vehicles fitted with catalytic-converters, leading to a gradual phasing out of leaded gasoline. In addition, sulphur and phosphorus in fuels can also poison the catalyst, requiring further cleaning of fuels. Automobile engines are generally fitted with one of two common types of catalytic converters. These are the oxidation or two-catalysts and the oxidation-reduction or three way catalysts. 1. Two-Way Catalysts: In most gas steams, CO and HC can be removed by combining them with oxygen, using an oxidation catalyst. Oxidation catalysts use platinum or palladium or a combination of both. These catalysts increase the reaction rate between oxygen, hydro-carbons and carbon-monoxide present in the exhaust. Without catalysts, this chemical reaction is very slow, especially for HCs like methane and ethane in fuels, which are linked with shorter chains and are therefore more difficult to oxidize. 2. Three-Way Catalysts and NOx-adsorbers: Also known as “non-selective catalytic reduction (NSCR) catalysts”, these generally use a combination of platinum, palladium and rhodium. They are designed to simultaneously

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

103

2.

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

convert three pollutants to harmless emissions, namely CO, HC and NOx to carbon-dioxide, water, nitrogen and oxygen, respectively. Thus, these catalysts not only promote oxidation of HCs and carbon-monoxide, but when combined with rhodium, help to reduce nitric oxide (NO) to its harmless components nitrogen and oxygen. Under “lean burn” conditions, conventional three-way catalysts are not effective in reducing nitrous oxides into its components. This drawback had restricted the use of lean-burn engines in passenger cars. However, lean-burn engines are still desirable, because they are more fuel-efficient. Recent research has identified zeolite catalytic materials as being effective in reducing about 50% of nitrous oxide in “lean-burn” engines. These catalysts were first introduced in Japan (Faiz et al., 1996). Later improvements to this technology (under richer conditions) not only adsorbs the NOx with platinum and palladium, but first releases the NOx, and reduces it to nitrogen using the rhodium.

104

F01N3/035

Gas-flow silencers or exhaust apparatus for machines or engines in general; gas-flow silencers or exhaust apparatus for internal combustion engines. Exhaust or silencing apparatus having means for purifying, rendering innocuous, or otherwise treating exhaust. [N: with catalytic reactors.]

F01N3/10

By thermal or catalytic conversion of noxious components of exhaust by using other chemical processes, chemical aspects of catalytic conversion, e.g. using specified catalysts.

F01N3/20

Specifically adapted for catalytic conversion; methods of operation or regulation of catalytic converters.

F01N3/28

Catalytic reactors combined or associated with other devices, e.g. exhaust silencers or other exhaust purification devices.

B01D53/92

Of engine exhaust gases (exhaust apparatus having means for purifying or otherwise treating exhaust gases F01N 3/00).

B01D53/94

By catalytic processes.

B01D53/96

Regeneration, reactivation or recycling of reactants.

B01J23/42

Performing operations; transporting. Physical or chemical processes or apparatus in general. Chemical or physical processes, e.g. catalysis, colloid chemistry; their relevant apparatus. Catalysts comprising metals or metal oxides or hydroxides, not provided for in group B01J21/00 [mixed oxides: Platinum].

B01J23/44

Catalysts comprising metals or metal oxides or hydroxides, not provided for in group B01J21/00 [mixed oxides: Palladium].

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

2.

B01J23/46

ENVIRONMENTAL REGULATION AND INTERNATIONAL INNOVATION...

Catalysts comprising metals or metal oxides or hydroxides, not provided for in group B01J21/00 [mixed oxides: Rhodium].

Particulate filters and regeneration F01N3/00

Mechanical engineering; lighting; heating; blasting engines or pumps. Machines or engines in general. Gas-flow silencers or exhaust apparatus for machines or engines in general; gas-flow silencers or exhaust apparatus for internal combustion engines. Exhaust or silencing apparatus having means for purifying, rendering innocuous, or otherwise treating exhaust. [N: collecting or removing exhaust gases of vehicle engines on highways.]

F01N3/01

By means of electric or electrostatic separators.

F01N3/08

Exhaust or silencing apparatus having means for purifying, rendering innocuous, or otherwise treating exhaust. [N: for making innocuous using electric or electrostatic separators.]

Evaporative emission control technologies Charcoal canisters are not found as a separate IPC class or subclass and are included in categories covered in tailpipe emissions classifications – F01N3/2.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

105

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Chapter 3

Policy Versus Consumer Pressure: Innovation and Diffusion of Alternative Bleaching Technologies in the Pulp Industry by Popp and Tamara Hafner (Maxwell School of Public Policy, Syracuse University)*

In the late 1980s and early 1990s, concern over dioxin in both paper products and wastewater led to the development of techniques that reduced the use of chlorine in the pulp industry. In this chapter, the evolution of two completing bleaching technologies in five major paper-producing countries is reviewed. By the end of the 1990s, nearly all pulp production in these countries used one of these technologies. Evidence is provided that substantial innovation occurred before regulations were in place. Instead, pressure from consumers to reduce the chlorine content of paper appeared to drive the first round of innovation. However, while some companies chose to adopt these technologies in response to consumer pressure, not all firms differentiated their product in this way.

* The authors would like to thank Nick Johnstone and Ivan Hascic for helpful comments and assistance in locating data for the project and Dave Halliburton, Mimi Nameki and Grethe Torrissen for their assistance in clarifying regulations in their countries.

107

3.

POLICY VERSUS CONSUMER PRESSURE...

1. Introduction This Chapter uses patent data to examine the evolution of elemental chlorine free (ECF) and totally chlorine free (TCF) technologies used by the pulp and paper industry. In both cases, the technologies reduce (or eliminate) the use of chlorine in the bleaching stage of pulp production. Use of these technologies grew rapidly during the 1990s, beginning in the Nordic countries (Finland, Norway, and Sweden), then spreading to the US and Canada. One advantage of studying innovation on these technologies is that they are process technologies. Most previous studies of environmental innovation using patents examine end-of-pipe pollution control technologies, because it is easier to identify patents for specific end-of-pipe technologies than for modifications to the production process.1 However, for ECF and TCF technologies, there are welldefined patent classes related to these processes. Therefore, this study is among the first to study the evolution of a process environmental technology using patent data. Another advantage of studying ECF and TCF technologies is that they offer a window into the effects of different policy regimes on innovation. While there is a large theoretical literature on the different effects of various policy instruments on innovation, few empirical studies compare innovation under different types of policy incentives.2 Here, at least three types of policies are relevant. First, in each country, command and control regulation limits the amount of chlorine releases from the pulp bleaching process. In the US and Canada, national (or provincial, in the case of Canada) standards set the basic limits on chlorine usage. There is some variation across plants, as each plant operates under an environmental permit. In the Nordic countries, decisions on allowable emissions are made on a plant-by-plant basis as part of a plant permitting system. While national guidelines were eventually developed in Sweden, these play less of a role than the binding regulatory limits which exist in North America. In addition to these two regulatory systems, some European mills also chose to adopt TCF production because of consumer preferences in the European market (Reinstaller, 2005). Much of the early demand for reductions in chlorine came from consumers, rather than from regulators. Chlorine used in bleaching not only affects wastewater released from mills, but also persists in the final paper product (Galloway, Helminen, and Carter, 1989). Concerns over chlorine in paper products led to increased demand for chlorine-free

108

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

paper in the late 1980s and early 1990s. Product labelling requirements allow consumers to identify paper made without chlorine. A growing literature in economics looks at the possibility of voluntary provision of environmental quality by firms (see, for example, Lyon and Maxwell, 2002). One reason often proposed for such behaviour is that firms are responding to consumer demand. This Chapter considers the potential effects of such demand-side influences on environmentally-friendly innovation, and whether these demand-side influences are magnified by product labelling schemes that make information about paper quality more readily available to consumers. This Chapter also looks at the links between environmental regulation and innovation across countries. In previous work (Popp, 2006), it was found that innovations for sulphur dioxide and nitrogen oxide control at coal-fired power plants responds primarily to domestic regulation. Looking at patents in these fields, it was found that increases in patents assigned to own-country inventors when a country passes or strengthens environmental regulations for power plants, but little increase in innovations from other countries. An important question is whether this finding is robust for other technologies, or is unique to the electric power industry. One important difference between the electric power industry and the pulp and paper industry is that there is little trade in electric power. Regulations affecting air pollution from power production focus on the location of production. In contrast, while regulations addressing chlorine in pulp and paper production do focus on the location of production, the final paper product is a traded commodity, implying that consumer preferences for chlorine-free paper in trading partner countries may influence innovation in producer countries. Finally, this Chapter looks at who innovates. In the case of pollution control at power plants, innovation is not done by the plants themselves, but by firms that produce boilers and pollution abatement equipment. In the case of a new process, such as the ECF and TCF technologies, it would be expected that more innovation would be done by users of the process, rather than by a third party firm. This can be studied using assignee data from the patents.

2. The pulp and paper industry The pulp and paper industry consists of two main types of firms. Pulp and paper mills process raw wood fibre or recycled fibres to make pulp and paper. Converting facilities use these primary materials to manufacture specialised products such as paperboard products, writing paper, and sanitary paper (EPA, 2002). The focus is on pulp mills. Pulp mills are typically located near where trees are harvested. Sixty-five per cent of the world pulp market is supplied by the NORSCAN countries (the US, Canada, Sweden, Finland, and Norway) (Reinstaller, 2005).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

109

3.

POLICY VERSUS CONSUMER PRESSURE...

Table 3.1 summarises data for the main pulp producing nations in 2000, sorted by chemical pulp production and percentage of value-added from the pulp and paper industries.3 The first panel shows the top producers using chemical pulp methods. This panel includes countries from around the world. The second panel shows the top countries based on value added in the pulp, paper, paper products, and printing and publishing industry. Value-added data have been obtained from the OECD STAN database, and have two weaknesses. First, the data only include OECD countries. Second, the value-added data are for the entire paper industry, not just for pulp production. For example, some of the top 10 countries in terms of value-added have no chemical pulp production. Nonetheless, note that four countries (US, Canada, Finland and Sweden) appear in both panels. In addition, Japan comes close to being on both lists, as it ranks 12th among OECD nations in percentage of value-added Table 3.1. Pulp producers Top countries: pulp production Chemical pulp production (1 000 metric tons)

% Value added from Pulp and Paper

United States

48 198

2.25%

Canada

13 553

2.99%

Japan

9 792

1.73%

Sweden

7 979

3.52%

Finland

7 100

6.06%

Brazil

6 689

N/A

Russian Federation

4 195

N/A

Indonesia

3 626

N/A

Chile

2 220

N/A

France

1 817

1.50%

Top countries: % total value added from pulp, paper, paper products, printing and publishing Chemical pulp production (1 000 metric tons)

Percentage

7 100

6.06%

0

3.82%

Sweden

7 979

3.52%

Canada

13 553

2.99%

754

2.43%

0

2.35%

48 198

2.25%

1 190

2.02%

0

2.01%

1 774

1.84%

Finland Ireland

New Zealand United Kingdom United States Austria Netherlands Portugal

Source: Pulp production from FAOSTAT (2006). Value-added percentages are the authors’ calculations, based on data from the OECD STAN database.

110

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

from the pulp and paper industry. As it is also an important source of patents for ECF and TCF technologies, it is included as well.4 Because consumer pressure played an important role in the reduction of chlorine bleaching, it is also important to consider where paper produced in these countries is sold. Table 3.2 shows the percentage of exports going to each of the countries in the study, along with Germany, other EU countries, and the rest of the world. Germany is included separately because, as shown in the next Section, consumer pressure in that country played an important role in the diffusion of chlorine-free paper. These data show that much trade in paper products is regional. Most exports from Sweden and Finland go to other European countries. Most exports from Canada and the US flow between the two countries. Japanese paper production is primarily exported to other countries. Given that consumer pressure varied across the world, its effects are also likely to vary by region. Table 3.2. Percentage of exports to each country: paper and paperboard Exports to: Exporter

Year

Canada

Finland

Japan

Sweden

USA

Canada

1988

Finland

1988

0.9%

Japan

1988

1.5%

0.8%

Sweden

1988

0.3%

1.7%

0.8%

USA

1988

19.3%

0.1%

12.4%

0.4%

Canada

1993

0.0%

2.7%

0.0%

Finland

1993

0.5%

2.2%

Japan

1993

0.7%

0.1%

Sweden

1993

0.1%

2.2%

0.4%

USA

1993

27.5%

0.0%

9.2%

0.3%

Canada

1998

0.0%

1.6%

0.0%

Finland

1998

0.8%

1.7%

Japan

1998

0.8%

0.1%

Sweden

1998

0.1%

2.5%

0.3%

USA

1998

30.5%

0.1%

6.5%

0.0%

2.0%

0.0%

82.3%

1.2%

4.7%

9.8%

2.9%

2.9%

6.3%

12.4%

50.1%

24.5%

1.1%

18.2%

4.5%

8.6%

65.3%

4.6%

17.1%

58.7%

16.9%

2.8%

12.3%

52.9%

81.9%

0.8%

5.3%

9.3%

2.6%

7.7%

14.5%

51.9%

20.6%

0.2%

14.8%

1.7%

5.6%

76.8%

2.0%

19.3%

57.6%

18.4%

2.4%

9.6%

51.0%

87.1%

0.5%

3.1%

7.6%

3.0%

7.3%

16.1%

49.1%

21.9%

0.1%

21.4%

1.7%

5.9%

70.0%

2.0%

19.5%

60.3%

15.3%

2.1%

10.5%

50.2%

0.1%

Germany

Other EU

Other

Source: Authors’ calculations, using data from Comtrade (http://comtrade.un.org). Includes exports in SITC2 categories 641 (Paper and paperboard) and 642 (paper and paperboard, precut and articles of paper or paperboard).

2.1. Pulp and paper manufacturing Paper manufacturing first requires the preparation of cellulose-based fibres in a dilute slurry of about 0.5% solids. This ensures good separation of the fibres from one another, and extensive inter-fibre bonding when water is removed in the paper-making stage. Water is removed when the slurry is fed

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

111

3.

POLICY VERSUS CONSUMER PRESSURE...

to paper machines, drawing the fibres together, and forming a sheet. Combinations of suction, pressing and drying are used. Pulp fibres are obtained from wood, other fibrous materials and recovered paper products. The intended use of the paper products influences the types of fibre used. Wood can be turned into fibres by either chemical or mechanical means. For mechanical pulping, physical force breaks the wood into fine fibres with a composition similar to the incoming wood. In contrast, chemical pulping selectively attacks lignin polymers that hold the cellulose fibres together in the wood. Bleached chemical pulps are often referred to as “wood free” since all the lignin has been removed and the fibres have a different composition to the wood. In contrast, mechanical pulps closely resemble wood owing to their high lignin content. Certain use of paper requires high brightness and long life. This is provided by bleached chemical pulps which account for a large portion of world output. The absence of lignin and the pulps pure cellulose nature ensures brightness and permanence. Bright cellulose pulps are made using a two-stage process. The dominant pulping process is kraft which uses a mixture of sodium sulphide and sodium hydroxide to selectively dissolve away most of the lignin. Some lignin is retained following pulping to minimise damage to the cellulose fibres by the strong pulping solutions. The remaining lignin is removed by milder bleaching stages. Prior to the discovery of dioxins discovery, treatment with chlorine gas was the traditional approach used to solubilise most of the remaining lignin. This was then removed by subsequent sodium hydroxide washing. The small amounts of lignin remaining after hydroxide washing was removed by treatment with a substance known as chlorine dioxide ClO2 and further sodium hydroxide washing. Using C to represent elemental chlorine, D to represent chlorine dioxide, and E to represent caustic soda, the predominant five stage bleaching process in used could be denoted as CEDED (NorbergBohm, 1998; Reinstaller, 2005; and EPA, 2002).5 The dioxin-organochlorine issue emerged in the mid-1980s, following earlier concerns in the 1970s that lead to measures to address suspended solids and biochemical oxygen demand. This stimulated installation of reasonably efficient washing of chemical pulps prior to bleaching, providing good removal of spent cooking solutions. These were incinerated in recovery boilers. Incineration destroyed all environmentally damaging materials in pulping liquors, and allowed for the recovery of valuable pulping chemicals, providing energy to generate power and heat to operate the mill at the same time. Due to the chlorine content of bleach-plant effluents and their corrosiveness to the recovery systems, bleach-plant wastes remained the major effluent source in the 1980s. Primary and in most cases secondary

112

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

treatment had emerged as a means to minimise the environmental impacts of effluent from the bleach plant and other areas of the mills. The adverse impacts of bleach-plant wastes were not recognised until advances in chemical analysis methods and environmental science enabled detection of low concentrations of pollutants in the wastes. The dioxin and organochlorine of the 1980s stemmed from such advances. As such, in seeking to address both the dioxin issue, as well as organochlorines, many mills initiated pollution prevention based programs directed at avoidance of both dioxin and organochlorine formation. The formation of dioxins and furans presented a major product quality concern owing to risks attributed to use of the products. Ex ante prevention (rather than ex post treatment) was the only way to alleviate this. In essence, remedy of dioxin-furan in mill effluents was indirectly addressed by the common measures applied to prevent formation during manufacturing. However, the resolution of the organochlorine issue presented a more fundamental challenge to the sector as knowledge was lacking regarding the composition and specific effects on biota of organo-chlorine substances in the bleach-plant effluents. Over time Adsorbable Organic Halide (AOX), emerged as a frequently applied regulatory measure to address organochlorines. However the reliability of AOX as a measure of effluent toxicity was unclear. For example, the test did not differentiate between chlorine attached to low molecular weight and high molecular weight compounds, which had vastly different toxicities. Considerable scientific debates arose in the 1990s concerning the value of AOX as the primary target of regulatory standards (Harrison, 2002).

2.2. Elemental chlorine free bleaching Elemental chlorine free bleaching replaces the use of chlorine gas by chlorine dioxide in the first bleaching stage. Its use prevents formation of chlorinated dioxins and furans and results in a mix of organochlorine compounds that have lower toxicities than those produced using chlorine since the different molecular structure of chlorine dioxide greatly reduces levels of AOX. At many mills technologies such as oxygen bleaching and extended cooking were added to reduce levels of lignin in pulp prior to bleaching. Thus, the typical bleaching process changed from OCEDED to ODEDED. This reduced the amount of chlorine dioxide required and provided lower AOX discharge per tonne of pulp produced.

2.3. Total chlorine free bleaching Total chlorine free bleaching differs from ECF in that no chlorine compounds are used and use of elemental chlorine (such as chlorine dioxide as well as molecular chlorine as chlorine gases) are displaced. Hydrogen

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

113

3.

POLICY VERSUS CONSUMER PRESSURE...

peroxide (H 2O 2) and ozone (O 3) are used as substitute bleaching agents. Denoting peroxide as P, ozone as Z, and chelation as Q,6 a typical TCF bleaching sequence is OQPZP. Discharges of AOX from TCF processes are lower and the only AOX present in theory would come from trace amounts of organically bound material naturally present in mill process waters and incoming raw materials. TCF processing provides the ability to eliminate releases of all but the naturally present AOX and dioxin furans. However use of the process comes at the expense of reduced pulp brightness and poorer runnability, owing to slower drainage of water on the paper machine. Additionally the energy demand per tonne of TCF pulp made is higher than in the case of ECF and the yield of pulp per tonne of wood fed is lower. Nonetheless, despite these disadvantages, TCF technologies were adopted by some firms, particularly in the Nordic paper producing countries (Reinstaller, 2005).

3. Pollution and the pulp and paper industry As noted earlier, increased awareness of the links between chlorine and dioxin led to dramatic reductions in chlorine use by the pulp and paper industry. One notable feature of this reduction was that consumer demand played a critical role in both the innovation and adoption of chlorine-free technologies. A series of environmental concerns led to increased environmental awareness among consumers. Increasing consumer awareness and activism regarding the environment spurred governments into action, creating regulatory pressure on the industry to respond. More importantly, competition among major industry players to satisfy consumer demand for chlorine-free paper played a vital role. Companies took competitive advantage of the demand for environmentally-friendly products, through the use of environmental labelling. This in turn led to the diffusion of more environmentally-friendly technologies, even before regulations limiting chlorine bleaching were put in place, as companies tried to maintain (or increase) their share of the global market. Pulp and paper mill effluents have been of particular concern to environmental activists. Initially the concern was focused on discharges with high concentrations of biochemical oxygen demanding (BOD) substances (Burton, 1989). In the 1980s, the emphasis shifted to the release of halogenated organic compounds in effluent. Chlorinated organic compounds such as dioxins and furans (which are both recalcitrant and bioaccumulative) are byproducts of the bleaching process when chlorine gas is used as the bleaching agent. In 1980, the US EPA discovered furans and dioxins in paper mill waste; in 1983 dioxins were found in fish living downstream from pulp and paper mills. These studies were first reported to the public by Greenpeace in August 1987 (Gray, Lowther, and Todd, 1987). In addition, Greenpeace publicised

114

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

studies finding trace amounts of dioxin in consumer products such as diapers, milk cartons and coffee filters (Collins, 1992), creating consumer awareness of the environmental impacts caused by the pulp and paper industry. Some of the most publicised research on the accumulation and adverse effects of chlorinated organic compounds in the environment emerged from Sweden, and to a lesser extent Finland, in the mid-1980s. In Sweden, concerns about the discharge of chlorinated organic compounds in pulp mill effluents emerged in the late 1970s and reached very high levels by the mid-1980s (Galloway, Helminen, and Carter, 1989). In 1982, the Environment/Cellulose I project was initiated in Sweden to investigate possible links between chlorinated organics and possible adverse environmental effects. A 1983 report suggested that chlorinated organic compounds in the effluents of pulp and paper mills were to be blamed for the declining health of coastal waters. Governmentsponsored scientists did a more comprehensive assessment at the Norrsundet kraft mill at the Gulf of Bothnia in 1984. Researchers found altered fish populations with acute skeletal deformities and other adverse effects living in waters receiving mill discharges (Larsson, Andersson, Förlin, and Härdig, 1988; Thulin, Höglund, and Lindesjöö, 1988). Elevated levels of chlorinated organics were also found in sediments, but these effects were not found at non-kraft mills. Finally, in 1987 the discovery of dioxin in diapers prompted a call for the ban of chlorine in the bleaching of pulp to be used in the manufacture of disposable diapers (Anonymous, 1987). These findings led regulators to reassess discharge limits within the context of possible toxicological and bioaccumulative effects and pressure pulp and paper mills to address the problems associated with chlorine bleaching (Smith and Rajotte, 2001). Two other key events occurred in the German market, which was a major market for paper producers from Sweden and Finland. In 1989, a leading toilet tissue manufacturer, announced plans to abandon chlorine and ECF pulp altogether. All its competitors in Germany, Austria and Switzerland took similar steps within 3 months, switching to TCF or deinked secondary fibres. Arguably the most influential action occurred in 1991 when Greenpeace provided information on TCF technology and its benefits, including reply cards to the publishers of Der Spiegel, requesting that future issues be printed on TCF paper. This led to many publishers requesting TCF paper from their suppliers (Smith and Rajotte, 2001). In 1992, another influential campaign was started when a Swedish firm started its “Z pulp” campaign. This campaign publicised the company’s discussions with Greenpeace and embraced the goal of zero discharge. Most influentially, it adopted the stance that brilliant white paper might be poisoning its users (Smith and Rajotte, 2001). The importance of firms exploiting a perceived market niche, has therefore aided in the diffusion of new bleaching technologies throughout the global industry.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

115

3.

POLICY VERSUS CONSUMER PRESSURE...

3.1. Environmental labelling Linking chlorine to contamination in everyday consumer products helped drive consumer demand for paper products produced using chlorine-free technologies. However, for this demand to have an impact on production processes, it was important that consumers be informed as to the production processes used. For this, environmental labelling (ecolabelling) emerged as an early policy instrument to address increasing consumer demand for environmentally friendly products. Environmental labelling promotes more environmentally-friendly consumption for the consumer, in addition to acting as an economic instrument for the industry, which can tap into a perceived market niche for green products (Salzman, 1991). Environmental labelling is based mainly on the impact of the product on environment, and less so on the impact on the consumer. This type of labelling therefore represented a major change in product labelling and reflected the emergence of growing awareness of environmental issues among consumers, as well as their strong desire to participate in the protection of the environment through their consumption choices. Regarding pulp and paper products, most labelling schemes have tend to emphasize recycled fibre content, underscoring the perception that recycling is the best way to minimise the environmental impacts of paper products (Webb, 1994). Table 3.3 summarises the major labelling schemes, along with noting the limits on chlorine content for paper products for each ecolabel. Most labelling schemes began in the late 1980s. One prominent exception is Blue Angel. The earliest labelling scheme, Blue Angel was launched in Germany in 1978, with 200 labels across 33 product categories (Sammarco, 1997). The Blue Angel scheme did not cover chlorine usage in paper until February 1992, when a new category for newsprint was introduced. Many countries adopted similar schemes in the late 1980s, although requirements on chlorine content for paper typically came later. The Canadian government introduced Environmental Choice in 1988. Criterion for paper products were first proposed in 1991, stipulating limits for total absorbable organohalides (AOX), biological oxygen demand (BOD), and total suspended solids (TSS) in wastewater discharge, as well as requiring that bleached paper products not produce measurable concentrations of chlorinated dioxins in the wastewater or have an effect on rainbow trout (Webb, 1993). However, these proposed limits were postponed (Webb, 1994), so it was not until the second iteration of Environmental Choice in 1998 that dioxins were addressed.7 The Nordic Council launched the Nordic White Swan in 1989, with the first paper standards beginning in November 1991. Halogenated and aromatic hydrocarbon cleaning solvents and fluorescent brightening agents were prohibited in fine papers. While a product could be made from virgin and/or

116

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Table 3.3. Summary of Ecolabel programs related to pulp and paper manufacturing Blue Angel (Germany), begins 1978 • 1992: new category for newsprint: no halogenated bleaches Environmental Choice (Canada), begins 1988 • 1991: limits for AOX Nordic White Swan, begins 1989 • 1991: Chlorine bleaching prohibited • 1994: Revised to allow ECF Ecocheck (UK), begins 1991 • 1991: Included limits for COD, AOX, BOD and TSS in wastewater Green Seal (US), begins 1989 • 1992: Bathroom and facial tissue standards limit AOX to 1 kg/ton pulp • 1993: Standards for printing and writing paper prohibit chlorine bleaching • 1996: Chlorine bleaching prohibited for bathroom and facial tissues Eco-label (EU), begins 1993 • 1996: Criteria for copy papers includes pass/fail system in four areas, including AOX Ecomark (Japan), begins 1989 • 2004: Chlorine gas not to be used in bleaching process

recycled fibres, chlorine-based bleaching agents were banned for recycled fibres. For pulp and paper production in general, the Nordic Swan limited AOX releases to 0.5 kg/ton. A 1994 revision of the criteria for the Nordic White Swan label removed the ban of chlorine bleaching of recycled fibres and increased the constraint on the use of chemicals with a general ban on the use of chemicals containing more than 1% of any substance that has been classified as harmful to the environment by the EU (Webb, 1994). This new limit allows the use of ECF, rather than TCF bleaching, and had been criticised by environmental groups as being too lenient. Additionally, Sodrä Cell and the Swedish Society for Nature Conservation have argued that too many products qualify for the label, thereby decreasing its value in encouraging companies to improve their environmental performances (Sammarco, 1997). In the UK, Ecocheck was introduced in 1991 (Webb, 1993), and included limits for chemical oxygen demand (COD), AOX, BOD, and TSS in wastewater discharge; as well as sulfur dioxide and nitrogen oxides emission into the atmosphere (Webb, 1996). The US launched Green Seal in 1989. Green Seal is administered by a private US environmental labelling agency. Its first paper standards covered bathroom and facial tissues, and were introduced in 1992. Chlorine bleaching was allowed until 1996, if the wastewater AOX was below 1 kg/ton pulp, but banned thereafter. It also stipulated that the tissues be made of 100% waste paper with 10-20% post-consumer waste content. Standards for printing and writing paper were established in 1993, which included a complete ban on chlorine-containing bleaches (Webb, 1994).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

117

3.

POLICY VERSUS CONSUMER PRESSURE...

Japan introduced the EcoMark label in 1989. However, limitations on the use of chlorine were not part of the criteria until 2004, when the use of chlorine bleaching was not allowed for products receiving this label. Documentation on the EcoMark label notes that chlorine content were not considered because dioxin pollution had already been addressed by all the relevant emitters.8 The Eco-label launched by the EU in mid-1993 has experienced many difficulties. Forest management criteria were a major sticking point because they required all wood to be sourced from managed forests following the criteria established under the Helsinki Guidelines (Webb, 1996). Furthermore, this label was criticised as being biased towards recycled paper grades because manufacturers of tissue paper made from 100% virgin fibres were concerned that their tissues would not meet the AOX limits. At the global level, the Ecolabel was viewed as a form of protectionism against imported virgin pulp products, and therefore as a trade barrier (Sammarco, 1997). The criteria for toilet tissues and kitchen towels were finalised in 1994, but the first paper label was only awarded in December 1995. By mid-1996, agreement was reached on the criteria for copy papers based on a pass/fail system in four areas: COD and AOX content in wastewater discharge, sulfur-compounds air emissions, energy consumption (Webb, 1996).

4. Regulatory responses Increased awareness of the links between chlorine and dioxin also spurred governments into action, creating regulatory pressure on the industry to reduce the use of chlorine in pulp production. In the Nordic countries, regulation is done on a plant-by-plant basis.9 In Sweden, permits, which are reviewed every ten years, are issued by the National Licensing Board for Environmental Protection. Regulatory authority regarding pulp mill effluents comes from the Environmental Protection Act of 1969, which emphasized prevention (instead of control) of pollution. Two key principles played a role in innovation and adoption of pulp and paper technologies in Sweden. First, regulators adopted a substitution principle, which urged the industry to replace harsh chemicals with less benign ones wherever possible. Second, best available technologies were used as the starting point for approving permit applications (Smith and Rajotte, 2001). Whereas other countries relied on the industry to control effluent parameters, Sweden aimed at preventing discharge (Burton, 1989). This was done through in-process changes such as adopting the oxygen delignification process to recycle the waste stream. Such changes gave the industry experience with mill retrofitting and sourcing the best available technology and made chlorine-free technologies available when consumer demand for chlorine-free pulp reached its peak in the early 1990s (Smith and Rajotte, 2001).

118

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Because of the use of plant-by-plant licensing, Sweden has not imposed national discharge standards. However, legislative action has been used to set national goals for environmental performance. In 1987, the Swedish Environmental Protection Agency (NV) established goals for organochlorine substances. In 1992, Parliament established more stringent national goals, stating that the pulp and paper industry should work to attain no noticeable effect of effluents by the end of the century (OECD, 1999a). Recommended limits for AOX releases from kraft pulp mills were just 0.1-0.2 kg/t (OECD, 1999a). The final requirements of each permit are developed after negotiation with each plant. The focus is on application of the best available technology (BAT), which the NV defines as the “best technology used on a commercial scale at a similar plant anywhere in the world” (OECD, 1999b, p. 176). By 1990, ECF was considered to be BAT. The Licensing Board also considers technical, environmental, and economic factors. Typically, economic considerations focus on effects on the industry as a whole, rather than on a specific plant. Under special circumstances, a plant may be given more time to implement needed upgrades (OECD, 1999b). In contrast to Sweden, Finland moved more slowly towards chlorine reduction. Like Sweden, Finland has no regulation specifically limiting AOX emissions, as it issues permits to plants on a case-by-case basis. In 1988, Sweden proposed new discharge limits for chlorinated organics for Nordic states at a meeting of Nordic Ministries (Smith and Rajotte, 2001). Finland disagreed with these limits, opting to set less stringent targets for the kraft pulp industry in 1989, limiting AOX releases to 1.4 kg/ADt by 1994.10 Finland later accepted more stringent performance targets developed by a Nordic Working Group in 1993. These targets limited AOX releases to 0.2-0.4 kg/t for bleached kraft mills, with the more stringent guidelines applying to new mills (OECD, 1999b). Unlike the Nordic countries, both the US and Canada have binding regulations limiting AOX emissions. Individual permits are still needed for each plant, but the national performance standards must be met or exceeded. In the US, it was not until the Cluster Rules of 1997 that a binding limit on AOX releases took effect. Prior to this (1984), the EPA established an ambient water quality standard for dioxin of 0.013 ppq. However, pulp mills were not covered as they were not a known source of dioxin at that time. A follow-up study completed in 1989 confirmed that pulp mills were an important source of dioxin. The EPA responded by initially requiring pulp mills to meet the 0.013 ppq ambient standard in their wastewater, although this requirement was later eased to 1.2 ppq (Norberg-Bohm and Rossi, 1998). The EPA first proposed regulations for AOX releases in 1993. These standards could not have been met using existing ECF technology, suggesting that TCF

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

119

3.

POLICY VERSUS CONSUMER PRESSURE...

would be considered the BAT in the US. The final rules11 established ECF as the “BAT”, with less stringent AOX limits than the Nordic countries. The monthly average of AOX emissions from existing sources cannot exceed 0.62 kg/ton pulp. For new sources, the monthly average cannot exceed 0.27 kg/ton pulp. Daily discharges cannot exceed 0.95 kg/ton pulp and 0.48 kg/ton pulp, respectively. In addition, for bleached sulfite mills, both existing and new sources require TCF technology (Webb, 1998). Mills had until 2001 to comply. To encourage TCF usage, mills that voluntarily chose to install TCF technology are given an additional three years to comply (OECD, 1999b). Similarly, regulation in Canada addressed dioxins in the 1990s. Federal and provincial governments share responsibility for water pollution control. The federal role in water pollution control is outlined in the Fisheries Act of 1970. Pulp and Paper Effluent Regulations were first introduced in 1971. In May 1992, the federal government introduced new regulations for the pulp and paper industry, which included new standards for dioxins, but did not include specific limits for AOX (OECD, 1999b). However, several Provinces have established limits for AOX emissions. The first to do so was British Columbia, which in 1990 set a limit of 1.5 AOX kg/ADt, to be met by 1995. The 1990 legislation called for eliminating AOX emissions by 2002 (OECD, 1999b). However, this was later repealed after review by a panel of scientific experts. New standards now limit the monthly average AOX releases to 0.6 kg/ADt.12 Quebec established AOX limits in 1992; followed by Ontario in 1993. In both cases, the standards were phased in gradually, with monthly AOX averages needing to fall to 0.8 kg/ADt by 2000. In addition, new mills in Quebec face a more stringent standard of 0.25 kg/ADt (OECD, 1999b). During the period when dioxin and furan issues unfolded two new bleached kraft mills were built in Alberta, as well as the modernization and expansion of two existing bleached kraft mills. The Provincial Government conducted extensive assessments of technology and environmental requirements, setting case by case limits in each mill’s permit. In 2002 its mills had daily AOX limits in the range of 1 to 3.0 kg/t and monthly limits in the range of 0.55 to 1.5 kg/t.13 In Japan, regulation was slower to develop than voluntary industry measures. In 1991, the Japanese pulp and paper industry proposed that AOX levels be limited to 1.5 kg/metric ton by the end of 1993, and recommended the use of oxygen delignification equipment and chlorine dioxide substitution to meet this goal (Management Institute for Environment and Business, 1994). The first law pertaining to dioxin took effect in 2000, limiting dioxins in wastewater to 1 pg/l.14 This was a general regulation applying to all industries. No specific limits apply to the pulp and paper industry.15

120

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Table 3.4 summarises the key regulations in each country. Finland first regulated in 1988, and Sweden in 1991. While Canadian provinces also passed regulations in the early 1990s, these did not take effect immediately. The standards (0.6-0.8 kg/ADt) were less stringent than the permitting guidelines used in Sweden and Finland (0.1-0.4 kg/t). While the United States attempted to establish TCF as best available technology in 1993, the final Cluster Rule published in 1997 did not include this standard. The monthly average AOX releases from existing sources must be below 0.62 kg/t for existing sources – a standard comparable to British Columbia. And finally, Japan relied voluntary compliance until 2000. Table 3.4. Summary of key regulations Sweden 1991: Environmental legislation establishes strict guidelines for AOX (0.1-0.2 kg/t). Enforcement is through plant-by-plant permitting. Finland 1987: Issues first guidelines for AOX (1.4 kg/ADT), to be met by 1994. Enforcement is through plant-by-plant permitting. 1993: Accepts Nordic Working Group performance standards for AOX (0.2-0.4 kg/t). Enforcement is through plant-by-plant permitting. Canada 1990: British Columbia sets AOX limits of 1.5 kg/ADt, to be met by 1995. Since lowered to 0.6 kg/ADt. 1992: Quebec passes AOX limits that are phased in gradually. AOX limit of 0.8 kg/ADt by 2000. New mills limited to 0.25 kg/ADt. 1993: Ontario passes AOX limits that are phased in gradually. AOX limit of 0.8 kg/ADt by 2000. United States 1993: Proposed Cluster Rule suggests TCF as best available technology. Never took effect. 1997: Revised Cluster Rule limits monthly average AOX releases to 0.62 kg/t pulp for existing sources, and 0.27 kg/t pulp for new sources. Mills have until 2001 to comply. Japan 1991: Pulp and paper industry proposes voluntary AOX limit of 1.5 kg/metric ton by end of 1993. 2000: First law limiting dioxins in wastewater (1 pg/l). No specific limit for AOX or for the pulp and paper industry.

5. Data Using IPC classes, it is possible to identify patents specifically pertaining to ECF and TCF technology, as well as the source of each patent. One advantage of looking at ECF and TCF technologies is that there are specific patent classes pertaining to these technologies. (These are listed in Annex 3.A1.) Within these classes, the first, D21C 9/14, pertains to ECF production, as it covers the use of chlorine dioxide. Chlorine dioxide was used in the later stages of bleaching (even before the switch to ECF technologies), so there will be patents in this even class before regulations were in place. However, an increase in innovation would be expected once the shift from elemental chlorine to ECF began. The second class,

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

121

3.

POLICY VERSUS CONSUMER PRESSURE...

D21C 9/147, includes bleaching processes using oxygen. These can be either ECF or TCF. The last two classes, D21C 9/153 and D21C 9/16 cover bleaching using ozone or per compounds (e.g. hydrogen peroxide). Although these also can be used in both ECF and TCF bleaching, they are particularly important for TCF, as these chemicals substitute for chlorine dioxide in the TCF process. Data was collected for each of the five countries, using the Delphion patent database.16 Within each country, patent counts were generated for the home country of the inventor. 17 The focus was primarily on patents by domestic inventors, which shows how inventors are reacting to incentives in that country. Patents were sorted by the priority year. In the case of Sweden and Finland, inventors might also choose to file patents through the European Patent Office (EPO), rather than through the national patent offices. EPO patent applicants may designate as many of the 32 EPO member-states for protection as desired. The application is examined by the EPO. If granted, the patent is transferred to the individual national patent offices designated for protection. Because EPO applications are more expensive, European inventors typically first file a patent application in their home country, and then apply to the EPO if they desire protection in multiple European countries. Thus, most Swedish and Finnish inventors will first file an application in their home country. However, inventors from other European countries are likely to use an EPO patent if they desire protection in Sweden and Finland. Thus, data on European Patents that designate Sweden and/or Finland for protection are also included.

6. Analysis 6.1. Patent trends Domestic patent applications in each country are first reviewed. This allows for a comparison of incentives for invention in each of the countries in the study. Figure 3.1 shows patents granted in each country to domestic inventors, sorted by the first priority year. What is most notable from this Figure is the role that perceived public pressure (in response to initial reports of dioxin in waterways) appears to have played in driving innovation with respect to ECF and TCF technologies. With the exception of Canada, every country experienced an increase in ECF and TCF patents that began after release of the Greenpeace report in 1987. While there was some regulation at this time, the initial regulations were not very strict. Sweden, the first country to pass stringent AOX guidelines, did so in 1992. While the US did announce plans for strict regulations that would declare TCF to be best available technology in 1993, the lack of innovative response from US inventors after this announcement suggests that this initial proposal was not perceived as credible (this proposal was eventually withdrawn, and replaced by the weaker Cluster Rules in 1998).18

122

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Figure 3.1. Domestic ECF and TCF patents by country SE Number 35

United States

Finland

SE tightens guidelines

30

Greenpeace report

Canada

US Cluster Rule

Japan

New JP dioxin limits

25 F1 tightents standards

20 15 10 5 0 1975

1977

1979 1981 1983 1985 1987

1989 1991

1993

1995 1997 1999 2001 2003 Priority year

The Figure shows patent applications of domestic inventors for each country in our sample. In all cases except the US, both successful and unsuccessful applications are included. Only patents that were subsequently granted are included in US applications.

The finding that patenting increased before regulations were put in place, rather than in response to regulation, and that these increases occurred even in countries that did not pass early regulation, suggests that increased public scrutiny played an important role in influencing this first wave of innovation.19 The American Paper Institute noted as early as November 1988 that several mills where dioxin was detected downstream had begun process modifications, even though no regulations were in place at the time (Chemical Week, 1988). Similarly, the discharge limits adopted by the Nordic States became redundant because green market demand had surpassed those limits for more stringent measures (Smith and Rajotte, 2001, p. 146). In addition, industry experts expressed concern over future regulation in response to increased scrutiny. An April 1988 article in Chemical Week cited a prediction from an industry source, that the bleaching process would soon be regulated, and that this would have a “heavy impact on purchasing over the next 5-10 years” (Agoos and Portnoy, 1988, p. 45). Another unnamed paper producer noted at the time that “there is likely to be a substantial change in the way pulp is bleached” (Agoos and Portnoy, op. cit.). A second implication of these trends is that stringent regulation seems to have spurred additional innovation. Here, it is worth comparing innovation in the Nordic countries to innovation in the US and Canada. Both Swedish and Finnish ECF and TCF patents peaked within two years of those countries passing more stringent guideline limits.20 By comparison, US and Canadian mandatory standards were less stringent than the Nordic guidelines, and

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

123

3.

POLICY VERSUS CONSUMER PRESSURE...

typically made use of existing technologies. There is little evidence of innovation occurring in response to US and Canadian regulations. Canadian patents experience no notable increases; US patenting activity increased dramatically in response to the initial news about dioxin, but not in response to new regulations. Neither the proposed (and ultimately defeated) standards of 1993, nor the final Cluster Rule of l998 led to additional patenting in the US. Further evidence of the importance of the regulatory framework can be seen by looking at the types of innovation that were taking place. Table 3.5 provides counts of patents using chlorine dioxide (ClO2) and chlorine substitutes for selected years. Patents using ClO2 could only be used for ECF processes, whereas those using chlorine substitutes could be used in either ECF or TCF processes. While it cannot be ascertained when research has shifted to TCF processes, it is possible to identify some (but not all) of those cases where ECF was clearly the main research goal, as reflected in the row labelled ClO2 in Table 3.5. While the “other” category can in theory reflect either ECF or TCF innovations, in some countries it is likely to be almost exclusively ECF technologies.21 Table 3.5. Number of domestic chlorine and non-chlorine patents, selected years Priority Year 1975 1980 1985 1988 1989 1990 1991 1992 1993 1998 2000 2002 Canada Finland Japan Sweden United States

ClO2

2

0

1

0

0

1

0

1

0

1

0

1

Other

1

0

2

1

2

2

2

5

2

3

2

2

ClO2

0

0

0

1

2

2

0

0

0

1

3

0

Other

0

0

2

0

2

1

3

1

7

3

0

1

ClO2

3

1

0

4

1

6

2

1

4

3

5

10

Other

2

7

0

8

15

7

14

22

25

11

14

21

ClO2

1

2

0

1

3

6

0

0

1

0

2

2

Other

2

4

1

4

1

2

5

11

22

5

4

0

ClO2

1

1

1

1

7

2

2

5

0

5

2

0

Other

3

0

3

4

11

15

27

13

21

0

2

0

The Table shows the number of domestic patent applications for selected years for technologies using chlorine dioxide (ClO2) and using substitutes for chlorine (other). ClO2 patents correspond to IPC class D21C 9/14, and other corresponds to the other three IPC classes. Note that ClO2 patents could only be used in ECF processes, whereas the other patents could be used in both ECF and TCF processes.

US patents focused on chlorine substitutes. Sweden had fewer patents at this time, and many used chlorine dioxide. However, after Sweden established its strict AOX standards in 1992, nearly all Swedish patents were for technologies which used chlorine substitutes. This was also true after Finland adopted stringent standards in 1993. In contrast, not only did US patenting activity fall after the adoption of the revised Cluster Rule in 1997, but the nature of innovation shifts. Once the revised Cluster Rule established ECF as acceptable, most

124

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

US patents made use of chlorine dioxide, because there was little incentive for plants to consider alternatives such as TCF bleaching. Similarly, a review of who received patents in each country illustrates differences in the types of innovations occurring across countries. Table 3.6 shows the domestic assignees receiving the most patents in each country, along with the type of firm. Innovations came from pulp producers, pulping equipment manufacturers, and from chemical companies. Chemical companies were more prominent in the Nordic countries than in North America. While the move to TCF appears to have required innovation from chemical suppliers, chlorine dioxide had been used in the pulping process even before the move to ECF. The switch to ECF simply required using more ClO2 than before. Also of note is that many of these patents come directly from the pulp and paper industry. The range of patent assignees found here contrasts with Popp (2006), which examined innovations for NOX and SO 2 control. There, nearly all patents came from equipment producers, rather than from the regulated firms themselves. Plant-by-plant permitting might be expected to induce more innovation when the regulated firms are the innovators (as they often are here), as these firms know what they can do when they agree to a permit. In contrast, outside suppliers should be more comfortable responding nationwide regulations, since such regulations provide a demand for new (cleaner) technologies. Unfortunately, it was not possible to pursue this hypothesis further, since the regulations vary not only by type, but also by stringency level. Examination of patent family data provided further evidence of the leading innovative role of Sweden and Finland. Patent protection is only valid in the issuing country. To receive protection in multiple countries, an applicant must obtain a patent in each country for which protection is desired. Additional fees apply for each application. However, inventors are given a one-year window after their first filing to decide to patent elsewhere. Only the most valuable inventions are therefore filed in several countries. Moreover, filing a patent application in a given country is a signal that the inventor expects the invention to be profitable in that country. Because of this, researchers such as Lanjouw and Schankerman (2004) have used data on patent families as proxies for the quality of individual patents. Figure 3.2 shows the average family size of domestic patents in each country by year. Sweden, Finland, and the US have consistently produced the largest family sizes, whereas most Japanese patents were filed only in Japan.22 Moreover, family sizes were largest after the initial outcry over dioxin in the late 1980s. This provides further evidence that global concern, rather than domestic regulation, drove innovation. Also notable is that the family size of

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

125

3.

POLICY VERSUS CONSUMER PRESSURE...

Table 3.6. Top domestic patent assignees Number

% total

Org. type

7 6.5 5 4 3

12.5% 11.6% 8.9% 7.1% 5.4%

Pulp/Paper Pulp/Paper N/A Chemicals N/A

27 23 6 3 3

35.1% 29.9% 7.8% 3.9% 3.9%

Chemicals Pulp/Paper Pulp/Paper Pulp/Paper Independent research org.

51.5 47 35 27 10.5 10 10

15.7% 14.3% 10.7% 8.2% 3.2% 3.0% 3.0%

Pulp/Paper Pulp/Paper Chemicals Pulp/Paper Chemicals Chemicals Pulp/Paper

32 26 14 14 7 6 5 5 4 4

22.4% 18.2% 9.8% 9.8% 4.9% 4.2% 3.5% 3.5% 2.8% 2.8%

Pulp and chemicals Pulping equipment Chemicals Pulp/Paper Pulp/Paper Pulping equipment Pulp/Paper Pulping equipment Chemicals Pulp/Paper

31 20 13 12 11.5 9 8 7 5 4.5 4 4

15.9% 10.3% 6.7% 6.2% 5.9% 4.6% 4.1% 3.6% 2.6% 2.3% 2.1% 2.1%

Pulp/Paper Pulp/Paper University Pulping equipment Pulping equipment N/A Chemicals Pulp/Paper Pulp/Paper Chemicals Chemicals Chemicals

Canada Pulp and Paper Research Institute of Canada MacMillan Bloedel Ltd. Individual Erco Industries Ltd. Unassigned Finland Kemira OY Ahlstrom Machinery OY Enos-Gutzeit OY Rauma-Repola OY Valtion Teknillinen Tutkimuskeskus Japan Mitsubishi Paper Mills Ltd. Oji Paper Co. Ltd. Mitsubishi Gas Chem Co. Inc. Nippon Paer Industries Co. Ltd. Sanyo Chem Ind. Ltd. Sumitomo Heavy Ind Ltd. New Oji Paper Co. Ltd. Sweden Kvaerner Pulping AB Sunds Defibrator Industries AB Eka Nobel AB Mo Och Domsjoe AB SCA Development AB Kamyr AB Metos Paper Inc. Valmet Fibretech AB AGA AB Stora Kopparbergs Bergslags AB United States Union Camp Patent Holding, Inc. International Paper Company Wayne State University Kamyr, Inc. Beloit Technologies, Inc. Individual Air Products and Chemicals, Inc. Weyerhaeuser Company Champion International Corporation E. I. Du Pont de Nemours and Company The BOC Group, Inc. The Dow Chemical Company

The Table shows the number of domestic patents granted to frequent assignees, by country. For firms, Org. Type describes the primary industry of the firm. Fractional counts are used for patents with multiple assignees. % total is the percentage of all domestic ECF and TCF patents assigned to the firm, including firms not listed above.

126

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Figure 3.2. Average patent family size by country and year Sweden

United States

Finland

Canada

Japan

Average family size 16 14 12 10 8 6 4 2 0 1985

1987

1989

1991

1993

1995

1997

1999

2001 2003 Priority year

The Figure shows the average patent family size of patent applications of domestic inventors for each country in our sample. In all cases except the US, both successful and unsuccessful applications are included. Only patents subsequently granted are included in US applications.

Swedish and Finnish patents remained high throughout the 1990s. In contrast, the average family size of US patents falls dramatically after 1996. A likely explanation for this is the technology-following nature of US regulations. Because the Nordic regulations were more stringent than those in the US, larger inventive steps would have been needed to comply with Nordic regulations. US regulations, in contrast, could be met with existing technology. As such, not only did the level of innovation fall once it was clear that TCF technology would not be required in the US, but so did the quality of innovation. Since the mid-1990s, most major patents for ECF and TCF technologies came from the Nordic countries. To better illustrate the flows of knowledge across countries, Figure 3.3 shows both domestic and foreign patents in Finland, Sweden, and the US. Both domestic and foreign regulations appear to have influenced innovation. For example, patents from US inventors peaked in 1990 in the US. However, they peak in 1992 in Sweden and 1993 in Finland – after passage of tighter regulations in those countries.23 Similarly, an increase in Swedish patents can be observed (both in Sweden and the US) after the 1997 Cluster Rule in the US.24 This contrasts with the results of Popp (2006), which found that domestic regulations were the primary drivers of innovation for air pollution control devices for coal-fired power plants. One difference here is that the pulp and paper industry is a global market, whereas most suppliers of pollution abatement equipment in Popp (2006) were domestic companies.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

127

3.

POLICY VERSUS CONSUMER PRESSURE...

Figure 3.3. ECF and TCF patent trends, selected countries SE

FI

United States

Patent applications 16 A. Finland 14

Other European

Other

F1 tightens standards

12 10 8 6 4 2 0 1975

1977

1979

1981

1983

SE

1985

1987

United States

Patent applications 25 B. Sweden

1989

1991

1993

1995

Other European

1997

1999 2001 2003 Year

Other

SE tightens guidelines

20 15 10 5 0 1975

1977

1979

1981

1983

United States

1985

1987

SE

Patent applications 35 C. United States 30

1989

1991

1993

1995

1997

Other European

CA

Failed Cluster Rule

1999 2001 2003 Year Other

Revised Cluster Rule

Greenpeace report

25 20 15 10 5 0 1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999 2001 2003 Year

The Figures show all ECF and TCF patents in Finland, Sweden, and the United States, grouped by inventor country.

128

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

6.2. Adoption of ECF and TCF technologies Figure 3.3 suggests that innovation on ECF and TCF technologies came early, in some cases preceding regulations. However, looking at innovation does not present a complete picture, as it does not consider whether newly developed technologies are actually put to use. Figures 3.4 and 3.5 show the percentage of pulp production using ECF technology and ECF or TCF technology respectively.25 It is here where the influence of regulation becomes clearer. By 1994, all pulp production in the Nordic countries used either ECF or TCF technology. In contrast, North American usage increased more slowly. It was not until the Cluster Rule deadline of 2001 that nearly 100% adoption was achieved. One important difference in North America is that public pressure for chlorine-free paper did not persist as long as it did in Europe. Moreover, the US industry served primarily a domestic market, exporting just 10% of its paper products (Norberg-Bohm and Rossi, 1998). While public pressure was sufficient to jumpstart innovation on ECF and TCF technologies, as well as to encourage some reductions of chlorine use, the adoption data make clear that complete diffusion will not occur unless binding regulations are also in place. Figure 3.4. Diffusion of ECF bleaching technologies World

Nordic countries

North America

Row

% of pulp production 100 90 80 70 60 50 40 30 20 10 0 1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001 2002 Year

The Figure shows the percentage of chemical pulp production using ECF technology in The Nordic countries, North America, and the rest of the world (ROW). Diffusion of ECF has been rapid in the Nordic countries, due to both strong consumer demand and early regulation. In contrast, diffusion in North America has been more gradual until 2001, the deadline for compliance with the US Cluster Rules. Source: Authors’ calculation, using data from Alliance for Environmental Technology (2006).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

129

3.

POLICY VERSUS CONSUMER PRESSURE...

Figure 3.5. Diffusion of ECF and TCF bleaching technologies World

Nordic countries

North America

Row

% of pulp production 100 90 80 70 60 50 40 30 20 10 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year The Figure shows the percentage of chemical pulp production using either ECF or TCF technology in The Nordic countries, North America, and the rest of the world (ROW). Use of elemental chlorine has been completely eliminated in The Nordic countries, and nearly eliminated in North America since the US Cluster Rules took effect. Source: Authors’ calculation using data from Alliance for Environmental Technology (2006).

6.3. Public Policy vs. Consumer Pressure The combination of early innovation and delayed adoption in North America provides some interesting lessons for the induced innovation literature. First, the early influence of public pressure, particularly after Greenpeace published EPA reports on dioxin, is striking. Typically, induced innovation studies focus on the effect of regulation on innovation. Here, innovation comes first. Regulation followed, made possible both by public pressure for action and the availability of alternative technologies for pulp production. The role of leading countries was also important. Sweden and Finland moved quickly to reduce chlorine usage. As discussed earlier, these decisions were made partly because the technologies that had been developed in response to news on dioxin discharges were deemed acceptable. However, the data suggest that additional research was needed in order to perfect these technologies, as both Sweden and Finland experienced an increase in ECF and TCF patenting after announcing stringent national guidelines in the early 1990s. In contrast, the US and Canada delayed regulation, and appeared to have developed their regulations in light of the availability of existing technologies. Early attempts to establish TCF as the “best available technology” in the US did not come into force. When the US finally adopted the Cluster Rule in 1998, ECF had been clearly established as a viable technology. As such, while the rule served to increase adoption of ECF technology, no further innovation was needed.

130

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

This differs from the conclusion reached by Popp (2006), which found that even late adopters of coal-fired power plant regulations needed to innovate to adapt technologies to local conditions. One difference was that domestic innovation did occur in the US prior to the Cluster Rule, whereas in the case of power plants, little domestic innovation occurred before regulations were enacted. In the case of power plants, even if there had been public pressure to reduce emissions, consumers’ only option before regulation would have been to reduce electricity usage. Alternative, clean suppliers of electricity were not available, as regulations for nitrogen dioxide and sulfur dioxide took effect before the movement towards deregulated electricity markets in the late 1990s. In contrast, some pulp manufacturers did face incentives to reduce chlorine usage before regulations were in place, because this enabled these manufacturers to differentiate their product and target environmentally-conscious consumers. Finally, the importance of early publicity suggests a possible role for labelling to encourage both innovation and diffusion of ECF and TCF technologies. However, the first labelling requirement restricting chlorine was the Nordic Swan in 1991. Most schemes did not address chlorine usage until later in the 1990s. While they may have played a role in the diffusion of ECF and TCF technology, labelling schemes appear to incorporate existing technologies in their criteria, rather than serve as technology-forcing standards.26 Given that these labels are voluntary measures, this is not surprising, as labelling programs do not offer consumer choices unless some products qualify for the label.

7. Conclusions This Chapter drew upon patent data to study the development of ECF and TCF bleaching technologies in the pulp and paper industry across five OECD countries. While most studies using patent data focus on end-of-the-pipe solutions to environmental problems, this Chapter examined patenting for a process technology. As in other studies of environmental innovation, regulation does seem to play a predominant role in both the development and diffusion of these technologies. However, it is not the only driver of innovation, and perhaps not even the most important. While the data used in this Chapter is insufficient to draw definitive conclusions, one striking finding from the data presented is the apparent role of public pressure. While ECF and TCF technologies are process technologies, they affect the quality of the final product. In the late 1980s, studies linking chlorine bleaching technologies to dioxin led to pressure from environmentalists for reduced chlorine bleaching in paper production. Detection of trace amounts of dioxin in products such as diapers and coffee filters led to increased awareness of the issue, particularly in Europe. In response, the development of alternative

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

131

3.

POLICY VERSUS CONSUMER PRESSURE...

bleaching technologies increased rapidly in each of the countries examined. Moreover, this increase occurred before new environmental regulations could be put in place limiting chlorine use, suggesting that public pressure, rather than regulation, was the primary driver of this first wave of innovation. Nonetheless, public policy did play an important role. In response to increased awareness of the problems of chlorine bleaching, both Sweden and Finland enacted relatively strict regulations in the early 1990s, typically through plant-by-plant permitting. In both countries, these regulations were followed by both increased innovation and increased adoption of ECF and TCF technologies. Moreover, these tighter regulations shaped the nature of innovation, as firms focused on technologies relevant for TCF production. In contrast, the US, Canada, and Japan all enacted weaker regulations that could be satisfied using ECF technology. Moreover, in the US and Japan, these regulations did not come until later in the decade. As such, patenting in the US did not remain at high levels (although it did in Japan, presumably in response to foreign regulations). Moreover, the majority of US patents at the end of the decade focused on ECF, rather than TCF, technology. Public policy also plays a role in diffusion. While the patent data show that the US was an early innovator of ECF and TCF technologies, adoption of these technologies was slower in the US than in Sweden or Finland. While some US plants seem to have adopted ECF technology in response to consumer pressure, it was not until regulations which required its use took effect in 2001 that near universal adoption of ECF or TCF technologies occurred. In contrast, Sweden and Finland achieved 100 percent diffusion of ECF and TCF technologies by 1994, due to earlier regulation requiring these technologies. Finally, while this Chapter illustrates the effect of different policy regimes on innovation, it says nothing about efficiency. TCF technology was more costly and produced lower quality paper than ECF. While the more stringent regulations in Sweden and Finland did hasten the development and diffusion of TCF technology, it is beyond the scope of this Chapter to assess whether the additional benefits from completely removing chlorine from the bleaching process, compared to the partial reduction achieved by ECF, were worth the additional costs of developing and using TCF technology.

Notes 1. Examples include Popp (2003, 2006), Taylor et al. (2003), and Lanjouw and Mody (1996). 2. The theoretical literature includes Magat (1978), Milliman and Prince (1989), and Fisher et al. (2003). These papers predict that market-based environmental policies, such as a tax or permit trading, will induce more innovation than a comparable command and control policy.

132

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

3. Pulp can be produced using chemical or mechanical methods, as described in the following section. Most production, particularly in developed countries, uses chemical methods. 4. Despite being a NORSCAN country, Norway appears on neither list, and has few ECF or TCF patents. Thus, Norway is not included in the analysis. 5. These five stages may be preceded by oxygen delignification, denoted by the letter O, so that the process is OCEDED. 6. Chelation is the addition of compounds to control the formation of free radicals to retard decomposition of the hydrogen peroxide (Reinstaller, 2005). 7. Personal communication, Dave Halliburton, Environment Canada. 8. Personal Communication, Mimi Nameki, First Secretary, Permanent Delegation of Japan to the OECD. 9. In addition, European Union integrated pollution and prevention control (IPPC) regulations cover European pulp mills. The European Union reached agreement on IPPC in 1996, and is based on similar legislation passed earlier in the United Kingdom (Webb 1999). Directives for pulp and paper production took effect in 2001. For bleached kraft pulp, the new AOX standard is < 0.25 kg/adt. Existing standards in both Finland and Sweden already satisfy this requirement. 10. ADt represents an air dry ton of pulp product. 11. These are called the “Cluster Rules”, because the standards address multiple pollutants, including both air and water, simultaneously. 12. Personal communication, Dave Halliburton, Environment Canada. 13. Personal communication, David Halliburton, Environment Canada. 14. There is no specific standard for AOX in Japan. 15. Personal communication, Mimi Nameki, First Secretary, Permanent Delegation of Japan to the OECD. 16. This database was preferred over the OECD Triadic Patent Family database since there would be little variation in the “counts” for the latter due to the restricted nature of the technologies examined. However, as with TPF data, the Delphion data does combine related applications and grants into families to ensure commensurability. 17. In the case of multiple inventors from different countries, the home country of the inventor is listed first. 18. In contrast, in the late 1980s, patenting activity for sulfur dioxide control technologies increased in response to early attempts at modifying the US Clean Air Act, suggesting that innovators expected that these attempts would eventually lead to regulatory changes (Taylor et al., 2003). 19. At the same time, the implementation of the Toxic Release Inventory (TRI) in the US in 1990 led firms to reduce releases of chemicals such as chlorine. However, the finding that innovation is global, rather than just in the US, suggests that the importance of consumer pressure was greater than the pressure from TRI. Of particular importance, TRI data only reports releases from the production process. It says nothing about the levels of chlorine that remain in paper. Consumer pressure, particularly in Europe, focused on the chlorine content of paper, rather than chlorine releases. 20. One interesting possibility is that case-by-case permitting may hasten the development of new technologies. Bjällås (1999) notes that Sweden’s approach to

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

133

3.

POLICY VERSUS CONSUMER PRESSURE...

permitting encourages innovation by allowing individual plants to propose different solutions. Moreover, the Swedish National Licensing Board can postpone decisions in order to investigate new technologies. Conversely, mandatory standards such as those applied in Canada and the United States may allow for greater certainty with respect to the potential size of the market for any innovations. Unfortunately sufficient patent data does not exist to allow testing of whether either of these factors were important. 21. See, for instance, the review of softwood bleaching practices in Canada by Earl and Pryke (2003). 22. Japan’s small family size is explained by Japan being a technology-follower in this field. Most pulping equipment used in Japan comes from foreign sources (Management Institute for Environment and Business, 1994). 23. Patents are sorted by their priority date, which is the date of the first patent filing anywhere in the world. As such, these trends are not simply results of delays in filing patent applications abroad. 24. Since the peaks are different for each destination country, at least some of the variation comes from inventors choosing to file more of their patent applications in other countries. However, the data does allow for conclusions concerning the extent to which inventors were responding to regulation in other countries by increasing innovative activity (e.g. creating more inventions) or simply by choosing to file more of their patent applications in other countries. 25. The chart combines Canada and the US as part of North America, because separate data are available for these countries only through 2001. Using separate data reveals similar trends. Canada had slightly higher adoption rates than the US, due to earlier regulation at the provincial level. The only deviation between the two countries was that increased North American diffusion in 2001 was entirely due to the US, which saw usage of ECF technology increase from 76% to 96% as the deadline for compliance with the Cluster Rule regulations passed (Alliance for Environmental Technology, 2002). Less than 1% of plants in the US and Canada use TCF, since it is possible to comply with regulation using ECF technology. 26. Even the influence of labelling schemes to encourage adoption is hard to discern from the data. Clearly, adoption of ECF and TCF technology increases in The Nordic countries in the early 1990s, but this is not solely due to labelling, as regulations in Sweden and Finland were also tightened at this time. However, the Nordic Swan standards are more stringent than the AOX requirements in these countries, and may help reduce emissions beyond what is required (personal communication, Grethe Torrissen, Advisor, Sustainable Production and Consumption/IPP, the Norwegian Ministry of the Environment). In contrast, the US Green Seal label allowed chlorine, as long as releases were below 1 kg/ton, until 1996. This may be one explanation for adoption of ECF technology prior to the Cluster Rule. However, although Green Seal banned chlorine bleaching after 1996, there was no shift to TCF technology in the US, with less than 1% of production using this technology (Alliance for Environmental Technology, 2002).

References Agnoos, Alice and Kristine PORTNOY (1988), “Chemicals Alter Their Roles in Papermaking”, Chemical Week, p. 45. Alliance for Environmental Technology (2002), Trends in World Bleached Chemical Pulp Production: 1990-2001.

134

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Alliance for Environmental Technology (2006), Trends in World Bleached Chemical Pulp Production: 1990-2005. Anonymous (1987), “Dioxins in Diapers Stir Sweden”, Chemical Week, December 23, 1987, p. 28. Ratthew Auer, M. (1996), “Negotiating Toxic Risks: a Case from the Nordic Countries”, Environmental Politics, Vol. 5, No. 4, pp. 687-699. Bjällås, Ulf (1999), “The Swedish Integrated Permitting System”, Environmental Requirements for Industrial Permitting, Vol. 2, OECD, Paris. Burton John, (1989), “A Global Leader – Sweden Sets the Pace in Tackling Environmental Issues”, Financial Times, December 13, 1989, p. IV. Chemical Week (1988), “Paper Industry Fights Dioxin”, p. 84. Collins, Lyndhurst (1992), “Environment versus Industry: a Case Study of how the Pulp and Paper Industry is Responding to Changing Attitudes to the Environment”, Business Strategy and the Environment, Vol. 1, No. 4, pp. 29-36. Earl, Paul F. and Douglas C. Pryke (2005), “Softwood Bleaching Practices in Canada: Analysis of the 2003 PAPTAC Bleaching Committee ‘Best Practices’ Survey”, paper presented at the 2005 Engineering, Pulping and Environmental Conference, 2005, / www.tappi.org/s_tappi/doc_bookstore.asp?CID=5021&DID=526517EPA Office of Compliance (2002), Profile of The Pulp And Paper Industry, 2nd Edition, Washington, DC. Faostat Data (2006), www.fao.org/faostat, Accessed May 2, 2006. Fisher, Carolyn, Ian W.H. Parry, and William A. PISER (2003), “Instrument Choice for Environmental Protection when Environmental Protection is Endogenous”, Journal of Environmental Economics and Management, Vol. 45, pp. 523-345. Gray, Malcolm, William LOWTHER and David TODD (1987), “Alarm over paper goods”, Maclean’s, October 26, 1987, p. 57. Griliches, ZVI (1990), “Patent Statistics As Economic Indicators: A Survey”, Journal of Economic Literature, Vol. 28, No. 4, pp. 1661-1707. Harrision, Kathryn (2002), “Ideas and Environmental Standard-Setting: A Comparative Study of Regulation of the Pulp and Paper Industry” in Governance: An International Journal of Policy, Administration and Institutions, Vol. 15, No. 1, pp. 65-96. Lanjouw, Jean O. and Ashoka MODY (1996), “Innovation and the International Diffusion of Environmentally Responsive Technology”, Research Policy, Vol. 25, pp. 549-571. Lanjouw, Jean O. and Mark Schankerman (2004), “Patent Quality and Research Productivity: Measuring Innovation with Multiple Indicators”, Economic Journal, Vol. 114, No. 495, pp. 441-465. Larsson, A., T. Andersson, L. Förlin and J. Härdig (1988), “Physiological Disturbances in Fish Exposed to Bleached Kraft Mill Effluents”, Water Science and Technology, Vol. 20, No. 2, pp. 67-76. Levin, Richard C., Alvin K. Klevorick, Richard R. Nelson, and Sidney Winter (1987), “Appropriating the Returns from Industrial Research and Development”, Brookings Papers on Economic Activity, Vol. 3, pp. 783-820. Lockie, Mark (1998), “Pulp Producers Hunt for Lower Cost Bleaching”, Pulp and Paper International, Vol. 40, No. 10, pp. 44-45. Lyon, Thomas P. and John W. Maxwell (2002), “Voluntary Approaches to Environmental Regulation: A Survey,” in Maurizio Franzini and Antonio Nicita (eds.), Economic

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

135

3.

POLICY VERSUS CONSUMER PRESSURE...

Institutions and Environmental Policy: Post Present and Future, Aldershot, Hampshire, UK, Ashgate Publishing Ltd. Magat, A. Wesley (1978), “Pollution Control and Technological Advance: a Dynamic Model of the Firm”, Journal of Environmental Economics and Management, Vol. 5, pp. 1-25. Management Institute for Environment and Business (1994), “Competitive Implications of Environmental Regulation: A Study of Six Industries”, US EPA, Washington, DC, Ava i l abl e o n l i n e a t h t t p : / / y o s e m i t e. e p a . g o v / e e / e p a / e e r m . n s f /v w S E R / C6E46F0830E7B87 78525644D0053BE5A?OpenDocument. Milliman, Scott R. and Raymond Prince (1989), “Firm Incentives to Promote Technological Change in Pollution Control”, Journal of Environmental Economics and Management, Vol. 17, pp. 247-165. Norberg-Bohm, Vicki and Mark Rossi (1998), “The Power of Incrementalism: Environmental Regulation and Technological Change in Pulp and Paper Bleaching in the US,”Technology Analysis and Strategic Management, Vol. 10, No. 2, pp. 225-245. OECD (2005), “Environmental Exposure Assessment: Final Draft for Emission Scenario Document on Kraft Pulp Mills”, document ENV/JM/EEA(2005)4. OECD (1999a), “Environmental Requirements for Industrial Permitting: Case Study on the Pulp and Paper Sector, Part One”, document ENV/EPOC/PPC(99)8/FINAL/PART1. OECD (1999b), “Environmental Requirements for Industrial Permitting: Case Study on the Pulp and Paper Sector, Part Two”, document ENV/EPOC/PPC(99)8/FINAL/PART2. Popp, David (2006), “International Innovation and Diffusion of Air Pollution Control Technologies: The Effects of NOX and SO2 Regulation in the US, Japan, and Germany”, Journal of Environmental Economics and Management, Vol. 51, No. 1, January 2006, pp. 46-71. Popp, David (2003), “Pollution Control Innovations and the Clean Air Act of 1990”, Journal of Policy Analysis and Management, Vol. 22, pp. 641-660. Reinstaller, Andreas (2005), “Policy Entrepreneurship in The Co-Evolution of Institutions, Preferences, and Technology. Comparing the Diffusion of Totally Chlorine Free Pulp Bleaching Technologies in the US And Sweden”, Research Policy, Vol. 34, pp. 1366-1384. Salzman, Jim (1991), Environmental labelling in OECD countries, Paris, France, OECD. Sammarco, L. (1997), “Eco-Labelling Paper and Board Products: The Story so far and What the Future Holds”, Paper and Packaging Analyst, Vol. 29, pp. 37-51. Smith, Adrian and Alain Rajotte (2001), “When Markets Meet Sociopolitics: The Introduction of Chlorine-Free Bleaching in the Swedish Pulp and Paper Industry”, in Rod Coombs, Ken Green, Albert Richards and Vivien Walsh (Eds.), Technology and the market. Demand, users and innovation, Cheltenham, UK, Edward Elgar. Taylor, Margaret, R., Edward S. Rubin, David H. Hounsell (2003), “Effect of Government Actions on Technological Innovation for SO2 Control”, Environmental Science and Technology, Vol. 37, pp. 4527-4534. Thulin, Jan, J. and E. Lindesjöö (1988), “Diseases and Parasites of Fish in a Bleached Kraft Mill Effluent”, Water Science and Technology, Vol. 20, No. 2, pp. 179-180. Webb, Leslie (1999), “IPPC Steps up to BAT”, Pulp and Paper International, Vol. 41, No. 4, pp. 29-32. Webb, Leslie (1998), “Wastewater Treatment: Regulations, Bugs and Beds”, Pulp and Paper International, Vol. 40, No. 6, pp. 39-43.

136

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

3.

POLICY VERSUS CONSUMER PRESSURE...

Webb, Leslie (1996), “Making the Labels Stick on a Complex Issue”, Pulp and Paper International, Vol. 38, No. 12, pp. 31-34. Webb, Leslie (1994), “Eco-labels Stuck on Search for Common Standards,”Pulp and Paper International, Vol. 36, No. 11, pp. 39-42. Webb, Leslie (1993), “How Green is your Labelling Scheme?”, Pulp and Paper International, Vol. 35, No. 5, pp. 32-35.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

137

3.

POLICY VERSUS CONSUMER PRESSURE...

ANNEX 3.A1

Relevant Patent Classes for Pulp Bleaching Technologies

138

D21C 9/14

Paper/Paper-Making; Production of Cellulose/Production of Cellulose by Removing Non-cellulose Substances from Cellulose-containing Materials; Regeneration of Pulping Liquors; Apparatus Therefor/After-treatment of cellulose pulp, e.g. of wood pulp, or cotton liners/Bleaching/with halogens or halogencontaining compounds/with ClO2 or chlorites

D21C 9/147

Paper/Paper-Making; Production of Cellulose/Production of Cellulose by Removing Non-cellulose Substances from Cellulose-containing Materials; Regeneration of Pulping Liquors; Apparatus Therefor/After-treatment of cellulose pulp, e.g. of wood pulp, or cotton liners/Bleaching/with oxygen or its allotropic modifications (9/16 takes precedence)

D21C 9/153

Paper/Paper-Making; Production of Cellulose/Production of Cellulose by Removing Non-cellulose Substances from Cellulose-containing Materials; Regeneration of Pulping Liquors; Apparatus Therefor/After-treatment of cellulose pulp, e.g. of wood pulp, or cotton liners/Bleaching/with oxygen or its allotropic modifications (9/16 takes precedence)/with ozone

D21C 9/16

Paper/Paper-Making; Production of Cellulose/Production of Cellulose by Removing Non-cellulose Substances from Cellulose-containing Materials; Regeneration of Pulping Liquors; Apparatus Therefor/After-treatment of cellulose pulp, e.g. of wood pulp, or cotton liners/Bleaching/with per compounds

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Chapter 4

Renewable Energy Policies and Technological Innovation: Energy Source and Instrument Choice* by Nick Johnstone and Ivan Hascic (OECD Environment Directorate)

This paper examines the effect of environmental policies on technological innovation in the specific case of renewable energy. Patent data is collected for wind, solar, wave/tide, biomass, geothermal, and waste-to-energy technologies. Empirical analysis is conducted using patent data on a panel of 25 countries over the period 1978-2003. It is found that in addition to the influence of market factors (i.e. electricity prices) and expenditures on R&D, public policy plays a significant role in determining patent applications. However, different types of policy instruments are effective for different renewable energy sources.

* The authors wish to thank Piotr Tulej formerly of the International Energy Agency for the policy data used in this study. Thanks also to David Popp, Frans de Vries, Paul Lanoie, Jeremy Luchetti, Tom Jones and Nils Axel Braathen for valuable comments on a previous version.

139

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

1. Introduction In this Chapter, an assessment is made of the effects on innovation of different policies implemented to favour the development of renewable energy. The Chapter reports on empirical analysis undertaken with respect to a wide variety of policy types, using patent counts as an indicator of technological innovation. The first Section provides an overview of trends and technologies in the area of renewable energy, as well as the policies implemented in different OECD countries in support of its development. In the Second section, data on patents with respect to renewable energy are presented. In the third Section, the empirical model and empirical results are discussed. The Chapter concludes with a brief summary of the main policy implications.

2. The renewable energy sector: trends, technologies and policies Investment in renewable energy sources – wind, solar, geothermal, ocean, waste-to-energy, and biomass – can contribute significantly to the realisation of public environmental objectives. In addition, it is sometimes argued that such investment contribute to other public policy objectives, such as increased energy security, in the face of uncertain markets for fossil fuels. The penetration of renewables, although increasing, remains limited. In the absence of public intervention favouring their development, full life-cycle costs (including development, investment and operating costs) remain higher than for substitute fossil fuels. Governments are now introducing a wide variety of policy instruments to seek to accelerate market penetration. As such, the assessment of the role of public policy in inducing innovation in the development of renewable energy technologies is an interesting context in which to assess the effectiveness of different types of measures. Three generations of renewable energy technologies can be distinguished (IEA, 2006a):

140

●

First-generation technologies which have already reached maturity, such as hydropower, biomass combustion, and geothermal energy.

●

Second-generation technologies which are undergoing rapid development such as solar energy, wind power, and modern forms of bio-energy.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

●

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Third-generation technologies which are presently in developmental stages such as concentrating solar power, ocean energy, improved geothermal, and integrated bio-energy systems.

Figure 4.1 gives the growth rates for different renewable energy sources amongst OECD countries and for the rest of the world between 1990 and 2004 (IEA, 2006b), compared with the rate of growth of total primary energy supply. The rapid growth in wind power is evident. However, the relative importance of slow-growing renewables, such as hydropower and geothermal, means that the overall rate of growth (1.9%) has only been marginally above the growth rate of total primary electricity supply (TPES) (1.8%). Figure 4.1. Annual growth rates for renewable energy in the world and the OECD (1990-2004) World

OECD

30 25 20 15 10 5

ES TP

d W

in

r la

ga

So

s/ l b i iqui om d as s

o dr Hy

ot

W

as

te

/b

io

Ge

So

lid

bi

he

om

rm

as

s

al

0

Source: IEA (2006b).

In 2004, among the three main regions of the OECD, Europe had the highest share of renewable use in total energy production (6.9%). In North America (Canada, US and Mexico) the figure was 5.7%; for OECD Pacific (Japan, Korea, Australia and New Zealand) it was 3.4%. Biomass was the single largest source (44.6%), followed by hydropower (34.6%). Figure 4.2 gives a breakdown of the contribution of different renewable energy sources in the OECD (IEA 2006b). There are important differences within regions with respect to overall use and the contribution of different sources. For instance, in 2003, over 15% of total energy production (minus hydropower) was attributable to renewable sources in Iceland (principally geothermal), and Denmark (wind). Conversely,

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

141

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.2. Renewable energy sources in the OECD in 2004 Gas from biomass, 2.6% Renewable municipal waste, 3.2% Hydro, 34.6%

Solid biomass, 44.6%

Solar/Tidal, 1.0% Geothermal, 8.9% Liquid biomass, 3%

Wind, 2.1%

Source: IEA (2006b).

in the Czech Republic, France, Hungary, Norway, Poland and Slovakia, less than 1% of total energy production was attributable to renewable sources. Table 4.1 gives figures for renewable energy use as a percentage of electricity consumption for European countries. As noted above many governments have recently sought to encourage the further development of renewable energies (IEA, 2004). For instance, a European Union Directive of 2001 (Directive 2001/77/EC) provides a framework for the development of renewable energies in Europe. Figure 4.3 gives national targets for renewable energy use in European countries for 2010 (CEC 2004). In March 2007 EU Heads of State agreed to set a binding target of using renewable energy to meet 20% of total EU energy needs by 2020. To meet these targets a variety of policies have been introduced. These vary according to the point of incidence, and IEA (2004) distinguishes between policies which address supply or demand, and whether this targets generation or capacity in either case. An example of a policy which affects supply and capacity would be investment tax credits for investment in renewable energy facilities (e.g. wind turbines). A supply-side policy which targets generation could be obligations for electricity generators to use a minimum amount of renewable energy in their fuel mix. Demand-side policies which relate to capacity would target firms and households as end-users, such as through sales tax rebates and exemptions or grants on small-scale capital equipment (e.g. solar panels). And finally, demand-side policies which target generation would include measures such as public procurement or “green” energy pricing. In addition to differences in terms of the point of incidence of a policy, it is important to also distinguish policies according to the renewable energy source(s) concerned, since specific incentives are often provided for specific renewable types. Indeed, it is relatively rare for policies to be strictly neutral

142

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.1. Share of electricity production from renewable sources (excluding hydro)* (%) by country 1990

1995

2000

2005

Australia

0.4

0.4

0.5

1.2

Austria

2.3

3.3

2.9

5.5

Belgium

0.4

0.4

0.7

2.1

Canada

0.8

1.0

1.4

1.7

–

0.7

0.7

0.9

3.1

5.3

16.1

28.1

Finland

–

10.3

12.3

13.7

France

0.5

0.5

0.6

0.9

Germany

0.3

0.8

2.4

6.9

Greece

0.0

0.1

0.8

2.3

Hungary

0.1

0.2

0.2

4.7

Iceland

6.7

5.8

17.2

19.1

Czech Republic Denmark

Ireland

–

0.1

1.4

4.8

Italy

1.5

1.5

2.4

4.2

Japan

1.4

1.5

1.6

1.9

Korea

0.0

0.1

0.0

0.1

Luxembourg

2.1

3.9

11.5

3.4

Mexico

4.1

3.6

3.1

4.2

Netherlands

1.0

1.6

3.2

7.4

New Zealand

8.1

7.7

9.3

9.7

Norway

0.2

0.2

0.2

0.6

Poland

0.0

0.0

0.2

1.1

Portugal

2.5

3.2

3.6

7.6

–

–

–

0.1

Spain

0.4

0.8

2.8

8.2

Sweden

1.3

1.7

3.1

5.3

Switzerland

0.8

0.9

1.3

1.8

Turkey

0.1

0.4

0.2

0.1

United Kingdom

0.2

0.6

1.3

3.0

United States

3.0

2.0

1.9

2.2

Slovak Republic

* Renewable sources include geothermal, solar thermal, solar PV, tide, wind, renewable municipal waste, solid biomass, liquid biomass and biogas. Source: IEA/OECD Renewables Statistics and IEA/OECD Energy Balances of OECD Countries.

with respect to different types of renewable energy. It is also important to distinguish between policy types according to the precise nature of the instrument being applied, with the following criteria being potentially important determinants of their impacts on innovation: ●

whether the measure is price-based (i.e. taxes, exemptions, subsidies, etc.) or quantity-based (obligations, procurement, etc.);

●

whether the measure is mandatory or voluntary; and

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

143

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.3. Percentage of energy to be provided by renewable energy % 90 80 70 60 50 40 30 20 10

m do

en

ng Ki

d Un

i te

l

n

ed Sw

ai Sp

s

ga r tu

Po

nd

g Ne

th

er

la

ur

ly

bo

It a

m Lu

xe

nd la

ce

y an

ee

Ir e

Gr

ce

rm Ge

an Fr

an

ar

d

k

nl Fi

m

nm

iu

De

lg Be

Au

st

ria

0

Source: IEA (2006b).

●

whether a positive incentive is provided for renewables or a negative incentive is imposed upon fossil fuel substitutes.

The Global Renewable Energy Policies and Measures Database (www.iea.org) provides data on policies applied in over 100 countries in support of renewable energy. Some examples of recent significant policy measures introduced in OECD countries are provided in Table 4.2. Figure 4.4 provides summary data on the point at which the first “significant” example of given type of measure was introduced in a particular country. This indicates clearly that different policy types have been introduced with some temporal regularity. First, in the 1970s, a number of countries introduced support for R&D. This was followed by investment incentives (third-party financing, investment guarantees), taxes (exemptions, rebates), and price-based policies (tariffs, guaranteed prices). More recently, a number of countries have introduced quantity obligations, often followed by certificates in which the obligations are tradable across generators. Figure 4.5 provides a graphical representation of the introduction of different policy types in different countries (IEA, 2004).

3. Patent applications for renewable energy As with the other studies contained in this report, patents were used as a measure of innovation. Relevant patents applications have identified using the International Patent Classification (IPC) codes, developed at the World Intellectual Property Organisation (www.wipo.org). Based on an extensive

144

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.2. Examples of policies aimed at supporting renewable energy Year

Description of policy

Type

Belgium

2001

Green Certificates – From Jan. 1, 2002 all electricity suppliers have the obligation to buy a specified amount of “green” certificates from electricity producers (6% in 2010).

Tradable certificates

Germany

2004

Renewable Energy Sources Act – Prescribes fixed tariffs which grid operators must pay for renewable energy.

Feed-in tariffs

Germany

2005

Fifth Energy Research Programme – Sets the framework for public RD&D support for the development of renewable energies.

Subsidies for RD&D

Austria

2006

Green Electricity Act – 2006 Amendment. Provision of investment subsidies for new “renewable” energy power plants.

Investment subsidies

Denmark

2001

New Rules for Payment of Green Electricity – Fixed settlement price per kWh for initial load-hours of energy from wind turbines.

Feed-in tariffs

Japan

2002

Renewable Portfolio Standards – Annual obligation on electricity retailer to use a certain amount of electricity generated from renewable sources. Can be met in three ways: own generation; purchase from other generators; and, purchase of certificates.

Obligations and tradable certificates

Canada

2006

Sustainable Technology Development Canada – Subsidy payments up to a maximum of 33% of project costs.

Subsidy for RD&D and investment

United States

2004

Tax Incentives for Renewable Energy (Extension) – Tax exemption for financing of investments in renewable energy technologies.

Tax exemption for investment

Italy

2000

Tax Credit for Geothermal and Biomass-Fuelled District Heating – Users connected to a geothermal or biomass fuelled district-heating grid receive a tax credit.

Tax credits for users

United Kingdom

2002

Renewables Obligation Order. Electricity supply companies are required to source a percentage of electricity from renewable sources (7.9% in 2007/2008). Can be met by either surrendering a certificate or paying a contribution.

Obligations

Netherlands

2003

MEP: Environmental Quality of Electricity Production – Subsidy paid to domestic producers of electricity generated from renewable sources.

Production subsidy

Poland

2000

Obligation for Power Purchase from Renewable Sources: Distribution companies obliged to provide a certain share of energy produced from renewable sources (5% in 2008).

Obligations

Source: IEA/JRC Global Renewable Energy Policies and Measures Database (www.iea.org/textbase/pm/grbackground.htm) accessed Nov. 22, 2007).

literature of technology developments in the area of renewable energy, a set of keywords were identified for this study. These were used to determine appropriate IPC codes which relate directly to renewable energy in the areas of wind, solar, geothermal, ocean, biomass, and waste. Table 4.3 gives the relevant codes and their definitions. (A sample patent for technological development with respect to wind energy is provided in Annex 4.A1. More information on the sample patent is available at http://v3.espacenet.com/textdoc?DB=EPODOC&IDX= WO9404820&F=0&QPN=WO9404820.) The patent data used in this study were drawn from EPO/OECD PATSTAT database. The count represents the number of patent applications to the EPO, based on the priority date and inventor country, using fractional counts.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

145

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.4. Introduction of policies by type for renewable energy in OECD countries (1973-2003) Research and Development

D FI DK

B

NO C NZ NE CH US E ITA SW UK J

IR K

GR A

T

F

L

K

GR

H

P

AUS

CZ

FI

Investment incentives

US

ITA

T

F

D

A

P

CH

J

B

L

C

NE

US

F

T IR

L

E

Tariffs

US

D

UK DK

D FI NO P CZ

DK

GR ITA SW B

NZ

NO

SW

C NE

Taxes

E UK

AUS

H DK

A

UK E K

ITA L

P CH

A

GR

B

F

SW

K

H NO

IR

NE

C

CZ

Voluntary programmes

US CH

D

P

Obligations

AUS

DK A AUS

C

NZ

B

J

L

K EU CZ L H D J GR NO F UK

E

CH FI

ITA

NE ITA

IR SW AUS H

Tradable permits

NE

ITA

NO A FI

1970

1973

1976

1979

1982

1985 1988

1991

1994

1997

2000

B

SW UK J DK K

2003

AUS : Australia – C : Canada – FI : Finland – GR : Greece – ITA : Italy – L : Luxembourg NO : Norway – SW : Sweden – UK : United Kingdom – A : Austria – CZ : Czech Republic F : France – H : Hungary – J : Japan – NE : Netherlands – P : Portugal – CH : Switzerland US : United States Source: An updated version of the table published in IEA (2004) was kindly provided by Piotr Tulej of the International Energy Agency.

Counts were obtained for all important patenting countries, including nonEuropean countries. While the European market is significant, it is still expected that there will be some bias toward applications from European inventors (see Dernis and Guellec, 2001). In the empirical analysis undertaken here this bias is addressed through the inclusion of both fixed effects and data on total EPO applications to reflect different propensities to patent in general across countries and through time. Figure 4.5 gives data for total patent applications for six different renewable energy sources. Geothermal applications fell off dramatically after the late 1970s, while there has been continuous growth in patenting for solar power technologies. Wind power and waste-to-energy exhibited even more rapid growth, particularly since the mid-1990s. There were very few patents for wave-tidal energy and geothermal, but they are also now increasing.1 Figure 4.6 compares total patent applications for a selection of OECD countries which exhibited significant levels of innovation. Germany had the highest number of patents, but relative to the US and Japan, this partly reflects the “home bias” in EPO applications. France and the UK both have at least

146

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.3. IPC Classifications for Renewable Energy Class

Sub-Classes

WIND Wind motors with rotation axis substantially in wind direction Wind motors with rotation axis substantially at right angle to wind direction Other wind motors Controlling wind motors Adaptations of wind motors for special use; Details, component parts, or accessories not provided for in, or of interest apart from, the other groups of this subclass

F03D F03D F03D F03D F03D F03D

1/00-06 3/00-06 5/00-06 7/00-06 9/00-02 11/00-04

Electric propulsion with power supply from force of nature, e.g. sun, wind Effecting propulsion by wind motors driving water-engaging propulsive elements

B60L

8/00

B63H

13/00

F03G F24J

6/00 – 08 2/00 – 54

F25B F26B

27/00B 3/28

H01L

31/042

Aspects of roofing for the collection of energy – i.e. Solar panels

H02N E04D

6/00 13/18

Electric propulsion with power supply from force of nature, e.g. sun, wind

B60L

8/00

F24J

3/00 – 08

F03G H02N

4/00-06 10/00

F03B

13/12-24

F03G F03G

7/05 7/04

F03B

7/00

C10L F02B

5/42-44 43/08

C10L B01J

1/14 41/16

C10L F25B F02G F23G F012K C10J F23G H01M

5/46-48 27/02 5/00-04 5/46 25/14 3.86 7/10 8/06

SOLAR Devices for producing mechanical power from solar energy Use of solar heat, e.g. solar heat collectors Machine plant or systems using particular sources of energy – sun Drying solid materials or objects by processes involving the application of heat by radiation – e.g. sun Semiconductor devices sensitive to infra-red radiation – including a panel or array of photoelectric cells, e.g. solar cells Generators in which light radiation is directly converted into electrical energy

GEOTHERMAL Other production or use of heat, not derived from combustion – using natural or geothermal heat Devices for producing mechanical power from geothermal energy Electric motors using thermal effects WAVE/TIDE Adaptations of machines or engines for special use – characterised by using wave or tide energy Mechanical-power producing mechanisms – ocean thermal energy conversion Mechanical-power producing mechanisms – using pressure differentials or thermal differences Water wheels BIOMASS Solid fuels based on materials of non-mineral origin – animal or vegetable Engines operating on gaseous fuels from solid fuel – e.g. wood Liquid carbonaceous fuels – organic compounds Anion exchange – use of materials, cellulose or wood WASTE Solid fuels based on materials of non-material origin – refuse or waste Machine plant or systems using particular sources of energy – waste Hot gas or combustion – Profiting from waste heat of exhaust gases Incineration of waste – recuperation of heat Plants or engines characterised by use of industrial or other waste gases Prod. of combustible gases – combined with waste heat boilers Incinerators or other apparatus consuming waste – field organic waste Manufacture of fuel cells – combined with treatment of residues

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

147

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.5. Number of EPO patent applications by type of renewable Wind Wave-tide

Solar Biomass

Geothermal Waste

180 160 140 120 100 80 60 40 20 0 1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

Source: OECD Patent Project (www.oecd.org/document/10/0,2340,en_2649_33703_1901066_1_1_1_1,00.html).

Figure 4.6. Number of EPO patent technologies for renewables by country DE

US

JP

GB

FR

160 140 120 100 80 60 40 20 0 1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

Source: OECD Patent Project (www.oecd.org/document/10/0,2340,en_2649_33703_1901066_1_1_1_1,00.html).

200 patent applications over the period. In addition to these countries, there were specific areas in which individual countries have been important innovators for specific renewables. In Table 4.4, the counts are averaged across all years. In addition to Germany, Japan and the US (countries which are consistently important for most renewables), other significant innovating countries for particular sources have included Denmark (wind), Switzerland (solar, geothermal), France (geothermal, biomass, waste), the UK (ocean, biomass and waste), Italy

148

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.4. Number of EPO patent filings in renewable energy technologies (annual average 1978-2003, by inventor country) Wind

Solar

Geothermal

Ocean

Biomass

Waste

All 1978-2003 renewables Total

AT

0.46

1.19

1.23

0.15

0.27

0.92

4.23

110

AU

0.19

1.88

0.42

0.19

0.12

0.42

3.23

84

BE

0.92

0.50

0.42

0.04

0.15

0.31

2.35

59

BR

0.00

0.00

0.00

0.04

0.12

0.04

0.19

5

CA

0.58

0.54

0.23

0.08

0.12

1.15

2.65

66

CH

0.50

2.08

1.31

0.08

0.15

1.31

5.42

138

DE

13.88

12.81

7.00

0.69

3.54

11.77

49.46

1 285

DK

3.38

0.46

0.19

0.42

0.15

0.73

5.35

137

ES

0.96

0.81

0.08

0.46

0.00

0.08

2.38

61

FI

0.27

0.35

0.12

0.08

0.00

0.50

1.31

34

FR

1.81

1.92

2.85

0.35

1.62

1.88

10.38

267

GB

1.96

1.31

0.92

0.96

5.46

1.92

12.38

322

GR

0.19

0.19

0.00

0.08

0.00

0.08

0.54

14

HU

0.08

0.19

0.19

0.08

0.04

0.00

0.58

15

IE

0.19

0.15

0.00

0.19

0.00

0.00

0.54

14

IT

1.19

1.35

0.92

0.62

0.31

1.31

5.69

148

JP

1.73

7.19

1.73

0.42

0.77

13.42

25.27

656

KR

0.31

0.04

0.00

0.04

0.04

0.15

0.58

15

NL

2.00

1.58

0.96

0.19

0.35

1.19

6.19

161

NO

0.31

0.27

0.19

0.46

0.04

0.12

1.38

36

NZ

0.04

0.00

0.04

0.04

0.04

0.12

0.27

7

PL

0.04

0.08

0.08

0.00

0.04

0.04

0.27

7

PT

0.19

0.19

0.00

0.04

0.00

0.04

0.46

12

SE

1.35

0.62

1.12

0.62

0.15

0.42

4.27

109

TW

0.19

0.15

0.04

0.04

0.00

0.19

0.62

16

US

3.77

5.88

3.69

2.04

8.65

11.65

35.73

925

942

1 079

616

216

566

1 285

4 702

Total

Note: The Table provides the annual mean number of patent filings during 1978-2003, classified by inventor country. The last column (row) gives the total number of patent filings in renewables in a given country (technology).

(ocean), Netherlands (wind), and Sweden (ocean). (Countries in the top five for each renewable are indicated in bold face.) In Table 4.5, the counts are weighted by the country’s GDP, to yield a measure of patent intensity, relative to the size of the economy. On this basis, a number of smaller countries, such as Denmark, Switzerland, Austria, and Sweden achieve the highest innovation output per unit of GDP. Of the three countries which have the highest absolute counts, only Germany continues to rank consistently in the top five. However, Japan and the US remain first and third among non-EPO countries, with Australia being second.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

149

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.5. Number of EPO patent filings in renewable energy technologies (annual average 1978-2003, per unit of GDP, by inventor country) Wind

Solar

Geothermal

Ocean

Biomass

Waste

All 1978-2003 renewables Total

AT

2.50

6.20

8.14

0.67

1.40

4.85

23.76

110

AU

0.41

4.41

1.48

0.43

0.38

0.78

7.89

84

BE

3.96

2.12

2.29

0.21

0.58

1.22

10.39

59

BR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

5

CA

0.75

0.76

0.44

0.12

0.15

1.52

3.68

66

CH

2.72

10.36

7.96

0.45

0.81

6.83

29.14

138

DE

7.20

6.96

4.97

0.42

1.87

6.29

27.59

1 285

DK

23.10

3.28

1.89

2.99

1.48

5.81

38.56

137

ES

1.27

1.08

0.14

0.68

0.00

0.11

3.28

61

FI

2.63

3.10

1.35

0.56

0.00

4.27

11.90

34

FR

1.44

1.54

2.84

0.29

1.31

1.41

8.81

267

GB

1.58

1.06

1.00

0.84

4.30

1.53

10.18

322

GR

1.20

1.07

0.00

0.45

0.00

0.53

3.24

14

HU

0.39

1.95

0.37

0.00

0.39

0.00

3.11

15

IE

3.97

2.14

0.00

2.43

0.00

0.00

8.54

14

IT

0.91

1.03

0.89

0.49

0.23

0.98

4.52

148

JP

0.54

2.48

0.72

0.15

0.27

4.33

8.49

656

KR

0.44

0.05

0.00

0.07

0.05

0.21

0.81

15

NL

5.40

3.97

3.27

0.53

1.02

3.09

17.09

161

NO

2.12

1.99

1.76

3.41

0.25

0.70

10.24

36

NZ

0.54

0.00

0.54

0.60

0.60

1.45

3.73

7

PL

0.10

0.23

0.00

0.00

0.10

0.13

0.56

7

PT

1.59

1.17

0.00

0.22

0.00

0.26

3.25

12

SE

7.10

3.36

6.55

3.00

0.79

2.19

22.99

109

TW

0.00

0.00

0.00

0.00

0.00

0.00

0.00

16

US

0.53

0.77

0.64

0.31

1.15

1.43

4.84

925

Total

942

1 079

616

216

566

1 285

4 702

Note: The Table provides the annual mean number of patent filings during 1978-2003, classified by inventor country, and weighted by country’s GDP (in trillions of US dollars, using PPP and 2000 prices).

As noted in the introductory Chapter of this volume, there can be differences in the propensity to patent across countries. In order to control for this effect, Table 4.6 normalises the patent counts with respect to renewable energy by total EPO patents for all technological fields. In this case it is Denmark, Norway, Spain, Australia and Austria which have the highest number of patents. The “big three” have no particular revealed comparative advantage in renewable energy patenting. And finally, in order to address the concern that the inventive activity of non-European countries may be biased downward, due to the fact that the data used here are based on patent filings with the EPO, Figure 4.7 shows the European and non-European countries separately.

150

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.6. Number of EPO patent applications in renewable energy technologies, normalised by overall patenting activity (1978-2003) Wind

Solar

Geothermal

Ocean

Biomass

Waste

All renewables

AT

0.67

1.75

AU

0.39

3.75

1.76

0.22

0.39

1.33

6.13

0.86

0.42

0.25

0.84

BE

1.31

6.50

0.69

0.61

0.06

0.17

0.44

CA

3.28

0.73

0.68

0.30

0.08

0.16

1.38

3.34

CH

0.29

1.15

0.75

0.03

0.08

0.74

3.03

DE

1.10

1.01

0.55

0.05

0.27

0.93

3.91

DK

7.65

1.03

0.44

0.92

0.35

1.64

12.04

ES

2.62

2.29

0.24

1.31

0.00

0.24

6.70

FI

0.47

0.60

0.20

0.13

0.00

0.85

2.25

FR

0.37

0.39

0.60

0.07

0.33

0.39

2.15

GB

0.51

0.35

0.24

0.25

1.46

0.50

3.32

IT

0.53

0.61

0.42

0.28

0.14

0.59

2.57

JP

0.16

0.65

0.16

0.04

0.07

1.21

2.29

KR

0.62

0.08

0.00

0.08

0.08

0.31

1.16

NL

1.11

0.88

0.55

0.10

0.20

0.68

3.52

NO

1.68

1.41

1.01

2.39

0.20

0.61

7.31

SE

1.05

0.49

0.90

0.49

0.11

0.31

3.36

TW

1.48

1.19

0.30

0.30

0.00

1.48

4.75

US

0.21

0.33

0.21

0.11

0.48

0.66

2.01

Figure 4.7. EPO patent filings in renewable energy technologies (annual mean, per unit of GDP) 40

10

35 30 25 20

5

15 10 5 0

0 DK CH DE AT SE NL FI BE NO GB FR IE IT ES PT GR HU PL

JP AU US NZ CA KR TW BR

Note: The figure provides the annual mean number of patent filings during 1978-2003, classified by inventor country, and weighted by country’s GDP (in trillions of US dollars, using PPP and 2000 prices).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

151

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.8. Relationship between point of introduction of policies and patent counts Germany 150 125 R-D

INV

TAR

VOL

100 75 50 TAX

OBLIG

25 0 1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2000

2002

Denmark 20 18

R-D

16

INV

TAR

1980

1982

OBLIG

TAX

14 12 10 8 6 4 2 0 1978

1984

1986

1988

1990

1992

1994

1996

1998

Japan 100

80 INV 60 OBLIG 40 VOL 20

0 1978

152

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Clearly, natural conditions play a role in explaining national variations in patent activity by renewable type in many cases, but this is not always evident. Other factors are also at play. In order to get a first indication of the relative importance of public policy factors on patenting, a comparison of total renewable energy patent applications and the introduction of specific policy tools is provided in Figure 4.8 for Germany, Denmark and Japan. There is no obvious correlation between the introduction of different policies and “spikes” in patent activity, except perhaps the introduction of tariffs in Germany (with some lag), obligations and taxes in Denmark, and investment credits in Japan.

4. Empirical analysis Using the patent count data obtained it is possible to assess the determinants of patenting activity for renewable energy in a more formal manner. In total, a panel of data for 26 countries and 26 years is available, but the presence of missing observations for some of the variables (often R&D data) reduces the size of the samples to between 400 and 500 in the models estimated. Explanatory variables included in the model were the following:

4.1. Expenditures on R&D Patent activity is clearly a result of scientific capacity. While it is difficult to get a precise measure of national capacity for innovation, data was obtained from the IEA on national public sector expenditures on R&D disaggregated by type of renewable energy (IEA, Energy Technology Research and Development Database, 2006, http://data.iea.org.) The sign on this variable is expected to be positive.

4.2. Electricity consumption Returns on innovation are affected by the potential market for this innovation. In the case of renewable energy, this is best reflected in trends for electricity consumption. A growing market for electricity should increase incentives to innovate with respect to renewable energy technologies. This data was obtained from the IEA (IEA, Energy Balances of OECD Countries, 2006, http://data.iea.org.).

4.3. Price of electricity The commercial viability of renewable energy is dependent in large part on the price of electricity. Since electricity production costs are often greater than for fossil fuels, an increase in the price of electricity should increase incentives for innovation in the area of renewable energies. Since renewable sources represent a relatively small proportion of total electricity generation, it is assumed that this is exogenous. This data was obtained from the IEA (IEA, Energy Prices and Taxes Data Service, 2006, http://data.iea.org.), weighting price indices for residential and industrial use by consumption levels. The sign is expected to be positive.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

153

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

4.4. EPO applications There will be a changing propensity to patent within a country across time, both because different strategies may be adopted to capture the rents from innovation, and because legal conditions may change through time. In addition, there will be different propensities across countries to patent at a specific office, due, for instance, to issues such as the home bias that was discussed earlier. As such a variable was included reflecting overall EPO patent applications in all areas (OECD, Triadic Patent Family Database, 2006).2 The sign is expected to be positive.

4.5. Public policies Binary variables were also created for different policy “types”, such as R&D support, taxes, investment incentives, differentiated tariffs, voluntary programmes, obligations, and tradable certificates (see IEA 2004). This variable takes on a value of 0 prior to introduction of the policy, and 1 thereafter. The policies reflect the first introduction of “significant” policy measures as determined by the delegates to the IEA Working Party on Renewable Energy. While there are likely to be important differences between instruments in terms of the “stringency” of the measures introduced, this shortcoming is unavoidable for any cross-comparative analysis in which multiple instruments are included.3 Table 4.7 provides basic descriptive statistics for the explanatory variables. A negative binomial model was estimated to evaluate the effects of environmental policy and market-related determinants on patenting activity in renewable energy technologies. A dataset consisting of a panel of 26 countries over the period 1978-2003 was used to estimate the regression models. The full model is specified as:

ln( PatentCountit ) = β1.PRICEit + β 2 .R & Dit + β3 .CONSit + β 4 .EPOit + β5 .POLICYit + α i + ε it

[1]

where i = 1,…,26 indexes the cross-sectional unit (country) and t = 1978, …, 2003 indexes time. In different specifications a variety of lag structures were imposed to reflect the dynamic nature of innovation. This is further discussed below. The dependent variable, patenting activity, is measured by the number of successful and unsuccessful patent applications in each of the technological areas of renewable energy (wind, solar, geothermal, ocean, biomass, and waste). Fixed effects (i) are introduced to capture unobservable country-specific heterogeneity. All the residual variation is captured by the error term (i) , with exp(i) assumed to be gamma-distributed with mean 0 and variance 1.4

154

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.7. Descriptive statistics of explanatory variables (1978-2003) Variable

Obs.

Mean

Std. dev.

Electricity price (US$/unit, using PPP)

583

0.0849

0.0345

Electricity consumption (millions GWh)

624

0.0158

0.0323

Total EPO patent filings (thousands)

673

2.3964

4.9912

Wind R&D (10E + 9 USD, 2005 prices and PPP)

478

0.0063

0.0140

Solar R&D (10E + 9 USD, 2005 prices and PPP)

479

0.0237

0.0702

Ocean R&D (10E + 9 USD, 2005 prices and PPP)

477

0.0016

0.0077

Bioenergy R&D (10E + 9 USD, 2005 prices and PPP)

478

0.0086

0.0157

Renewables R&D (10E + 9 USD, 2005 prices and PPP)

482

0.0481

0.1261

R&D support

676

0.8432

0.3639

Investment incentives

676

0.4127

0.4927

Tax measures

676

0.2722

0.4454

Tariffs

676

0.3151

0.4649

Voluntary programmes

676

0.1050

0.3068

Obligations

676

0.2130

0.4097

Tradable certificates

676

0.0577

0.2333

Technology-specific R&D expenditures

Policy dummies

Several alternative specifications of the model were estimated. Table 4.8 presents the estimation results when all policy dummies are included in the regressions, except the dummy for R&D programmes (due to correlation with the intercept).5 The coefficient of the electricity price has a positive sign in every equation. It is statistically significant at the 1% and 5% levels in the solar and biomass equations, respectively. This suggests that higher electricity prices provide an incentive for increased patenting activity in the solar and biomass technologies. The results also suggest that technology-specific R&D spending is a significant determinant of patenting in renewable energy overall, and especially in the case of wind and ocean technologies. The estimated coefficient of electricity consumption is statistically significant only in the “waste-toenergy” equation. The estimated coefficient of the total number of EPO filings is statistically significant at the 1% level in every technological area, suggesting that a part of the variation in patenting activity in renewable energies is due to changes in the propensity to patent.6 The results on the policy dummies suggest that public policy plays a significant role in inducing innovations in renewable energies. For wind technology and for renewable energies overall, tax measures, obligations, and tradable certificates are statistically significant (at the 5% level and higher) policy instruments. However, the efficacy of alternative policy instruments in inducing innovations varies by the technological area. For example, providing investment incentives is a statistically significant policy for solar energy innovations. Investment incentives, as well as voluntary programmes, are

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

155

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.8. Estimated coefficients of the negative binomial fixed effects models with individual policy variables

Electricity price

Wind

Solar

3.187

18.718**

(0.488) Specific R&D expenditures

17.789**

Waste

All renewables

2.181

14.769*

2.957

0.994

(0.000)

(0.737)

(0.035)

(0.469)

(0.683)

0.966

13.889*

–7.473*

0.479

(0.038)

(0.043)

(0.249)

–9.630

–8.060

–15.200

–15.900

–13.600**

–5.030

(0.123)

(0.141)

(0.335)

(0.115)

(0.005)

(0.162)

0.106**

Total EPO filings

Biomass

(0.153)

(0.000) Electricity consumption

Ocean

(0.001)

0.074** (0.000)

0.069

0.121**

(0.188)

(0.001)

–0.097

–0.176

(0.740)

(0.481)

0.122** (0.000)

1.063** (0.000)

0.081** (0.000)

Policy dummies Investment incentives

–0.214 (0.292)

Tax measures Tariffs Voluntary programmes

Tradable certificates Intercept Number of observations Log-likelihood

(0.000)

0.723** (0.000)

0.145 (0.146)

0.371*

–0.021

0.538

0.500*

0.083

0.235*

(0.040)

(0.881)

(0.089)

(0.050)

(0.578)

(0.017)

–0.434

0.116

0.015

(0.053)

(0.547)

(0.964)

0.783** (0.000)

0.192

–0.043

(0.336)

(0.717)

0.089

0.020

–0.066

–0.240

0.334*

0.119

(0.718)

(0.898)

(0.863)

(0.307)

(0.043)

(0.318)

0.181

0.472

–0.212

0.045

(0.000)

(0.214)

(0.155)

(0.372)

(0.761)

0.485*

0.064

0.192

–0.081

0.245

0.305*

(0.034)

(0.718)

(0.597)

(0.798)

(0.159)

(0.016)

–0.214

0.267

15.394

1.012

0.372

(0.598)

(0.685)

(0.992)

(0.371)

(0.509)

452

427

450

334

441

463

–477.65

–488.20

–238.56

–289.98

–482.30

–926.40

1.157**

Obligations

0.626**

0.384** (0.001)

0.995** (0.000)

Wald chi2

250.63

209.21

33.22

80.03

337.01

398.90

(Prob > chi2)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

Notes: * and ** refer to 5% and 1% level of statistical significance. P-values are in parentheses. The dependent variable is the natural log of patent counts (successful and unsuccessful applications) in a given technological area. Intercept represents the average value of the country-specific fixed effects. Results for geothermal energy are not reported because they represent a significant outlier. The R&D dummy is not included in the regressions because it is (almost perfectly) correlated with the intercept (fixed effects). This is due to the fact that most countries introduced R&D support programmes as early as 1970s, but the data here covers only the period after 1978.

statistically significant policy instruments for waste-to-energy incineration. Finally, putting in place preferential tariff structures, and to a lesser extent tax measures, are statistically significant policies with respect to biomass energy. There are two concerns related to including all policy dummies in the regressions. First, correlation among the dummies may cause multicollinearity problems. In particular, dummies representing investment incentives, tax measures, and tariffs are highly correlated Table 4.9).7 Similarly, obligations and tradable permits are highly correlated.8 Second, it is possible that there are interaction effects among alternative policy instruments (e.g. investment

156

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.9. Correlation coefficients between policy variables Investment incentives

Tax measures

Tariffs

Voluntary programmes

Tax measures

0.572

Tariffs

0.531

0.524

Voluntary programmes

0.368

0.222

0.347

Obligations

0.442

0.369

0.470

0.343

Tradable certificates

0.283

0.264

0.205

0.197

Obligations

0.484

incentives for capital goods may be accompanied by preferential tax rates for final goods). In an alternative specification, the individual policy dummies were included one-by-one in the regressions. The results (not reported) suggest that the key qualitative findings remain unaffected. Policies which are found to be statistically significant when all dummies are included in the regression remain statistically significant, and with the same signs, when they are included separately. Additional policy dummies are statistically significant when included separately. Thus, including all policy dummies may cause multicollinearity. However, including policy dummies one-by-one may lead to incorrect conclusions (due to omitted variables), and possible interaction effects among the different policies. In order to address these issues: a) a composite policy variable was constructing representing the number of policies in place; and b) clusters of policy variables were developed by clustering similar policies in groups. A composite policy variable is an index that reflects differences in the “richness” of policy approaches across countries and over time. It is constructed by summing across all seven policy dummies. It thus represents the (cumulative) number of policies in place (including the R&D dummy in this case). The disadvantage of this approach is that it does not distinguish among the individual policy instruments. The advantage is that it implicitly deals with the correlation and potential interaction effects among policy instruments. In addition, the composite policy variable can be lagged, allowing analysis of dynamic issues. Table 4.10 presents the estimation results using the composite policy variable. Overall, the results remain robust. The estimated coefficient of the policy variable is positive and statistically significant at the 5% level in every equation. This suggests that public policy is a significant driver of innovative activity in renewable energy overall, as well as in the specific technological areas. Using the same model, the dynamic aspects of innovative activity were also analysed by lagging the composite policy variable. Including the current policy variable or its 1-year, 2-year, or 3-year lag separately yields qualitatively identical results. When the policy variable and its first three lagged values are all included in a regression, the current policy variable is positive and statistically

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

157

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.10. Estimated coefficients of the negative binomial fixed effects models with a composite policy variable

Electricity price Specific R&D expenditures

Wind

Solar

–4.362

19.835**

1.501

12.394

5.028

0.406

(0.304)

(0.000)

(0.811)

(0.079)

(0.188)

(0.862)

15.067**

Intercept Number of observations Log-likelihood

All renewables

13.999*

–7.304

0.496

(0.027)

(0.059)

(0.238)

(0.000)

–15.567**

–6.144

–16.908

–1.146

–9.817*

–8.125*

(0.005)

(0.208)

(0.253)

(0.868)

(0.020)

(0.015)

0.055

0.063*

(0.249)

(0.025)

(0.000) Composite policy variable

Waste

0.971

0.109**

Total EPO filings

Biomass

(0.144)

(0.000) Electricity consumption

Ocean

0.374**

0.064** (0.000) 0.173**

0.238**

0.111*

0.119** (0.000) 0.237**

(0.000)

(0.000)

(0.000)

(0.030)

(0.000)

–0.774*

–0.134

14.147

0.041

–0.317

(0.036)

(0.805)

(0.985)

(0.950)

(0.487)

1.110**

0.089** (0.000) 0.207** (0.000) 0.759** (0.001)

452

427

450

334

441

463

–496.61

–494.88

–240.15

–299.02

–487.99

–931.35

Wald chi2

144.73

200.24

30.35

48.76

304.14

363.56

(Prob > chi2)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

Notes: * and ** refer to the 5% and 1% level of statistical significance. P-values are in parentheses. The dependent variable is the natural log of patent counts (successful and unsuccessful applications) in a given technological area. The coefficient on the intercept represents the average value of the country-specific fixed effects. Results for geothermal energy are not reported, because they represent a significant outlier.

significant,9 while the lags are insignificant.10 Finally, there could be a concern that variables such as electricity consumption, EPO patent filings, and fixed effects may all, to a certain extent, reflect the same tendencies. Including all of these variables in a regression may cause “over-fitting” of the model. However, even when country-specific fixed effects are dropped from the regression, the key qualitative findings remain robust. As an alternative to a simple summation of the policy types, hierarchical cluster analysis can be used to identify clusters of policy instruments which are then used as explanatory variables. Cluster analysis is a method that can be applied in order to reduce a set of correlated variables into a smaller number of cluster components, with little loss of information. It groups variables in such a manner that variables within one group are correlated, but uncorrelated with variables in the other groups. Figure 4.9 shows the estimated tree diagram of hierarchical clusters (dendogram) for the six policy variables. The choice of the number of clusters to retain must be made ad hoc. The clustering of policy variables which yields three conceptually distinct groups of policies was chosen: 1) price-based policy instruments (including investment incentives, tax measures, and tariffs), 2) voluntary programmes; and 3) quantitybased policy instruments (obligations and tradable certificates). The voluntary programmes dummy is relatively equally correlated with either of the remaining

158

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Figure 4.9. Dendogram of policy variable clustering Investment Tax measures Tariff Voluntary Obligations Tradable 6.0

5.5

5.0

4.5

4.0

3.5

3.0

2.5

2.0 1.5 1.0 Number of clusters

two clusters. The estimated scoring coefficients were used to compute component scores for each cluster. Table 4.11 shows the regression estimates when policy variables are divided into three clusters. The estimated results suggest that innovation effects of alternative policy instruments differ by the type of renewable energy technology. The evidence of differential policy effectiveness is particularly straightforward for wind, solar, and waste-to-energy technologies. For wind power, the coefficient of quantity-based policy instruments (cluster 3) is positive and statistically significant at the 1% level. For solar and waste energy, the coefficients of price-based policy instruments (cluster 1) are positive and statistically significant at the 1% level. In addition, the coefficient of voluntary programmes (cluster 2) in the waste equation is positive and statistically significant at the 5% level or higher. For ocean energy, none of the policy cluster variables are significant. For biomass, the coefficient of price-based instruments (cluster 1) is positive and statistically significant at the 1% level. Overall, for all renewables, the innovation effects of both price- and quantitybased instruments are highly statistically significant. Voluntary approaches seem to play a minor role. In sum, the major finding of this Chapter is that different policies work better for some technologies than for others. In particular, quantity-based policy instruments such as obligations and tradable certificates are most effective in inducing innovations in wind power technology. Price-based instruments, such as investment incentives, tax measures and tariffs are most effective in encouraging innovation in solar, biomass, and waste-to-energy technologies. Voluntary programmes are not significant, except perhaps in the case of waste. These findings are robust to alternative policy measures and model specifications.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

159

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Table 4.11. Estimated coefficients of the negative binomial fixed effects models with clusters of policy variables

Electricity price Specific R&D expenditures

Wind

Solar

–2.465

20.112**

1.787

12.459

5.013

0.094

(0.547)

(0.000)

(0.775)

(0.094)

(0.190)

(0.968)

16.944**

Waste

All renewables

1.100

15.028*

–6.705

0.490

(0.023)

(0.067)

(0.245)

–11.551

–7.088

–13.868

–9.658

–11.160*

–5.825

(0.073)

(0.175)

(0.361)

(0.253)

(0.011)

(0.109)

0.121**

Total EPO filings

Biomass

(0.091)

(0.000) Electricity consumption

Ocean

(0.000)

0.075** (0.000)

0.064 (0.199)

0.094** (0.004)

0.116** (0.000)

1.069** (0.000)

0.087** (0.000)

Policy clusters 0.614**

0.728**

0.310**

–0.018

(incl. inv, tax, tar)

(0.941)

(0.000)

(0.216)

(0.000)

(0.000)

Policy cluster 2

–0.006

0.030

–0.075

–0.127

0.366*

0.072

(incl. vol)

(0.980)

(0.843)

(0.842)

(0.573)

(0.031)

(0.521)

0.209

0.665

–0.508

0.219

1.632**

Policy cluster 3

0.419

1.053**

Policy cluster 1

(incl. oblig, trad)

(0.000)

(0.184)

(0.055)

(0.074)

(0.184)

Intercept

–0.132

0.006

13.727

0.346

–0.027

(0.726)

(0.991)

(0.982)

(0.646)

(0.955)

Number of observations Log-likelihood

(0.005)

0.639** (0.000) 1.029** (0.000)

452

427

450

334

441

463

–483.95

–493.77

–239.69

–293.72

–487.22

–927.89

Wald chi2

216.29

200.81

31.89

64.08

297.49

393.14

(Prob > chi2)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

Notes: * and ** refer to 5%, and 1% level of statistical significance. P-values are in parentheses. The dependent variable is the natural log of patent counts (successful and unsuccessful applications) in a given technological area. The coefficient on the intercept represents the average value of the country-specific fixed effects. Results for geothermal energy are not reported because they represent a significant outlier. Policy cluster 1 includes investment incentives, tax measures, and tariffs; Policy cluster 2 includes voluntary programmes; Policy cluster 3 includes obligations and tradable certificates.

Interpretation of these results is complicated by the fact that it is difficult to clearly distinguish the relevance of the policy variable by type of renewable. However, there are good economic reasons to explain some of the findings. For instance, the significance of investment incentives for solar energy is likely due to the fact that many solar installations are the most capital intensive (in terms of investment costs required per kW) among the studied renewable energy technologies (see, for example, Dickson and Fanelli 2004). Waste-to-energy investments can also involve significant up-front fixed costs, and the coefficient is significant and positive in this case as well. In addition, the relative importance of voluntary programs for waste may be explained by the importance of the public sector in waste management, perhaps obviating the need for mandatory regulations. The significant and positive coefficient on price-based measures (feed-in tariffs) for biomass may be explained by the fact that this source is relatively mature and competitive, and thus changes in relative prices are sufficient to have an effect on innovation. However, further work is required to support (or reject) these conclusions.

160

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

5. Conclusions This Chapter has examined the effects of public policies on innovation in the area of renewable energies in a cross-section of OECD countries over the period 1978-2003. Patent counts were used as the most suitable proxy for innovation, and the effects of a wide variety of different policy types were assessed. The descriptive data indicates rapid growth in wind and waste-to-energy patent activity, particularly since the mid-1990s. There continues to be innovation with respect to solar energy, perhaps reflecting the opportunities presented by developments in concentrating solar power. Innovation with respect to biomass and ocean are also growing, but from a very low base. And finally, there appears to have been little innovation in the area of geothermal energy since the 1970s. At the same time there have been significant changes in the public policy framework put in place to support renewable energy. Initially R&D programmes were put in place in a number of countries. This was followed by investment incentives, and later tax incentives and preferential tariffs. Later still, voluntary programmes were developed. More recently, quantitative obligations, and finally tradable certificates, have been applied. The empirical results indicate that public policy has had a very significant influence on the development of new technologies in the area of renewable energy. Using the composite policy variable, statistical significance at the 1% level is found for all renewable energy sources, except biomass (where it is significant at the 5% level). However, the results suggest that instrument choice also matters. With respect to patent activity in renewable energy overall, taxes, obligations and tradable certificates are the only statistically significant policy instruments. Interestingly however, source-specific models indicate that there is variation in the effects of instrument type on different renewables. Broadly, investment incentives are effective in supporting innovation in solar and waste-to-energy technologies, tariff structures are important for biomass, obligations and tradable certificates (which are closely related) support wind technology, and voluntary programmes induce waste-to-energy innovations. Overall, only investment and other tax incentives have wide influence on innovation for a number of renewable energy sources. While the results are interesting and robust, further work in the area could be undertaken. This includes accounting for variation in natural conditions as determinants of patenting in renewable energy technologies, and better examination of dynamic issues, with a particular focus on addressing the possible simultaneity of R&D expenditures and patenting activity. In addition, there are a number of technologies which are particularly important for electricity generation from renewable energies which could also be examined.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

161

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

For instance, advanced batteries can overcome some of the losses associated with intermittent energy supplies. Further work on assessing the effect of innovation on the costs of generation and on the share of renewables in the total energy mix is already underway.

Notes 1. Interestingly, EPO applications for patents in total increased approximately tenfold over the period in question, while patents for renewables increased just over four-fold. However, in recent years the rate of growth in the area of renewables has been higher than the rate of growth of total EPO applications. 2. The assistance of Hélène Dernis, OECD Directorate for Science, Technology and Industry in the collection of the data is gratefully acknowledged. 3. This is further complicated by inclusion of a large number of countries in the panel. 4. For further details on negative binomial models, see Cameron and Trivedi (1998). 5. Most countries introduced R&D support programmes as early as the 1970s. Since the data begins in 1978, this policy variable is (almost perfectly) correlated with the intercept (fixed effects), so it was dropped from the regression. 6. The coefficient is insignificant in the ocean equation. This is most likely due to the low ocean patent counts. 7. The negative and statistically significant coefficient of tariffs in the wind energy equation could be considered a consequence of multicollinearity. However, it remains robust, even if the tax dummy is dropped. 8. In spite of this, a positive and significant coefficient is found, even for tradable certificates. This suggests that allowing obligations to be traded provides a strong incentive for innovation. 9. The exception is the ocean equation, where it is statistically insignificant. 10. The exception is the biomass equation, where the 1-year lag is negative and significant.

References Cameron, A. C. and P. K. Trivedi (1998), Regression Analysis of Count Data, Cambridge, New York, Cambridge University Press. Commission of the European Communities (2004), The Share of Renewable Energy in the EU: Country Profiles (Brussels, COM). Commission of the European Communities (2007), “An Energy Policy for Europe” (eurlex.europa.eu/LexUriServ/site/en/com/2007/com2007_0001en01.pdf) – COM(2007)1 Final. Dernis, Hélène and Dominique Guellec (2001), “Using Patent Counts For Cross-Country Comparisons Of Technology Output”, STI mimeo, OECD, Paris (www.oecd.org/ dataoecd/26/11/21682515.pdf). Dernis, Hélène and Mosahid Kahn (2004), “Triadic Patent Families Methodology,”STI Working Paper 2004/2, OECD, Paris.

162

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

Dickson, Mary H. and Mario Fanelli (2004), What is Geothermal Energy? Instituto di Geoscienze e Georisorse, CNR, Pisa, Italy. International Energy Agency (2004), Renewable Energy – Market and Policy Trends in IEA Countries, IEA, Paris. International Energy Agency/Johannesburg Renewable Energy Coalition (2005), Renewable Energy Policies and Measures Database, OECD, Paris (www.iea.org/textbase/envissu/ pamsdb/index.html). International Energy Agency (2006a), Renewable Energy: RD&D Priorities, Insights from the IEA Technology Programmes, IEA, Paris. International Energy Agency (2006b), Renewables Information: 2006, IEA, Paris. OECD (2006a), Main Science and Technology Indicators, OECD, Paris (www.oecd.org/department/ 0,2688,en_2649_34451_1_1_1_1_1,00.html). OECD (2006b), Research and Development Statistics, OECD, Paris (www.oecd.org/department/ 0,2688,en_2649_34451_1_1_1_1_1,00.html). OECD Patent Project. Directorate for Science, Technology and Industry, OECD, Paris (www.oecd.org/document/10/0,2340,en_2649_33703_1901066_1_1_1_1,00.html). Popp, David (2005), “Using the Triadic Patent Family Database to Study Environmental Innovation”, OECD Environment Directorate Working Paper ENV/EPOC/WPNEP/ RD(2005)2.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

163

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

ANNEX 4.A1

First Page of Sample Patent Application

164

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

4.

RENEWABLE ENERGY POLICIES AND TECHNOLOGICAL INNOVATION...

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

165

ISBN 978-92-64-04681-8 Environmental Policy, Technological Innovation and Patents © OECD 2008

Chapter 5

Policy Conclusions and Further Work by Nick Johnstone (OECD Environment Directorate)

A number of conclusions have emerged from the three case studies that were undertaken as part of this study: 1) environmental policy does have an effect on technological innovation; 2) general scientific capacity is important in bringing about specifically “environmental” innovations; 3) relative prices induce innovations in which commercial and environmental benefits co-exist; 4) public concerns about the environment appeared to encourage the development of some technologies; 5) international diffusion of environmental innovation is common; and 6) there is some limited evidence of a “first-mover” advantage.

167

5.

POLICY CONCLUSIONS AND FURTHER WORK

1. Policy conclusions This report has examined the effects of environmental policy (and other factors) on eco-innovation. A number of interesting conclusions have emerged from the three case studies that were undertaken:

168

●

Environmental policy does have an effect on technological innovation. For instance, in the study on renewable energy, the implementation of different policy measures had a measurable impact on innovation, with both tax measures and quota obligations being statistically significant determinants of patent activity. However, the effect of different policies varied by the type of renewable energy involved.

●

General scientific capacity matters. Again in the case study on renewable energy innovation, the variable reflecting expenditures on targeted R&D was statistically significant in every model estimated.

●

Relative prices induce particular kinds of innovation. In the case of motor vehicle emissions abatement, fuel prices encouraged investment in “integrated” innovation (in which fuel efficiency gains also arose), but not in “post-combustion” technologies. In the case of renewable energy, the role of electricity prices was rarely significant, except for solar energy. However, as fossil fuel prices rise (and renewables become more competitive), the price substitution effect is likely to become more important.

●

Other market factors can also be important spurs to innovation. In the case of bleaching technologies in the pulping process, public concerns about the environment appeared to encourage the development of ECF and TCF technologies, pre-dating the introduction of regulatory standards. Interestingly, eco-labelling did not appear to have any influence on innovation in this case.

●

The type of innovation changes through time. In the renewables sector, different energy sources have reached maturity at different points, and there have been different “generations” of innovation within particular renewable energy sources. In the case of motor vehicle emissions abatement, there has been a shift from post-combustion technologies to integrated technologies.

●

International diffusion of environmental innovation is common. In the case of both bleaching technologies and motor vehicle emissions, abatement patent families (for some countries) were large, reflecting significant

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

technology transfer. In the case of motor vehicle emissions abatement, the transfer of Japanese technologies to the US was striking.1 ●

There is some evidence of a “first-mover” advantage. For example, in the pulp and paper sector, the early policy interventions introduced by Finland and Sweden resulted in a strong comparative advantage in TCF technologies.

While not arising directly out of the results of the empirical work reported on in this volume, a number of more general conclusions emerge from the work reviewed as part of this project and summarised in the introductory Chapter. These can be summarised as follows: ●

The general policy framework is of overarching importance. The factors which drive eco-innovation are likely to be the same as those which drive innovation in general. This includes macroeconomic factors, market regulation, and the protection of intellectual property rights.

●

Environmental policy continuity is important. Investing in R&D is risky, and it is important that the environmental policy framework not add to this risk. If markets have difficulty efficiently dealing with commercial risk associated with innovative activity, they will be even less likely to deal efficiently with political risk. As such, stability in policy regimes is important.

●

Environmental and technology policy coordination is also key. Innovation and environmental policy have different objectives. While the former is largely concerned with internalising knowledge spillovers (and thus increasing competitiveness and productivity), the latter is concerned with addressing negative environmental externalities. While no single instrument is likely to be able to address both market failures, co-ordination between the two is vital, if both the rate and the direction of innovation are to be optimal.

●

Innovation arises out of a combination of both supply and demand factors. This is no less true in the environmental sphere than it is elsewhere. As such, an efficient “environmental innovation” policy framework must address both sides of the market.

●

Policy incidence should encourage technology neutrality. Governments have limited resources at their disposal, as well as limited information about optimal technological trajectories. Moreover, with the potential for “lock in”, it is important to develop policies which minimise the downside risks of “picking losers”. In general, this means targeting environmental policies as close as possible to the environmental objective itself, not some proxy.

●

Partnerships are important, but there should be separation of responsibilities between public and private actors. This is clearest in the distinction between “basic” and “applied” research. With the active role generally played by the public sector in the environmental sphere, this issue is particularly important here.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

169

5.

POLICY CONCLUSIONS AND FURTHER WORK

And finally, it is important to emphasise that this is an area where the empirical evidence is limited, and a strong case can be made for the benefits of further research. More importantly, efforts should be devoted to ensuring that, in the process of policy development, the importance of ex post evaluation of their innovation effects is recognised and accommodated – i.e. through the incorporation of ex ante data collection requirements.

2. Further work The use of patent data to assess the extent, nature and determinants of eco-innovation is a promising area of policy research. A number of different areas could be explored, some of the most promising areas or which are summarized below.

2.1. Development of robust indicators of eco-innovation based on patent data A key building block in assessing the role played by public policy factors in the capacity of a country to benefit from environment-related innovation is the development of an appropriate indicator of eco-innovation, which is disaggregated by country, by year, and by type of environmental impact. As has been noted, patent data has many advantages for the development of such an indicator: ●

Patent classifications have been developed as detailed application-based definitions. This means it is possible to focus on areas which are of direct relevance to environmental concerns in a much more targeted manner than is possible using sectoral or commodity classifications.

●

Unlike other commonly used data (i.e.“environmental” research and development, scientific personnel, etc.) patents are an output-based measure of innovation. The data g enerated is quantitative and commensurable, allowing for the generation of indicators which are comparable across time and across countries.

●

A time series for the indicators can be developed for all countries, stretching back to 1977. In the event that users wish to modify the patent classifications used in the indicator, the data can be modified at no cost, with no inputs from government statistical agencies or other government officials.

●

Patent data can be “married” within other data sources, whether at the level of the country, sector, or even enterprise. This allows for the empirical analysis of a number of related policy-relevant questions.

While the existing work has focussed on specific areas of environmental technologies, further work would allow for a broader perspective on ecoinnovation. The objective of such work would be to identify patent classes in

170

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

the International Patent Classification scheme, which are relevant for a broad set of “environmental” technologies, including: ●

waste management and recycling;

●

wastewater treatment;

●

carbon capture and storage;

●

climate mitigation in selected sectors;

●

building energy efficiency;

●

air pollution abatement from stationary sources;

●

motor vehicle fuel efficiency;

●

environmental monitoring equipment;

●

“green” chemistry.

On the basis of this work an aggregate indicator of eco-innovation could be developed, along with thematic indicators. This would be analogous to work previously undertaken at the OECD, which has developed indices of patent activity in the areas of biotechnology and information and communications technology.2 These indicators have been widely diffused and applied. Experience in these two areas indicates that in addition to its value as a contribution to measuring progress towards sustainable development, general and thematic indicators of eco-innovation would serve as valuable inputs to a wide variety of policy-oriented empirical studies.

2.2. Assessing the economic and environmental “returns” on eco-innovation While patents are an “output” from inventive activity, unlike other measures such as investment in R&D and recruitment of scientific personnel, they are “intermediate” outputs. As noted earlier, the value of patents can vary widely. Moreover, it can be difficult to develop patent counts which accurately reflect “eco-innovation”, particularly for technologies which represent integrated changes in production processes and process design.3 Therefore, it is important to assess the environmental and economic benefits from patented inventions. For instance, in areas in which it is possible to develop robust indicators of environmental impacts (e.g. emission levels) for different environmentally-preferable technologies (and their “conventional” substitutes), the environmental benefits of patented inventions could be assessed. Areas such as motor vehicle emissions controls, “green” chemistry, and wastewater treatment could be promising areas for such research. In addition, to the environmental implications of “eco-innovation”, the economic returns could be assessed. This would involve an assessment of the links between patent activity, knowledge stocks, and economic activity. Work on

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

171

5.

POLICY CONCLUSIONS AND FURTHER WORK

the links between renewable energy patents and the market share of renewable energy is already underway. However, as noted in the Introduction, it is important to take into account potential “crowding out” of inventive activity with respect to environmental technologies on inventive activity in the economy in general. This is an area which has not been previously addressed in the literature.

2.3. Eco-innovation in a globalizing world economy The work reported on this volume, as well as that undertaken elsewhere makes it clear that eco-innovation needs to be understood in the broader context of a globalising world economy. In this vein, there are three other pieces of information which can be obtained from patent applications and which can be used to further refine our understanding of international economic factors: ●

Patent citation data gives an indication of the international diffusion of knowledge, by providing a trace of the geographical origins of relevant previous innovations.

●

Data on foreign co-inventors gives an indication of the role played by international collaboration in the invention process; and.

●

Patent family data gives an indication of the potential for technology adoption, that is, the markets in which innovators feel they are likely to be able to export their technologies.

With this information, the links between environmental innovation, and issues such as technology transfer and international competitiveness could be examined. Questions such as the benefits of international research cooperation in environmental technologies, as well as the role of first-mover advantage in environmental policy implementation could be assessed. The relative “openness” of a country to overseas knowledge, research, and market opportunities can also be affected by the nature of the environmental policy instruments that are implemented. Conversely, some types of environmental regulations may “fragment” markets for innovation, resulting in significant inefficiency in technological development. Patent data is well-suited to the assessment of these important policy questions.

Notes 1. Danish exports of wind power technologies to the US are also important, but patent family data was not collected for this case. 2. www.oecd.org/dataoecd/54/57/33882780.pdf. 3. For a discussion of some of these issues see J. Labonne and N. Johnstone (2007) “Environmental Policy and Economies of Scope in Facility-Level Environmental Practices” (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1035661).

172

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

ANNEX 5.A1

Glossary of Relevant Patent and Related Terms Adoption: The point at which a technology is selected for use by an individual or an organisation. Applicant: The person or company that applies for the patent and intends to “work” the invention (i.e. to manufacture or licence the technology). In most countries the inventor(s) does not necessarily have to be the applicant. Application (or filing) date: The patent application date is the date on which the patent office received the patent application. Application for a patent: To obtain a patent, an application must be filed with the authorised body (Patent Office) with all the necessary documents and fees. The patent office will conduct an examination to decide whether to grant or reject the application. Assignee: The person(s) or corporate body to whom all or limited rights under a patent are legally transferred. Assignment Transfer of all or limited rights under a patent. Bibliometrics: Study of the quantitative data of the publication patterns of individual articles, journals, and books in order to analyze trends and make comparisons within a body of literature. Breadth (or scope): A measure of the extent of the invention covered by a single patent application. For example, one patent application to the EPO would generally more claims than an application to the JPO. Citations: They comprise a list of references that are believed to be relevant prior art and which may have contributed to the “narrowing” of the original application. Citations may be made by the examiner or the applicant/ inventor.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

173

5.

POLICY CONCLUSIONS AND FURTHER WORK

Claim(s): These define the invention that the applicant wishes to protect. A main claim will define the invention in its broadest form, by including its essential technical features. Further “dependant” claims can then relate to additional features of the invention. Copyright: The legal right granted to an author, editor or publisher of an article, chapter or complete work. Copyright applies to intellectual property in a variety of artistic fields and attempts to be format-neutral. Designated countries: Countries in which patent applicants wish to protect their invention. This concept is specific to European patent applications and international patent applications filed under the Patent Cooperation Treaty (PCT). Diffusion: The extent to which a technology spreads to general use and application in the economy. Duplicate: All patents relating to the same invention and sharing the same priority, but filed at patent offices other than the priority office. The count of such patents can be considered as the size of a “simple” patent family. ECLA: The European Patent Office’s patent classification system. It is based on the IPC Classification System, with greater disaggregation. Equivalent: A patent that relates to the same invention and shares the same priority application as a patent from a different issuing authority. Espacenet: European Patent Office web site for searching, displaying and downloading patent documents. European Patent Convention (EPC): The Convention on the Grant of European Patents (European Patent Convention, EPC) was signed in Munich 1973 and entered into force in 1977. As a result of the EPC, the European Patent Office (EPO) was created to grant European patents. European Patent Office (EPO): The European Patent Office (a regional patents office) was created by the EPC to grant European patents, based on a centralised examination procedure. By filing a single European patent application in one of the three official languages (English, French and German), it is possible to obtain patent rights in all the EPC member and extension countries by designating the countries in the EPO application. The EPO is not an institution of the European Union. European patent: A European patent can be obtained for all the EPC countries by filing a single application at the EPO in one of the three official languages (English, French or German). European patents granted by the EPO have the same legal rights and are subject to the same conditions as national patents (granted by the national patent office). It is important to note that a granted European patent is a “bundle” of national patents, which must be

174

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

validated at the national patent office for it to be effective in member countries. Examiner: An employee of a patent office to whom an application is assigned for handling prosecution. Grant date: The date when the patent office issues a patent to the applicant. On average it takes three years for a patent to be granted at the USPTO and five years at the EPO. Grant: A temporary right given by the authorised body for a limited time period (normally 20 years) to prevent unauthorised use of the technology outlined in the patent. A patent application does not automatically give the applicant a temporary right against infringement. A patent has to be granted for it to be effective and enforceable against infringement. Home Bias: Propensity for the priority country to be the same as the inventor or applicant country. Infringement: Unauthorised use of a patented invention. Innovation: The creation or introduction of something new, especially a new product or a new way of producing something. Intellectual property rights (IPR): IPR allows people to assert ownership rights on the outcomes of their creativity and innovative activity in the same way that they can own physical property. The four main types of intellectual property rights are: patents, trademarks, design and copyrights. International patent application: Patent applications filed under the Patent Cooperation Treaty (PCT) are commonly referred to as international patent applications. However, an international patent (PCT) application does not result in the issuance of “international patents”, i.e. at present, there is no global patent system that is responsible for granting international patents. The decision of whether to grant or reject a patent application filed under the PCT rests with the national or regional (e.g. EPO) patent offices. International Patent Classification (IPC): The International Patent Classification, which is commonly referred to as the IPC, is based on an international multilateral treaty administered by WIPO. The IPC is an internationally recognised patent classification system, which provides a common classification for patents according to technology groups. IPC is periodically revised in order to improve the system and to take account of technical development. The current (eighth) edition of the IPC entered into force on 1 January 2006. Inventor country: Country of the residence of the inventor, which is frequently used to count patents in order to measure inventive performance. Inventor: Inventor names are recorded for all patents. These appear in the standard last name-initial(s) format.

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

175

5.

POLICY CONCLUSIONS AND FURTHER WORK

Japan Patent Office (JPO): The JPO administers the examination and granting of patent rights in Japan. The JPO is an agency of the Ministry of Economy, Trade and Industry (METI). Kind Code: The letter, often with a further number, indicating the level of publication of a patent. For example DE-A1 is the German Offenlegungsschrift (application laid open for public inspection) while a DE-C1 is the German Patentschrift (first publication of the granted patent). Lapse: The date when a patent is no longer valid in a country or system due to failure to pay renewal (maintenance) fees. Often the patent can be reinstated within a limited period. Learning by doing: Refers to the improvement in technology that takes place in some industries, early in their history, as they learn by experience, so that average cost falls as accumulated output rises. See infant industry protection, dynamic economies of scale. Learning curve: Relationship representing either average cost or average product as a function of the accumulated output produced. Usually reflecting learning by doing, the learning curve shows cost falling, or average product rising. Licence: The means by which the owner of a patent gives permission to another person to carry out an action which, without such permission, would infringe on the patent. A licence can thus allow another person to legitimately manufacture, use or sell an invention protected by a patent. In return, the patent owner will usually receive royalty payments. A license, which can be exclusive or non-exclusive, does not transfer the ownership of the invention to the licensee. Novelty: If an application for a patent is to be successful, the invention must be novel (new). The invention must never have been made public in any way, anywhere, before the date on which the application for a patent is filed (or before the priority date). Obviousness: The concept that the claims defining an invention in a patent application must involve an inventive step if, when compared with what is already known (i.e.prior art), it would not be obvious to someone skilled in the art. OECD triadic patent families: The triadic patent families are defined at the OECD as a set of patents taken at the European Patent Office (EPO), the Japan Patent Office (JPO) and the US Patent and Trademark Office (USPTO) that share one or more priorities. Triadic patent families data are consolidated to eliminate double counting of patents filed at different offices (i.e. regrouping all the interrelated priorities in EPO, JPO and USPTO patent documents).

176

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

Paris Convention: The Paris Convention for the Protection of Industrial Property was established in 1883 and is generally referred to the Paris Convention. The Paris Convention established the system of priority rights. Under the priority rights, applicants have up to 12 months from first filing their patent application (usually in their own country) in which to make further applications in member countries and claim the original priority date. Patent Cooperation Treaty (PCT): Signed in 1970, the PCT entered into force in 1978. The PCT provides the possibility to seek patent rights in a large number of countries by filing a single international application (PCT application) with a single patent office (receiving office). The PCT procedure consists of two main phases: a) an “international phase”; and b) a PCT “national/regional phase”. PCT applications are administered by the World Intellectual Property Organisation (WIPO). Patent family: A patent family is a set of individual patents granted by various countries. The patent family is all the equivalent patent applications corresponding to a single invention, covering different geographical regions. Patent family size is a measure of the geographical breadth for which protection of the invention is sought. Patent number: A patent number is a unique identifier of a patent. Patent numbers are assigned to each patent document by the patent-issuing authority. The first two letters designate the issuing patent office i.e. EP for EPO patents and US for USPTO patents. Patent: A patent is an intellectual property right issued by authorized bodies to inventors to make use of, and exploit their inventions for a limited period of time (generally 20 years). The patent holder has the legal authority to exclude others from commercially exploiting the invention (for a limited time period). In return for the ownership rights, the applicant must disclose the invention for which protection is sought. The trade-off between the granting of monopoly rights for a limited period and full disclosure of information is an important aspect of the patenting system. Patentability: Patentability is the ability of an invention to satisfy the legal requirements for obtaining a patent. The basic conditions of patentability, which an application must meet before a patent is granted, are that the invention must be novel, contain an inventive step (or be nonobvious), be capable of industrial application and not be in certain excluded fields (e.g. scientific theories and mathematical methods are not regarded as inventions and cannot be patented at the EPO). PATSTAT: The EPO’s World Patent Statistical Database. Prior Art: Previously used or published technology that may be referred to in a patent application or examination report. a) In a broad sense, technology that is relevant to an invention and was publicly available (e.g. described in a

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

177

5.

POLICY CONCLUSIONS AND FURTHER WORK

publication or offered for sale) at the time an invention was made. b) In a narrow sense, any such technology which would invalidate a patent or limit its scope. The process of prosecuting a patent or interpreting its claims largely consists of identifying relevant prior art and distinguishing the claimed invention from that prior art. Priority country: Country where the patent is first filed before being (possibly) extended to other countries. Priority date: The priority date is the first date of filing of a patent application, anywhere in the world (normally in the applicant’s domestic patent office), to protect an invention. The priority date is used to determine the novelty of the invention, which implies that it is an important concept in patent procedures. For statistical purposes, the priority date is the closest date to the date of invention. Publication lag: In most countries, a patent application is published 18 months after the priority date. For example, all pending EPO and JPO patent applications are published 18 months after the priority date. Prior to a change in rules under the American Inventors Protection Act of 1999, USPTO patent applications were held in confidence until a patent was granted. Patent applications filed at the USPTO on or after 29 November 2000 are required to be published 18 months after the priority date. R&D expenditures: The basic measure of R&D expenditures is “intramural expenditures”; i.e. all expenditures for R&D performed within a statistical unit or sector of the economy. R&D: Research and experimental development (R&D) comprises creative work undertaken on a systematic basis in order to increase the stock of knowledge, including knowledge of man, culture and society, and the use of this stock of knowledge to devise new applications. Renewal fees: Once a patent is granted, annual renewal fees are payable to patent offices to keep the patent in force. In the USPTO these payments are referred to as maintenance fees. Rent: The premium that the owner of a resource receives over and above its opportunity cost. Reverse engineering: The process of learning how a product is made by taking it apart and examining it. Revocation: Termination of the protection given to a patent on one or more grounds, e.g. lack of novelty. Scientometrics: The quantitative study of the disciplines of science based on published literature and communications. This could include identifying emerging areas of scientific research, examining the development of research over time, or geographic and organisational distributions of research.

178

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

5.

POLICY CONCLUSIONS AND FURTHER WORK

Search report: The search report is a list of citations of all published prior art documents which are relevant to the patent application. The search process, conducted by a patent examiner, seeks to identify patent and nonpatent documents constituting the relevant prior art to be taken into account in determining whether the invention is novel and includes an inventive step. Technology Transfer: The communication or transmission of a technology from one country to another. This may be accomplished in a variety of ways, ranging from deliberate licensing to reverse engineering. Term of patent: The maximum number of years that the monopoly rights conferred by the grant of a patent may last. Trade-Related Aspects of Intellectual Property Rights (TRIPS): Agreement on trade-related aspects of intellectual property rights requires members to comply with certain minimum standards for the protection of IPR. But members may choose to implement laws which provide more extensive protection than is required in the agreement, so long as the additional protection does not contravene the provisions of the agreement. The WTO’s TRIPS agreement, negotiated in the 1986-94 Uruguay round, introduced intellectual property rules into the multilateral trading system for the first time. United States Patent and Trademark Office (USPTO): The USPTO administers the examination and granting of patent rights in the United States. It falls under the jurisdiction of the US Department of Commerce. Utility model: Also known as “petty patent”, these are available in some countries (e.g. Japan). This type of patent involves a simpler inventive step than that in a traditional patent and it is valid for a shorter time period. World Intellectual Property Organization (WIPO): An intergovernmental organisation responsible for the negotiation and administration of various multilateral treaties dealing with the legal and administrative aspects of intellectual property. In the patent area, the WIPO is notably in charge of administering the Patent Cooperation Treaty (PCT) and the International Patent Classification system (IPC).

Primary Sources: OECD (2006), STI/EAS Division Glossary of Patent Terminology (2006) (www.oecd.org/dataoecd/5/39/37569498.pdf). Deardoff’s Glossary of International Economic Terms (www-personal.umich.edu/~alandear/glossary/). OECD (2006), Economics Glossary: English-French (2006). Thomson Scientific Glossary of Thomson Scientific Terminology (http://scientific.thomson.com/support/patents/patinf/terms/).

ENVIRONMENTAL POLICY, TECHNOLOGICAL INNOVATION AND PATENTS – ISBN 978-92-64-04681-8 – © OECD 2008

179

OECD PUBLICATIONS, 2, rue André-Pascal, 75775 PARIS CEDEX 16 PRINTED IN FRANCE (97 2008 06 1 P) ISBN 978-92-64-04681-8 – No. 56339 2008

Environmental Policy, Technological Innovation and Patents Technological innovation can help realise environmental objectives in a less costly manner than would otherwise be the case. Thus, understanding the role that technological innovation can play in achieving environmental objectives is important for policy debates. However, the relationship between environmental policy and technological innovation remains an area in which empirical evidence is scant. In an attempt to bridge this gap, the OECD has examined the relevant issues, using patent activity as a measure of technological innovation.

The full text of this book is available on line via these links: www.sourceoecd.org/environment/9789264046818 www.sourceoecd.org/scienceIT/9789264046818 Those with access to all OECD books on line should use this link: www.sourceoecd.org/9789264046818 SourceOECD is the OECD online library of books, periodicals and statistical databases. For more information about this award-winning service and free trials, ask your librarian, or write to us at [email protected].

ISBN 978-92-64-04681-8 97 2008 06 1 P

XXXPFDEPSHQVCMJTIJOH

-:HSTCQE=UY[]V]:

Environmental Policy, Technological Innovation and Patents

Three case studies have been undertaken: abatement technologies for wastewater effluent from pulp production, abatement of motor vehicle emissions, and development of renewable energy technologies. On the basis of patent data, the nature, extent, and causes of innovation in each of these areas have been explored. While a particular focus has been placed on the role of environmental policy in bringing about the innovation documented, it is recognised that other factors play a key role in inducing innovation that has positive environmental implications.

OECD Studies on Environmental Innovation

OECD Studies on Environmental Innovation

OECD Studies on Environmental Innovation

Environmental Policy, Technological Innovation and Patents