207
3
Social, Economic, and Ethical
Concepts and Methods
Coordinating Lead Authors:
Charles Kolstad (USA), Kevin Urama (Nigeria / UK / Kenya)
Lead Authors:
John Broome (UK), Annegrete Bruvoll (Norway), Micheline Cariño Olvera (Mexico), Don
Fullerton (USA), Christian Gollier (France), William Michael Hanemann (USA), Rashid Hassan
(Sudan / South Africa), Frank Jotzo (Germany / Australia), Mizan R. Khan (Bangladesh), Lukas Meyer
(Germany / Austria), Luis Mundaca (Chile / Sweden)
Contributing Authors:
Philippe Aghion (USA), Hunt Allcott (USA), Gregor Betz (Germany), Severin Borenstein (USA),
Andrew Brennan (Australia), Simon Caney (UK), Dan Farber (USA), Adam Jaffe (USA / New
Zealand), Gunnar Luderer (Germany), Axel Ockenfels (Germany), David Popp (USA)
Review Editors:
Marlene Attzs (Trinidad and Tobago), Daniel Bouille (Argentina), Snorre Kverndokk (Norway)
Chapter Science Assistants:
Sheena Katai (USA), Katy Maher (USA), Lindsey Sarquilla (USA)
This chapter should be cited as:
Kolstad C., K. Urama, J. Broome, A. Bruvoll, M. Cariño Olvera, D. Fullerton, C. Gollier, W. M. Hanemann, R. Hassan, F. Jotzo,
M. R. Khan, L. Meyer, and L. Mundaca, 2014: Social, Economic and Ethical Concepts and Methods. In: Climate Change
2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovern-
mental Panel on Climate Change [Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler,
I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel and J.C. Minx (eds.)].
Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
208208
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
Contents
Executive Summary � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 211
3�1 Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 213
3�2 Ethical and socio-economic concepts and principles � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 214
3�3 Justice, equity and responsibility � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 215
3�3�1 Causal and moral responsibility
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 215
3�3�2 Intergenerational justice and rights of future people
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 216
3�3�3 Intergenerational justice: distributive justice
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 216
3�3�4 Historical responsibility and distributive justice
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 217
3�3�5 Intra-generational justice: compensatory justice and historical responsibility
� � � � � � � � � � � � � � � � � � � � � � � � � � � � 217
3�3�6 Legal concepts of historical responsibility
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 218
3�3�7 Geoengineering, ethics, and justice
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 219
3�4 Values and wellbeing � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 220
3�4�1 Non-human values
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 220
3�4�2 Cultural and social values
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 221
3�4�3 Wellbeing
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 221
3�4�4 Aggregation of wellbeing
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 221
3�4�5 Lifetime wellbeing
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 222
3�4�6 Social welfare functions
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 222
3�4�7 Valuing population
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 223
3�5 Economics, rights, and duties � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 223
3�5�1 Limits of economics in guiding decision making
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 224
209209
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
3�6 Aggregation of costs and benefits � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 225
3�6�1 Aggregating individual wellbeing
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 225
3.6.1.1 Monetary values
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
3�6�2 Aggregating costs and benefits across time
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 228
3�6�3 Co-benefits and adverse side-effects
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 232
3.6.3.1 A general framework for evaluation of co-benefits and adverse side-effects
. . . . . . . . . . . . . . . . . . . . . . . . 232
3.6.3.2 The valuation of co-benefits and adverse side-effects
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
3.6.3.3 The double dividend hypothesis
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
3�7 Assessing methods of policy choice � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 235
3�7�1 Policy objectives and evaluation criteria
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 235
3.7.1.1 Economic objectives
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3.7.1.2 Distributional objectives
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3.7.1.3 Environmental objectives
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3.7.1.4 Institutional and political feasibility
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
3�7�2 Analytical methods for decision support
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 238
3.7.2.1 Quantitative-oriented approaches
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
3.7.2.2 Qualitative approaches
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
3�8 Policy instruments and regulations � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 239
3�8�1 Economic incentives
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 239
3.8.1.1 Emissions taxes and permit trading
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
3.8.1.2 Subsidies
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
3�8�2 Direct regulatory approaches
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 240
3�8�3 Information programmes
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 241
3�8�4 Government provision of public goods and services, and procurement
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 241
3�8�5 Voluntary actions
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 241
3�8�6 Policy interactions and complementarity
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 241
3�8�7 Government failure and policy failure
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 241
3.8.7.1 Rent-seeking
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
3.8.7.2 Policy uncertainty
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
3�9 Metrics of costs and benefits � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 242
3�9�1 The damages from climate change
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 243
3�9�2 Aggregate climate damages
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 245
210210
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
3�9�3 The aggregate costs of mitigation � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 247
3�9�4 Social cost of carbon
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 249
3�9�5 The rebound effect
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 249
3�9�6 Greenhouse gas emissions metrics
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 250
3�10 Behavioural economics and culture � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 252
3�10�1 Behavioural economics and the cost of emissions reduction
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 252
3.10.1.1 Consumer undervaluation of energy costs
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
3.10.1.2 Firm behaviour
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
3.10.1.3 Non-price interventions to induce behavioural change
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
3.10.1.4 Altruistic reductions of carbon emissions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
3.10.1.5 Human ability to understand climate change
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
3�10�2 Social and cultural issues
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 254
3.10.2.1 Customs
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
3.10.2.2 Indigenous peoples
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
3.10.2.3 Women and climate change
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
3.10.2.4 Social institutions for collective action
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
3�11 Technological change � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 256
3�11�1 Market provision of TC
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 256
3�11�2 Induced innovation
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 256
3�11�3 Learning-by-doing and other structural models of TC
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 257
3�11�4 Endogenous and exogenous TC and growth
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 257
3�11�5 Policy measures for inducing R&D
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 257
3�11�6 Technology transfer (TT)
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 257
3�12 Gaps in knowledge and data � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 258
3�13 Frequently Asked Questions � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 259
References � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 260
211211
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
Executive Summary
This framing chapter describes the strengths and limitations of the
most widely used concepts and methods in economics, ethics, and
other social sciences that are relevant to climate change. It also pro-
vides a reference resource for the other chapters in the Intergovern-
mental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5),
as well as for decision makers.
The significance of the social dimension and the role of ethics and
economics is underscored by Article 2 of the United Nations Frame-
work Convention on Climate Change, which indicates that an ultimate
objective of the Convention is to avoid dangerous anthropogenic inter-
ference with the climate system. Two main issues confronting society
(and the IPCC) are: what constitutes ‘dangerous interference’ with the
climate system and how to deal with that interference. Determining
what is dangerous is not a matter for natural science alone; it also
involves value judgements a subject matter of the theory of value,
which is treated in several disciplines, including ethics, economics, and
other social sciences.
Ethics involves questions of justice and value. Justice is concerned with
equity and fairness, and, in general, with the rights to which people
are entitled. Value is a matter of worth, benefit, or good. Value can
sometimes be measured quantitatively, for instance, through a social
welfare function or an index of human development.
Economic tools and methods can be used in assessing the positive
and negative values that result from particular decisions, policies, and
measures. They can also be essential in determining the mitigation
and adaptation actions to be undertaken as public policy, as well as
the consequences of different mitigation and adaptation strategies.
Economic tools and methods have strengths and limitations, both of
which are detailed in this chapter.
Economic tools can be useful in designing climate change miti-
gation policies (very high confidence). While the limitations of eco-
nomics and social welfare analysis, including cost-benefit analysis, are
widely documented, economics nevertheless provides useful tools for
assessing the pros and cons of taking, or not taking, action on climate
change mitigation, as well as of adaptation measures, in achieving
competing societal goals. Understanding these pros and cons can help
in making policy decisions on climate change mitigation and can influ-
ence the actions taken by countries, institutions and individuals. [Sec-
tion 3.2]
Mitigation is a public good; climate change is a case of ‘the
tragedy of the commons (high confidence). Effective climate change
mitigation will not be achieved if each agent (individual, institution or
country) acts independently in its own selfish interest, suggesting the
need for collective action. Some adaptation actions, on the other hand,
have characteristics of a private good as benefits of actions may accrue
more directly to the individuals, regions, or countries that undertake
them, at least in the short term. Nevertheless, financing such adaptive
activities remains an issue, particularly for poor individuals and coun-
tries. [3.1, 3.2]
Analysis contained in the literature of moral and political phi-
losophy can contribute to resolving ethical questions that are
raised by climate change (medium confidence). These questions
include how much overall climate mitigation is needed to avoid ‘dan-
gerous interference’, how the effort or cost of mitigating climate
change should be shared among countries and between the present
and future, how to account for such factors as historical responsibility
for emissions, and how to choose among alternative policies for miti-
gation and adaptation. Ethical issues of wellbeing, justice, fairness, and
rights are all involved. [3.2, 3.3, 3.4]
Duties to pay for some climate damages can be grounded in
compensatory justice and distributive justice (medium confi-
dence). If compensatory duties to pay for climate damages and adap-
tation costs are not due from agents who have acted blamelessly,
then principles of compensatory justice will apply to only some of
the harmful emissions [3.3.5]. This finding is also reflected in the pre-
dominant global legal practice of attributing liability for harmful emis-
sions [3.3.6]. Duties to pay for climate damages can, however, also be
grounded in distributive justice [3.3.4, 3.3.5].
Distributional weights may be advisable in cost-benefit analysis
(medium confidence). Ethical theories of value commonly imply that
distributional weights should be applied to monetary measures of ben-
efits and harms when they are aggregated to derive ethical conclu-
sions [3.6.1]. Such weighting contrasts with much of the practice of
cost-benefit analysis.
The use of a temporal discount rate has a crucial impact on the
evaluation of mitigation policies and measures The social dis-
count rate is the minimum rate of expected social return that com-
pensates for the increased intergenerational inequalities and the
potential increased collective risk that an action generates. Even with
disagreement on the level of the discount rate, a consensus favours
using declining risk-free discount rates over longer time horizons (high
confidence). [3.6.2]
An appropriate social risk-free discount rate for consumption
is between one and three times the anticipated growth rate in
real per capita consumption (medium confidence). This judgement
is based on an application of the Ramsey rule using typical values in
the literature of normative parameters in the rule. Ultimately, however,
these are normative choices. [3.6.2]
Co-benefits may complement the direct benefits of mitigation
(medium confidence). While some direct benefits of mitigation are
reductions in adverse climate change impacts, co-benefits can include
a broad range of environmental, economic, and social effects, such as
212212
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
reductions in local air pollution, less acid rain, and increased energy
security. However, whether co-benefits are net positive or negative in
terms of wellbeing (welfare) can be difficult to determine because of
interaction between climate policies and pre-existing non-climate poli-
cies. The same results apply to adverse side-effects. [3.6.3]
Tax distortions change the cost of all abatement policies (high
confidence). A carbon tax or a tradable emissions permit system can
exacerbate tax distortions, or, in some cases, alleviate them; carbon tax
or permit revenue can be used to moderate adverse effects by cutting
other taxes. However, regulations that forgo revenue (e. g., by giving
permits away) implicitly have higher social costs because of the tax
interaction effect. [3.6.3]
Many different analytic methods are available for evaluating
policies Methods may be quantitative (for example, cost-benefit
analysis, integrated assessment modelling, and multi-criteria analysis)
or qualitative (for example, sociological and participatory approaches).
However, no single-best method can provide a comprehensive analysis
of policies. A mix of methods is often needed to understand the broad
effects, attributes, trade-offs, and complexities of policy choices; more-
over, policies often address multiple objectives. [3.7]
Four main criteria are frequently used in evaluating and choos-
ing a mitigation policy (medium confidence). They are: cost-effec-
tiveness and economic efficiency (excluding environmental benefits,
but including transaction costs); environmental effectiveness (the
extent to which the environmental targets are achieved); distributional
effects (impact on different subgroups within society); and institutional
feasibility, including political feasibility. [3.7.1]
A broad range of policy instruments for climate change miti-
gation is available to policymakers These include: economic
incentives, direct regulatory approaches, information programmes,
government provision, and voluntary actions. Interactions between
policy instruments can enhance or reduce the effectiveness and cost
of mitigation action. Economic incentives will generally be more
cost-effective than direct regulatory interventions. However, the
performance and suitability of policies depends on numerous con-
ditions, including institutional capacity, the influence of rent-seek-
ing, and predictability or uncertainty about future policy settings.
The enabling environment may differ between countries, including
between low-income and high-income countries. These differences
can have implications for the suitability and performance of policy
instruments. [3.8]
Impacts of extreme events may be more important economi-
cally than impacts of average climate change (high confidence).
Risks associated with the entire probability distribution of outcomes
in terms of climate response [WGI] and climate impacts [WGII] are
relevant to the assessment of mitigation. Impacts from more extreme
climate change may be more important economically (in terms of the
expected value of impacts) than impacts of average climate change,
particularly if the damage from extreme climate change increases more
rapidly than the probability of such change declines. This is important
in economic analysis, where the expected benefit of mitigation may be
traded off against mitigation costs. [3.9.2]
Impacts from climate change are both market and non-market�
Market effects (where market prices and quantities are observed)
include impacts of storm damage on infrastructure, tourism, and
increased energy demand. Non-market effects include many ecological
impacts, as well as changed cultural values, none of which are gen-
erally captured through market prices. The economic measure of the
value of either kind of impact is ‘willingness-to-pay’ to avoid damage,
which can be estimated using methods of revealed preference and
stated preference. [3.9]
Substitutability reduces the size of damages from climate
change (high confidence). The monetary damage from a change in the
climate will be lower if individuals can easily substitute for what is
damaged, compared to cases where such substitution is more difficult.
[3.9]
Damage functions in existing Integrated Assessment Models
(IAMs) are of low reliability (high confidence). The economic assess-
ments of damages from climate change as embodied in the damage
functions used by some existing IAMs (though not in the analysis
embodied in WGIII) are highly stylized with a weak empirical foun-
dation. The empirical literature on monetized impacts is growing but
remains limited and often geographically narrow. This suggests that
such damage functions should be used with caution and that there
may be significant value in undertaking research to improve the preci-
sion of damage estimates. [3.9, 3.12]
Negative private costs of mitigation arise in some cases,
although they are sometimes overstated in the literature
(medium confidence). Sometimes mitigation can lower the private
costs of production and thus raise profits; for individuals, mitigation
can raise wellbeing. Ex-post evidence suggests that such ‘negative cost
opportunities’ do indeed exist but are sometimes overstated in engi-
neering analyses. [3.9]
Exchange rates between GHGs with different atmospheric life-
times are very sensitive to the choice of emission metric The
choice of an emission metric depends on the potential application and
involves explicit or implicit value judgements; no consensus surrounds
the question of which metric is both conceptually best and practical to
implement (high confidence). In terms of aggregate mitigation costs
alone, the Global Warming Potential (GWP), with a 100-year time hori-
zon, may perform similarly to selected other metrics (such as the time-
dependent Global Temperature Change Potential or the Global Cost
Potential) of reaching a prescribed climate target; however, various
metrics may differ significantly in terms of the implied distribution of
costs across sectors, regions, and over time (limited evidence, medium
agreement). [3.9]
213213
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
The behaviour of energy users and producers exhibits a variety
of anomalies (high confidence). Understanding climate change as a
physical phenomenon with links to societal causes and impacts is a
very complex process. To be fully effective, the conceptual frameworks
and methodological tools used in mitigation assessments need to take
into account cognitive limitations and other-regarding preferences that
frame the processes of economic decision making by people and firms.
[3.10]
Perceived fairness can facilitate cooperation among individu-
als (high confidence). Experimental evidence suggests that reciprocal
behaviour and perceptions of fair outcomes and procedures facilitate
voluntary cooperation among individual people in providing public
goods; this finding may have implications for the design of interna-
tional agreements to coordinate climate change mitigation. [3.10]
Social institutions and culture can facilitate mitigation and
adaptation (medium confidence). Social institutions and culture can
shape individual actions on mitigation and adaptation and be comple-
mentary to more conventional methods for inducing mitigation and
adaptation. They can promote trust and reciprocity and contribute to
the evolution of common rules. They also provide structures for acting
collectively to deal with common challenges. [3.10]
Technological change that reduces mitigation costs can be
encouraged by institutions and economic incentives (high con-
fidence). As pollution is not fully priced by the market, private indi-
viduals and firms lack incentives to invest sufficiently in the develop-
ment and use of emissions-reducing technologies in the absence of
appropriate policy interventions. Moreover, imperfect appropriability of
the benefits of innovation further reduces incentives to develop new
technologies. [3.11]
3.1 Introduction
This framing chapter has two primary purposes: to provide a frame-
work for viewing and understanding the human (social) perspective on
climate change, focusing on ethics and economics; and to define and
discuss key concepts used in other chapters. It complements the two
other framing chapters: Chapter 2 on risk and uncertainty and Chapter
4 on sustainability. The audience for this chapter (indeed for this entire
volume) is decision makers at many different levels.
The significance of the social dimension and the role of ethics and eco-
nomics is underscored by Article 2 of the United Nations Framework
Convention on Climate Change (UNFCCC), which indicates that the
ultimate objective of the Convention is to avoid dangerous anthropo-
genic interference with the climate system. Two main issues confront-
ing society are: what constitutes ‘dangerous interference’ with the
climate system and how to deal with that interference (see box 3.1).
Providing information to answer these inter-related questions is a pri-
mary purpose of the IPCC. Although natural science helps us under-
stand how emissions can change the climate, and, in turn, generate
physical impacts on ecosystems, people, and the physical environment,
determining what is dangerous involves judging the level of adverse
consequences, the steps necessary to mitigate these consequences,
and the risk that humanity is willing to tolerate. These are questions
requiring value judgement. Although economics is essential to evaluat-
ing the consequences and trade-offs associating with climate change,
how society interprets and values them is an ethical question.
Our discussion of ethics centres on two main considerations: justice
and value. Justice requires that people and nations should receive
what they are due, or have a right to. For some, an outcome is just
if the process that generated it is just. Others view justice in terms
of the actual outcomes enjoyed by different people and groups and
the values they place on those outcomes. Outcome-based justice can
range from maximizing economic measures of aggregate welfare to
rights-based views of justice, for example, believing that all countries
have a right to clean air. Different views have been expressed about
what is valuable. All values may be anthropocentric or there may be
non-human values. Economic analysis can help to guide policy action,
provided that appropriate, adequate, and transparent ethical assump-
tions are built into the economic methods.
The significance of economics in tackling climate change is widely rec-
ognized. For instance, central to the politics of taking action on climate
change are disagreements over how much mitigation the world should
undertake, and the economic costs of action (the costs of mitigation)
and inaction (the costs of adaptation and residual damage from a
changed climate). Uncertainty remains about (1) the costs of reducing
emissions of greenhouse gases (GHGs), (2) the damage caused by a
change in the climate, and (3) the cost, practicality, and effectiveness
of adaptation measures (and, potentially, geoengineering). Prioritiz-
ing action on climate change over other significant social goals with
more near-term payoffs is particularly difficult in developing countries.
Because social concerns and objectives, such as the preservation of
traditional values, cannot always be easily quantified or monetized,
economic costs and benefits are not the only input into decision mak-
ing about climate change. But even where costs and benefits can be
quantified and monetized, using methods of economic analysis to
steer social action implicitly involves significant ethical assumptions.
This chapter explains the ethical assumptions that must be made for
economic methods, including cost-benefit analysis (CBA), to be valid,
as well as the ethical assumptions that are implicitly being made
where economic analysis is used to inform a policy choice.
The perspective of economics can improve our understanding of the
challenges of acting on mitigation. For an individual or firm, mitigation
involves real costs, while the benefits to themselves of their own miti-
gation efforts are small and intangible. This reduces the incentives for
individuals or countries to unilaterally reduce emissions; free-riding on
the actions of others is a dominant strategy. Mitigating greenhouse
214214
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
gas (GHG) emissions is a public good, which inhibits mitigation. This
also partly explains the failure of nations to agree on how to solve the
problem.
In contrast, adaptation tends not to suffer from free-riding. Gains to
climate change from adaptation, such as planting more heat tolerant
crops, are mainly realized by the parties who incur the costs. Associated
externalities tend to be more localized and contemporaneous than for
GHG mitigation. From a public goods perspective, global coordination
may be less important for many forms of adaptation than for mitiga-
tion. For autonomous adaptation in particular, the gains from adapta-
tion accrue to the party incurring the cost. However, public adaptation
requires local or regional coordination. Financial and other constraints
may restrict the pursuit of attractive adaptation opportunities, particu-
larly in developing countries and for poorer individuals.
This chapter addresses two questions: what should be done about
action to mitigate climate change (a normative issue) and how the
world works in the multifaceted context of climate change (a descrip-
tive or positive issue). Typically, ethics deals with normative questions
and economics with descriptive or normative questions. Descriptive
questions are primarily value-neutral, for example, how firms have
reacted to cap-and-trade programmes to limit emissions, or how soci-
eties have dealt with responsibility for actions that were not known to
be harmful when they were taken. Normative questions use economics
and ethics to decide what should be done, for example, determining
the appropriate level of burden sharing among countries for current
and future mitigation. In making decisions about issues with norma-
tive dimensions, it is important to understand the implicit assumptions
involved. Most normative analyses of solutions to the climate problem
implicitly involve contestable ethical assumptions.
This chapter does not attempt to answer ethical questions, but rather
provides policymakers with the tools (concepts, principles, arguments,
and methods) to make decisions. Summarizing the role of economics
and ethics in climate change in a single chapter necessitates several
caveats. While recognizing the importance of certain non-economic
social dimensions of the climate change problem and solutions to it,
space limitations and our mandate necessitated focusing primarily on
ethics and economics. Furthermore, many of the issues raised have
already been addressed in previous IPCC assessments, particularly AR2
(published in 1995). In the past, ethics has received less attention than
economics, although aspects of both subjects are covered in AR2. The
literature reviewed here includes pre-AR4 literature in order to pro-
vide a more comprehensive understanding of the concepts and meth-
ods. We highlight ‘new’ developments in the field since the last IPCC
assessment in 2007.
3.2 Ethical and socio-economic
concepts and principles
When a country emits GHGs, its emissions cause harm around the
globe. The country itself suffers only a part of the harm it causes. It is
therefore rarely in the interests of a single country to reduce its own
emissions, even though a reduction in global emissions could benefit
every country. That is to say, the problem of climate change is a “trag-
edy of the commons” (Hardin, 1968). Effective mitigation of climate
change will not be achieved if each person or country acts indepen-
dently in its own interest.
Consequently, efforts are continuing to reach effective international
agreement on mitigation. They raise an ethical question that is widely
recognized and much debated, namely, ‘burden-sharing’ or ‘effort-
sharing’. How should the burden of mitigating climate change be
divided among countries? It raises difficult issues of justice, fairness,
and rights, all of which lie within the sphere of ethics.
Burden-sharing is only one of the ethical questions that climate change
raises.
1
Another is the question of how much overall mitigation should
1
A survey of the ethics of climate change is Gardiner (2004), pp. 555 600.
Box 3�1 | Dangerous interference with the climate system
Article 2 of the United Nations Framework Convention on Climate
Change states that “the ultimate objective of the Convention
[…] is to achieve […] stabilization of greenhouse gas concentra-
tions in the atmosphere at a level that would prevent dangerous
anthropogenic interference with the climate system.” Judging
whether our interference in the climate system is dangerous, i. e.,
risks causing a very bad outcome, involves two tasks: estimat-
ing the physical consequences of our interference and their
likelihood; and assessing their significance for people. The first
falls to science, but, as the Synthesis Report of the IPCC Fourth
Assessment Report (AR4) states, “Determining what constitutes
‘dangerous anthropogenic interference with the climate system’
in relation to Article 2 of the UNFCCC involves value judgements”
(IPCC, 2007, p.42). Value judgements are governed by the theory
of value. In particular, valuing risk is covered by decision theory
and is dealt with in Chapter 2. Central questions of value that
come within the scope of ethics, as well as economic methods for
measuring certain values are examined in this chapter.
215215
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
take place. UNFCCC sets the aim of “avoiding dangerous anthropo-
genic interference with the climate system”, and judging what is dan-
gerous is partly a task for ethics (see Box 3.1). Besides justice, fairness,
and rights, a central concern of ethics is value. Judgements of value
underlie the question of what interference with the climate system
would be dangerous.
Indeed, ethical judgements of value underlie almost every decision
that is connected with climate change, including decisions made by
individuals, public and private organizations, governments, and group-
ings of governments. Some of these decisions are deliberately aimed at
mitigating climate change or adapting to it. Many others influence the
progress of climate change or its impacts, so they need to take climate
change into account.
Ethics may be broadly divided into two branches: justice and value.
Justice is concerned with ensuring that people get what is due to them.
If justice requires that a person should not be treated in a particular
way uprooted from her home by climate change, for example then
the person has a right not to be treated that way. Justice and rights are
correlative concepts. On the other hand, criteria of value are concerned
with improving the world: making it a better place. Synonyms for
‘value’ in this context are ‘good’, ‘goodness’ and ‘benefit’. Antonyms
are ‘bad’, ‘harm’ and ‘cost’.
To see the difference between justice and value, think of a transfer of
wealth made by a rich country to a poor one. This may be an act of
restitution. For example, it may be intended to compensate the poor
country for harm that has been done to it by the rich country’s emis-
sions of GHG. In this case, the transfer is made on grounds of justice.
The payment is taken to be due to the poor country, and to satisfy a
right that the poor country has to compensation. Alternatively, the rich
country may make the transfer to support the poor country’s mitiga-
tion effort, because this is beneficial to people in the poor country,
the rich country, and elsewhere. The rich country may not believe the
poor country has a right to the support, but makes the payment simply
because it does ‘good’. This transfer is made on grounds of value. What
would be good to do is not necessarily required as a matter of justice.
Justice is concerned with what people are entitled to as a matter of
their rights.
The division between justice and value is contested within moral phi-
losophy, and so is the nature of the interaction between the two.
Some authors treat justice as inviolable (Nozick, 1974): justice sets
limits on what we may do and we may promote value only within
those limits. An opposite view called ‘teleological’ by Rawls
(1971) is that the right decision to make is always determined
by the value of the alternatives, so justice has no role. But despite
the complexity of their relationship and the controversies it raises,
the division between justice and value provides a useful basis for
organizing the discussion of ethical concepts and principles. We
have adopted it in this chapter: sections 3.3 and 3.4 cover justice
and value, respectively. One topic appears in both sections because
it bridges the divide: this topic is distributive justice viewed one way
and the value of equality viewed the other. Section 3.3.7 on geoen-
gineering is also in an intermediate position because it raises ethical
issues of both sorts. Section 3.6 explains how some ethical values
can be measured by economic methods of valuation. Section 3.5
describes the scope and limitations of these methods. Later sections
develop the concepts and methods of economics in more detail. Prac-
tical ways to take account of different values in policy-making are
discussed in Section 3.7.1.
3.3 Justice, equity and
responsibility
Justice, fairness, equity, and responsibility are important in interna-
tional climate negotiations, as well as in climate-related political deci-
sion making within countries and for individuals.
In this section we examine distributive justice, which, for the purpose
of this review, is about outcomes, and procedural justice or the way in
which outcomes are brought about. We also discuss compensation for
damage and historic responsibility for harm. In the context of climate
change, considerations of justice, equity, and responsibility concern the
relations between individuals, as well as groups of individuals (e. g.,
countries), both at a single point in time and across time. Accordingly,
we distinguish intra-generational from intergenerational justice. The
literature has no agreement on a correct answer to the question, what
is just? We indicate where opinions differ.
3�3�1 Causal and moral responsibility
From the perspective of countries rather than individuals or groups of
individuals, historic emissions can help determine causal responsibil-
ity for climate change (den Elzen etal., 2005; Lamarque etal., 2010;
Höhne etal., 2011). Many developed countries are expected to suf-
fer relatively modest physical damage and some are even expected to
realize benefits from future climate change (see Tol, 2002a; b). On the
other hand, some developing countries bear less causal responsibil-
ity, but could suffer significant physical damage from climate change
(IPCC, 2007, WG II AR4 SPM). This asymmetry gives rise to the follow-
ing questions of justice and moral responsibility: do considerations of
justice provide guidance in determining the appropriate level of pres-
ent and future global emissions; the distribution of emissions among
those presently living; and the role of historical emissions in distribut-
ing global obligations? The question also arises of who might be con-
sidered morally responsible for achieving justice, and, thus, a bearer of
duties towards others. The question of moral responsibility is also key
to determining whether anyone owes compensation for the damage
caused by emissions.
216216
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
3�3�2 Intergenerational justice and rights of
future people
Intergenerational justice encompasses some of the moral duties owed
by present to future people and the rights that future people hold
against present people.
2
A legitimate acknowledgment that future or
past generations have rights relative to present generations is indica-
tive of a broad understanding of justice.
3
While justice considerations
so understood are relevant, they cannot cover all our concerns regard-
ing future and past people, including the continued existence of
humankind and with a high level of wellbeing.
4
What duties do present generations owe future generations given that
current emissions will affect their quality of life? Some justice theo-
rists have offered the following argument to justify a cap on emissions
(Shue, 1993, 1999; Caney, 2006a; Meyer and Roser, 2009; Wolf, 2009).
If future people’s basic rights include the right to survival, health, and
subsistence, these basic rights are likely to be violated when tempera-
tures rise above a certain level. However, currently living people can
slow the rise in temperature by limiting their emissions at a reason-
able cost to themselves. Therefore, living people should reduce their
emissions in order to fulfil their minimal duties of justice to future
generations. Normative theorists dispute the standard of living that
corresponds to people’s basic rights (Page, 2007; Huseby, 2010). Also
in dispute is what level of harm imposed on future people is morally
objectionable. Some argue that currently living people wrongfully
harm future people if they cause them to have a lower level of well-
being than their own (e. g., Barry, 1999); others that currently living
people owe future people a decent level of wellbeing, which might be
lower than their own (Wolf, 2009). This argument raises objections on
grounds of justice since it presupposes that present people can violate
the rights of future people, and that the protection of future people’s
rights is practically relevant for how present people ought to act.
Some theorists claim that future people cannot hold rights against
present people, owing to special features of intergenerational rela-
tions: some claim that future people cannot have rights because they
cannot exercise them today (Steiner, 1983; Wellman, 1995, ch. 4). Oth-
ers point out that interaction between non-contemporaries is impos-
sible (Barry, 1977, pp. 243 244, 1989, p.189). However, some justice
theorists argue that neither the ability to, nor the possibility of, mutual
interaction are necessary in attributing rights to people (Barry, 1989;
Buchanan, 2004). They hold that rights are attributed to beings whose
interests are important enough to justify imposing duties on others.
2
In the philosophical literature, “justice between generations” typically refers to
the relations between people whose lifetimes do not overlap (Barry, 1977). In
contrast, “justice between age groups” refers to the relations of people whose
lifetimes do overlap (Laslett and Fishkin, 1992). See also Gardiner (2011),
pp.145 – 48.
3
See Rawls (1971, 1999), Barry (1977), Sikora and Barry (1978), Partridge (1981),
Parfit (1986), Birnbacher (1988), and Heyd (1992).
4
See Baier (1981), De-Shalit (1995), Meyer (2005), and for African philosophi-
cal perspectives see, Behrens (2012). See Section 3.4 on the wellbeing of future
people.
The main source of scepticism about the rights of future people and
the duties we owe them is the so-called ‘non-identity problem’. Actions
we take to reduce our emissions will change people’s way of life and
so affect new people born. They alter the identities of future people.
Consequently, our emissions do not make future people worse off than
they would otherwise have been, since those future people would not
exist if we took action to prevent our emissions. This makes it hard to
claim that our emissions harm future people, or that we owe it to them
as a matter of their rights to reduce our emissions.
5
It is often argued that the non-identity problem can be overcome
(McMahan, 1998; Shiffrin, 1999; Kumar, 2003; Meyer, 2003; Harman,
2004; Reiman, 2007; Shue, 2010). In any case, duties of justice do not
include all the moral concerns we should have for future people. Other
concerns are matters of value rather than justice, and they too can be
understood in such a way that they are not affected by the non-iden-
tity problem. They are considered in Section 3.4.
If present people have a duty to protect future people’s basic rights,
this duty is complicated by uncertainty. Present people’s actions or
omissions do not necessarily violate future people’s rights; they create
a risk of their rights being violated (Bell, 2011). To determine what cur-
rently living people owe future people, one has to weigh such uncer-
tain consequences against other consequences of their actions, includ-
ing the certain or likely violation of the rights of currently living people
(Oberdiek, 2012; Temkin, 2012). This is important in assessing many
long-term policies, including on geoengineering (see Section 3.3.7),
that risk violating the rights of many generations of people (Crutzen,
2006; Schneider, 2008; Victor etal., 2009; Baer, 2010; Ott, 2012).
3�3�3 Intergenerational justice: distributive
justice
Suppose that a global emissions ceiling that is intergenerationally just
has been determined (recognizing that a ceiling is not the only way to
deal with climate change), the question then arises of how the ceil-
ing ought to be divided among states (and, ultimately, their individ-
ual members) (Jamieson, 2001; Singer, 2002; Meyer and Roser, 2006;
Caney, 2006a). Distributing emission permits is a way of arriving at a
globally just division. Among the widely discussed views on distribu-
tive justice are strict egalitarianism (Temkin, 1993), indirect egalitarian
views including prioritarianism (Parfit, 1997), and sufficientarianism
(Frankfurt, 1999). Strict egalitarianism holds that equality has value
in itself. Prioritarianism gives greater weight to a person’s wellbeing
the less well off she is, as described in Section 3.4. Sufficientarianism
recommends that everyone should be able to enjoy a particular level
of wellbeing.
5
For an overview of the issue see Meyer (2010). See also Schwartz (1978), Parfit
(1986), and Heyd (1992). For a different perspective see Perrett (2003).
217217
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
For example, two options can help apply prioritarianism to the dis-
tribution of freely allocated and globally tradeable emission permits.
The first is to ignore the distribution of other goods. Then strict egali-
tarianism or prioritarianism will require emission permits to be distrib-
uted equally, since they will have one price and are thus equivalent
to income. The second is to take into account the unequal distribution
of other assets. Since people in the developing world are less well off
than in the developed world, strict egalitarianism or prioritarianism
would require most or all permits to go to the developing world. How-
ever, it is questionable whether it is appropriate to bring the overall
distribution of goods closer to the prioritarian ideal through the dis-
tribution of just one good (Wolff and de-Shalit, 2007; Caney, 2009,
2012).
3�3�4 Historical responsibility and distributive
justice
Historical responsibility for climate change depends on countries’ con-
tributions to the stock of GHGs. The UNFCCC refers to “common but
differentiated responsibilities” among countries of the world.
6
This is
sometimes taken to imply that current and historical causal responsi-
bility for climate change should play a role in determining the obliga-
tions of different countries in reducing emissions and paying for adap-
tation measures globally (Rajamani, 2000; Rive et al., 2006; Friman,
2007).
A number of objections have been raised against the view that his-
torical emissions should play a role (see, e. g., Gosseries, 2004; Caney,
2005; Meyer and Roser, 2006; Posner and Weisbach, 2010). First, as
currently living people had no influence over the actions of their ances-
tors, they cannot be held responsible for them. Second, previously liv-
ing people may be excused from responsibility on the grounds that
they could not be expected to know that their emissions would have
harmful consequences. Thirdly, present individuals with their particu-
lar identities are not worse or better off as a result of the emission-
generating activities of earlier generations because, owing to the non-
identity problem, they would not exist as the individuals they are had
earlier generations not acted as they did.
From the perspective of distributive justice, however, these objections
need not prevent past emissions and their consequences being taken
into account (Meyer and Roser, 2010; Meyer, 2013). If we are only
concerned with the distribution of benefits from emission-generating
activities during an individual’s lifespan, we should include the ben-
efits present people have received from their own emission-generating
activities. Furthermore, present people have benefited since birth or
conception from past people’s emission-producing actions. They are
6
Specifically, Article 3 of the UNFCCC includes the sentence: “The Parties should
protect the climate system for the benefit of present and future generations of
humankind, on the basis of equity and in accordance with their common but dif-
ferentiated responsibilities and respective capabilities.”
therefore better off as a result of past emissions, and any principle of
distributive justice should take that into account. Some suggest that
taking account of the consequences of some past emissions in this
way should not be subject to the objections mentioned in the previous
paragraph (see Shue, 2010). Other concepts associated with historical
responsibility are discussed in Chapter 4.
3�3�5 Intra-generational justice: compensatory
justice and historical responsibility
Do those who suffer disproportionately from the consequences of cli-
mate change have just claims to compensation against the main per-
petrators or beneficiaries of climate change (see, e. g., Neumayer, 2000;
Gosseries, 2004; Caney, 2006b)?
One way of distinguishing compensatory from distributive claims is to
rely on the idea of a just baseline distribution that is determined by
a criterion of distributive justice. Under this approach, compensation
for climate damage and adaptation costs is owed only by people who
have acted wrongfully according to normative theory (Feinberg, 1984;
Coleman, 1992; McKinnon, 2011). Other deviations from the baseline
may warrant redistributive measures to redress undeserved benefits or
harms, but not as compensation. Some deviations, such as those that
result from free choice, may not call for any redistribution at all.
The duty to make compensatory payments (Gosseries, 2004; Caney,
2006b) may fall on those who emit or benefit from wrongful emis-
sions or who belong to a community that produced such emissions.
Accordingly, three principles of compensatory justice have been sug-
gested: the polluter pays principle (PPP), the beneficiary pays princi-
ple (BPP), and the community pays principle (CPP) (Meyer and Roser,
2010; Meyer, 2013). None of the three measures is generally accepted,
though the PPP is more widely accepted than the others. The PPP
requires the emitter to pay compensation if the agent emitted more
than its fair share (determined as outlined in Section 3.3.2) and it
either knew, or could reasonably be expected to know, that its emis-
sions were harmful. The victim should be able to show that the emis-
sions either made the victim worse off than before or pushed below a
specified threshold of harm, or both.
The right to compensatory payments for wrongful emissions under PPP
has at least three basic limitations. Two have already been mentioned
in Section 3.3.4. Emissions that took place while it was permissible
to be ignorant of climate change (when people neither did know nor
could be reasonably be expected to know about the harmful conse-
quences of emissions) may be excused (Gosseries, 2004, pp. 39 41).
See also Section 3.3.6. The non-identity problem (see Section 3.3.2)
implies that earlier emissions do not harm many of the people who
come into existence later. Potential duty bearers may be dead and can-
not therefore have a duty to supply compensatory measures. It may
therefore be difficult to use PPP in ascribing compensatory duties and
identifying wronged persons. The first and third limitations restrict the
218218
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
assignment of duties of compensation to currently living people for
their most recent emissions, even though many more people are caus-
ally responsible for the harmful effects of climate change. For future
emissions, the third limitation could be overcome through a climate
change compensation fund into which agents pay levies for imposing
the risk of harm on future people (McKinnon, 2011).
According to BPP, a person who is wrongfully better off relative to a
just baseline is required to compensate those who are worse off. Past
emissions benefit some and impose costs on others. If currently liv-
ing people accept the benefits of wrongful past emissions, it has been
argued that they take on some of the past wrongdoer’s duty of com-
pensation (Gosseries, 2004). Also, we have a duty to condemn injustice,
which may entail a duty not to benefit from an injustice that causes
harm to others (Butt, 2007). However, BPP is open to at least two
objections. First, duties of compensation arise only from past emissions
that have benefited present people; no compensation is owed for other
past emissions. Second, if voluntary acceptance of benefits is a con-
dition of their giving rise to compensatory duties, the bearers of the
duties must be able to forgo the benefits in question at a reasonable
cost.
Under CPP, moral duties can be attributed to people as members of
groups whose identity persists over generations (De-Shalit, 1995;
Thompson, 2009). The principle claims that members of a community,
including a country, can have collective responsibility for the wrongful
actions of other past and present members of the community, even
though they are not morally or causally responsible for those actions
(Thompson, 2001; Miller, 2004; Meyer, 2005). It is a matter of debate
under what conditions present people can be said to have inherited
compensatory duties. Although CPP purports to overcome the problem
that a polluter might be dead, it can justify compensatory measures
only for emissions that are made wrongfully. It does not cover emis-
sions caused by agents who were permissibly ignorant of their harm-
fulness. (The agent in this case may be the community or state).
The practical relevance of principles of compensatory justice is limited.
Insofar as the harms and benefits of climate change are undeserved,
distributive justice will require them to be evened out, independently
of compensatory justice. Duties of distributive justice do not presup-
pose any wrongdoing (see Section 3.3.4). For example, it has been
suggested on grounds of distributive justice that the duty to pay for
adaptation should be allocated on the basis of people’s ability to pay,
which partly reflects the benefit they have received from past emis-
sions (Jamieson, 1997; Shue, 1999; Caney, 2010; Gardiner, 2011).
However, present people and governments can be said to know about
both the seriously harmful consequences of their emission-generating
activities for future people and effective measures to prevent those
consequences. If so and if they can implement these measures at a rea-
sonable cost to themselves to protect future people’s basic rights (see,
e. g., Birnbacher, 2009; Gardiner, 2011), they might be viewed as owing
intergenerational duties of justice to future people (see Section 3.3.2).
3�3�6 Legal concepts of historical
responsibility
Legal systems have struggled to define the boundaries of responsibility
for harmful actions and are only now beginning to do so for climate
change. It remains unclear whether national courts will accept lawsuits
against GHG emitters, and legal scholars vigorously debate whether
liability exists under current law (Mank, 2007; Burns and Osofsky,
2009; Faure and Peeters, 2011; Haritz, 2011; Kosolapova, 2011; Kysar,
2011; Gerrard and Wannier, 2012). This section is concerned with moral
responsibility, which is not the same as legal responsibility. But moral
thinking can draw useful lessons from legal ideas.
Harmful conduct is generally a basis for liability only if it breaches
some legal norm (Tunc, 1983), such as negligence, or if it interferes
unreasonably with the rights of either the public or property owners
(Mank, 2007; Grossman, 2009; Kysar, 2011; Brunée etal., 2012; Gold-
berg and Lord, 2012; Koch etal., 2012). Liability for nuisance does not
exist if the agent did not know, or have reason to know, the effects
of its conduct (Antolini and Rechtschaffen, 2008). The law in connec-
tion with liability for environmental damage still has to be settled.
The European Union, but not the United States, recognizes exemption
from liability for lack of scientific knowledge (United States Congress,
1980; European Union, 2004). Under European law, and in some US
states, defendants are not responsible if a product defect had not yet
been discovered (European Commission, 1985; Dana, 2009). Some
legal scholars suggest that assigning blame for GHG emissions dates
back to 1990 when the harmfulness of such emissions was established
internationally, but others argue in favour of an earlier date (Faure and
Nollkaemper, 2007; Hunter and Salzman, 2007; Haritz, 2011). Legal
systems also require a causal link between a defendant’s conduct and
some identified harm to the plaintiff, in this case from climate change
(Tunc, 1983; Faure and Nollkaemper, 2007; Kosolapova, 2011; Kysar,
2011; Brunée etal., 2012; Ewing and Kysar, 2012; Goldberg and Lord,
2012). A causal link might be easier to establish between emissions
and adaptation costs (Farber, 2007). Legal systems generally also
require causal foreseeability or directness (Mank, 2007; Kosolapova,
2011; van Dijk, 2011; Ewing and Kysar, 2012), although some statutes
relax this requirement in specific cases (such as the US Comprehensive
Environmental Response, Compensation, and Liability Act (CERCLA),
commonly known as Superfund. Emitters might argue that their contri-
bution to GHG levels was too small and the harmful effects too indirect
and diffuse to satisfy the legal requirements (Sinnot-Armstrong, 2010;
Faure and Peeters, 2011; Hiller, 2011; Kysar, 2011; van Dijk, 2011; Ger-
rard and Wannier, 2012).
Climate change claims could also be classified as unjust enrichment
(Kull, 1995; Birks, 2005), but legal systems do not remedy all forms of
enrichment that might be regarded as ethically unjust (Zimmermann,
1995; American Law Institute, 2011; Laycock, 2012). Under some legal
systems, liability depends on whether benefits were conferred without
legal obligation or through a transaction with no clear change of own-
219219
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
ership (Zimmermann, 1995; American Law Institute, 2011; Laycock,
2012). It is not clear that these principles apply to climate change.
As indicated, legal systems do not recognize liability just because a
positive or negative externality exists. Their response depends on the
behaviour that caused the externality and the nature of the causal
link between the agent’s behaviour and the resulting gain or loss to
another.
3�3�7 Geoengineering, ethics, and justice
Geoengineering (also known as climate engineering [CE]), is large-
scale technical intervention in the climate system that aims to cancel
some of the effects of GHG emissions (for more details see Working
Group I (WGI) 6.5 and WGIII 6.9). Geoengineering represents a third
kind of response to climate change, besides mitigation and adaptation.
Various options for geoengineering have been proposed, including dif-
ferent types of solar radiation management (SRM) and carbon dioxide
removal (CDR). This section reviews the major moral arguments for and
against geoengineering technologies (for surveys see Robock, 2008;
Corner and Pidgeon, 2010; Gardiner, 2010; Ott, 2010; Betz and Cacean,
2012; Preston, 2013). These moral arguments do not apply equally to
all proposed geoengineering methods and have to be assessed on a
case-specific basis.
7
Three lines of argument support the view that geoengineering tech-
nologies might be desirable to deploy at some point in the future. First,
that humanity could end up in a situation where deploying geoengi-
neering, particularly SRM, appears as a lesser evil than unmitigated
climate change (Crutzen, 2006; Gardiner, 2010; Keith et al., 2010;
Svoboda, 2012a; Betz, 2012). Second, that geoengineering could be
a more cost-effective response to climate change than mitigation or
adaptation (Barrett, 2008). Such efficiency arguments have been criti-
cized in the ethical literature for neglecting issues such as side-effects,
uncertainties, or fairness (Gardiner, 2010, 2011; Buck, 2012). Third,
that some aggressive climate stabilization targets cannot be achieved
through mitigation measures alone and thus must be complemented
by either CDR or SRM (Greene etal., 2010; Sandler, 2012).
Geoengineering technologies face several distinct sets of objections.
Some authors have stressed the substantial uncertainties of large-
scale deployment (for overviews of geoengineering risks see also
7
While the literature typically associates some arguments with particular types of
methods (e. g., the termination problem with SRM), it is not clear that there are
two groups of moral arguments: those applicable to all SRM methods on the one
side and those applicable to all CDR methods on the other side. In other words,
the moral assessment hinges on aspects of geoengineering that are not connected
to the distinction between SRM and CDR.
Schneider (2008) and Sardemann and Grunwald (2010)), while others
have argued that some intended and unintended effects of both CDR
and SRM could be irreversible (Jamieson, 1996) and that some cur-
rent uncertainties are unresolvable (Bunzl, 2009). Furthermore, it has
been pointed out that geoengineering could make the situation worse
rather than better (Hegerl and Solomon, 2009; Fleming, 2010; Hamil-
ton, 2013) and that several technologies lack a viable exit option: SRM
in particular would have to be maintained as long as GHG concentra-
tions remain elevated (The Royal Society, 2009).
Arguments against geoengineering on the basis of fairness and jus-
tice deal with the intra-generational and intergenerational distribu-
tional effects. SRM schemes could aggravate some inequalities if, as
expected, they modify regional precipitation and temperature patterns
with unequal social impacts (Bunzl, 2008; The Royal Society, 2009;
Svoboda etal., 2011; Preston, 2012). Furthermore, some CDR methods
would require large-scale land transformations, potentially competing
with agricultural land-use, with uncertain distributive consequences.
Other arguments against geoengineering deal with issues including
the geopolitics of SRM, such as international conflicts that may arise
from the ability to control the “global thermostat” (e. g., Schelling,
1996; Hulme, 2009), ethics (Hale and Grundy, 2009; Preston, 2011;
Hale and Dilling, 2011; Svoboda, 2012b; Hale, 2012b), and a critical
assessment of technology and modern civilization in general (Fleming,
2010; Scott, 2012).
One of the most prominent arguments against geoengineering sug-
gests that geoengineering research activities might hamper mitigation
efforts (e. g., Jamieson, 1996; Keith, 2000; Gardiner, 2010), which pre-
sumes that geoengineering should not be considered an acceptable
substitute for mitigation. The central idea is that research increases the
prospect of geoengineering being regarded as a serious alternative to
emission reduction (for a discussion of different versions of this argu-
ment see Hale, 2012a; Hourdequin, 2012). Other authors have argued,
based on historical evidence and analogies to other technologies, that
geoengineering research might make deployment inevitable (Jamie-
son, 1996; Bunzl, 2009), or that large-scale field tests could amount to
full-fledged deployment (Robock etal., 2010). It has also been argued
that geoengineering would constitute an unjust imposition of risks
on future generations, because the underlying problem would not be
solved but only counteracted with risky technologies (Gardiner, 2010;
Ott, 2012; Smith, 2012). The latter argument is particularly relevant to
SRM technologies that would not affect greenhouse gas concentra-
tions, but it would also apply to some CDR methods, as there may be
issues of long-term safety and capacity of storage.
Arguments in favour of research on geoengineering point out that
research does not necessarily prepare for future deployment, but can,
on the contrary, uncover major flaws in proposed schemes, avoid pre-
mature CE deployment, and eventually foster mitigation efforts (e. g.
Keith etal., 2010). Another justification for Research and Development
(R&D) is that it is required to help decision-makers take informed deci-
sions (Leisner and Müller-Klieser, 2010).
220220
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
3.4 Values and wellbeing
One branch of ethics is the theory of value. Many different sorts of
value can arise, and climate change impinges on many of them. Value
affects nature and many aspects of human life. This section surveys
some of the values at stake in climate change, and examines how far
these values can be measured, combined, or weighed against each
other. Each value is subject to debate and disagreement. For example,
it is debatable whether nature has value in its own right, apart from
the benefit it brings to human beings. Decision-making about climate
change is therefore likely to be contentious.
Since values constitute only one part of ethics, if an action will increase
value overall it by no means follows that it should be done. Many
actions benefit some people at the cost of harming others. This raises
a question of justice even if the benefits in total exceed the costs.
Whereas a cost to a person can be compensated for by a benefit to
that same person, a cost to a person cannot be compensated for by
a benefit to someone else. To suppose it can is not to “take seriously
the distinction between persons”, as John Rawls puts it (1971, p.27).
Harming a person may infringe their rights, or it may be unfair to them.
For example, when a nation’s economic activities emit GHG, they may
benefit the nation itself, but may harm people in other nations. Even if
the benefits are greater in value than the harms, these activities may
infringe other nations’ rights. Other nations may therefore be entitled
to object to them on grounds of justice.
Any decision about climate change is likely to promote some values
and damage others. These may be values of very different sorts. In
decision making, different values must therefore be put together or
balanced against each other. Some pairs of values differ so radically
from each other that they cannot be determinately weighed together.
For example, it may be impossible to weigh the value of preserving a
traditional culture against the material income of the people whose
culture it is, or to weigh the value of biodiversity against human well-
being. Some economists claim that one person’s wellbeing cannot be
weighed against another’s (Robbins, 1937; Arrow, 1963). When values
cannot be determinately weighed, they are said to be ‘incommensu-
rable’ or ‘incomparable’ (Chang, 1997). Multi-Criteria Analysis (MCA)
(discussed in Section 3.7.2.1) is a technique that is designed to take
account of several incommensurable values (De Montis etal., 2005;
Zeleny and Cochrane, 1982).
3�4�1 Non-human values
Nature provides great benefits to human beings in ways that range
from absorbing our waste, to beautifying the world we inhabit. An
increasing number of philosophers have argued in recent years that
nature also has value in its own right, independently of its benefits to
human beings (Leopold, 1949; Palmer, 2011). They have argued that
we should recognize animal values, the value of life itself, and even the
value of natural systems and nature itself.
In moral theory, rational adult humans, who are self-conscious subjects
of a life, are often taken (following Kant, 1956) to have a kind of uncon-
ditional moral worth sometimes called ‘dignity’ that is not found
elsewhere on earth. Others believe that moral worth can be found else-
where (Dryzek, 1997). Many human beings themselves lack rationality
or subjectivity, yet still have moral worth the very young, the very
old and people with various kinds of impairment among them. Given
that, why deny moral worth to those animals that are to some extent
subjects of a life, who show emotional sophistication (Regan, 2004),
and who experience pleasure, pain, suffering, and joy (Singer, 1993)?
An argument for recognizing value in plants as well as animals was
proposed by Richard Routley (1973). Routley gives the name ‘human
chauvinism’ to the view that humans are the sole possessors of intrin-
sic value. He asks us to imagine that the last man on earth sets out to
destroy every living thing, animal or plant. Most people believe this
would be wrong, but human chauvinists are unable to explain why.
Human chauvinism appears to be simply a prejudice in favour of the
human species (Routley and Routley, 1980). In contrast, some philoso-
phers argue that value exists in the lives of all organisms, to the extent
that they have the capacity to flourish (Taylor, 1986; Agar, 2001).
Going further, other philosophers have argued that biological com-
munities and holistic ecological entities also have value in their own
right. Some have argued that a species has more value than all of its
individuals have together, and that an ecosystem has still more value
(Rolston, 1988, 1999; compare discussion in Brennan and Lo, 2010).
It has further been proposed that, just as domination of one human
group by another is a moral evil, showing disrespect for the value of
others, then so is the domination of nature by humans in general.
If nature and its systems have moral worth, then the domination of
nature is also a kind of disrespect (Jamieson, 2010).
If animals, plants, species, and ecosystems do have value in their own
right, then the moral impact of climate change cannot be gauged by
its effects on human beings alone. If climate change leads to the loss
of environmental diversity, the extinction of plant and animal species,
and the suffering of animal populations, then it will cause great harms
beyond those it does to human beings. Its effects on species numbers,
biodiversity, and ecosystems may persist for a very long time, perhaps
even longer than the lifetime of the human species (Nolt, 2011).
It is very difficult to measure non-human values in a way that makes
them commensurate with human values. Economists address this
issue by dividing value into use value (associated with actual use of
nature instrumental value) and nonuse or existence value (intrinsic
value of nature). As an example, biodiversity might have value because
of the medical drugs that might be discovered among the diverse
biota (use value). Or biodiversity might be valued by individuals sim-
221221
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
ply because they believe that biologic diversity is important, over and
above any use to people that might occur. The total amount people are
willing to pay has sometimes been used as an economic measure of
the total value (instrumental and intrinsic) of these features (Aldred,
1994). As the discussion of the past few paragraphs has suggested,
nature may have additional value, over and above the values placed by
individual humans (Broome, 2009; Spash etal., 2009).
3�4�2 Cultural and social values
The value of human wellbeing is considered in Section 3.4.3, but the
human world may also possess other values that do not form part of
the wellbeing of individual humans. Living in a flourishing culture and
society contributes to a person’s wellbeing (Kymlicka, 1995; Appiah,
2010), but some authors claim that cultures and societies also pos-
sess values in their own right, over and above the contribution they
make to wellbeing (Taylor, 1995). Climate change threatens damage to
cultural artefacts and to cultures themselves (Adger etal., 2012). Evi-
dence suggests that it may already be damaging the culture of Arctic
indigenous peoples (Ford etal., 2006, 2008; Crate, 2008; Hassol, 2004;
see also WGII Chapter 12). Cultural values and indigenous peoples are
discussed in Section 3.10.2.
The degree of equality in a society may also be treated as a value that
belongs to a society as a whole, rather than to any of the individu-
als who make up the society. Various measures of this value are avail-
able, including the Gini coefficient and the Atkinson measure (Gini,
1912; Atkinson, 1970); for an assessment see (Sen, 1973). Section 3.5
explains that the value of equality can alternatively be treated as a
feature of the aggregation of individual people’s wellbeings, rather
than as social value separate from wellbeing.
3�4�3 Wellbeing
Most policy concerned with climate change aims ultimately at making
the world better for people to live in. That is to say, it aims to promote
people’s wellbeing. A person’s wellbeing, as the term is used here,
includes everything that is good or bad for the person everything
that contributes to making their life go well or badly. What things
are those what constitutes a person’s wellbeing? This question has
been the subject of an extensive literature since ancient times.
8
One
view is that a person’s wellbeing is the satisfaction of their prefer-
ences. Another is that it consists in good feelings such as pleasure. A
third is that wellbeing consists in possessing the ordinary good things
of life, such as health, wealth, a long life, and participating well in a
8
For example: Aristotle, Nicomachean Ethics. Recent work includes: Griffin (1986);
Sumner (1999); Kraut (2007).
good community. The ‘capabilities approach’ in economics (Sen, 1999)
embodies this last view. It treats the good things of life as ‘function-
ings’ and ‘capabilities’ things that a person does and things that
they have a real opportunity of doing, such as living to old age, having
a good job, and having freedom of choice.
A person’s wellbeing will be affected by many of the other values that
are mentioned above, and by many of the considerations of justice
mentioned in Section 3.3. It is bad for a person to have their rights
infringed or to be treated unfairly, and it is good for a person to live
within a healthy culture and society, surrounded by flourishing nature.
Various concrete measures of wellbeing are in use (Fleurbaey, 2009;
Stiglitz etal., 2009). Each reflects a particular view about what well-
being consists in. For example, many measures of ‘subjective wellbe-
ing’ (Oswald and Wu, 2010; Kahneman and Deaton, 2010) assume that
wellbeing consists in good feelings. Monetary measures of wellbeing,
which are considered in Section 3.6, assume that wellbeing consists
in the satisfaction of preferences. Other measures assume wellbeing
consists in possessing a number of specific good things. The Human
Development Index (HDI) is intended to be an approximate measure of
wellbeing understood as capabilities and functionings (UNDP, 2010). It
is based on three components: life expectancy, education, and income.
The capabilities approach has inspired other measures of wellbeing
too (Dervis and Klugman, 2011). In the context of climate change,
many different metrics of value are intended to measure particular
components of wellbeing: among them are the numbers of people at
risk from hunger, infectious diseases, coastal flooding, or water scar-
city. These metrics may be combined to create a more general measure.
Schneider etal. (2000) advocates the use of a suite of five metrics:
(1) monetary loss, (2) loss of life, (3) quality of life (taking account of
forced migration, conflict over resources, cultural diversity, and loss of
cultural heritage sites), (4) species or biodiversity loss, and (5) distribu-
tion and equity.
3�4�4 Aggregation of wellbeing
Whatever wellbeing consists of, policy-making must take into account
the wellbeing of everyone in the society. So the wellbeings of differ-
ent people have somehow to be aggregated together. How do they
combine to make up an aggregate value of wellbeing for a society as a
whole? Social choice theory takes up this problem (Arrow, 1963; Sen,
1970). Section 3.6 will explain that the aim of economic valuation is to
measure aggregate wellbeing.
Assume that each person has a level of wellbeing at each time they are
alive, and call this their ‘temporal wellbeing’ at that time. In a society,
temporal wellbeing is distributed across times and across the people.
When a choice is to be made, each of the options leads to a particular
distribution of wellbeing. Our aim is to assess the value of such distri-
butions. Doing so involves aggregating wellbeings across times and
across people, to arrive at an overall, social value for the distribution.
222222
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
3�4�5 Lifetime wellbeing
Next let us assume that each person’s temporal wellbeings can be
aggregated to determine a ‘lifetime wellbeing’ for the person, and that
the social value of the distribution of wellbeing depends only on these
lifetime wellbeings. This is the assumption that each person’s wellbe-
ing is “separable”, to use a technical term. It allows us to split aggre-
gation into two steps. First, we aggregate each person’s temporal well-
beings across the times in their life in order to determine their lifetime
wellbeing. The second step in the next section is to aggregate across
individuals using a social welfare function.
On one account, a person’s lifetime wellbeing is simply the total of
their temporal wellbeings at each time they are alive. If a person’s
wellbeing depended only on the state of their health, this formula
would be equivalent to ‘QALYs’ or ‘DALYs’ (quality-adjusted life years
or disability-adjusted life years), which are commonly used in the anal-
ysis of public health (Murray, 1994; Sassi, 2006). These measures take
a person’s lifetime wellbeing to be the total number of years they live,
adjusted for their health in each year. Since wellbeing actually depends
on other things as well as health, QALYs or DALYs provide at best an
approximate measure of lifetime wellbeing. If they are aggregated
across people by simple addition, it assumes implicitly that a year of
healthy life is equally as valuable to one person as it is to another.
That may be an acceptable approximation for the broad evaluation
of climate change impacts and policies, especially for evaluating their
effects on health (Nord etal., 1999; Mathers etal., 2009; but also see
Currie etal., 2008).
Other accounts give either increasing, (Velleman, 1991) or alternatively
decreasing, (Kaplow etal., 2010) weight to wellbeing that comes in
later years of life, in determining a person’s lifetime wellbeing.
3�4�6 Social welfare functions
Once we have a lifetime wellbeing for each person, the next step is
to aggregate these lifetime wellbeings across people, to determine an
overall value for society. This involves comparing one person’s wellbe-
ing with another’s. Many economists have claimed that interpersonal
comparisons of wellbeing are impossible.
9
If they are right, the wellbe-
ings of different people are incommensurable and cannot be aggre-
gated. In this section we set this view aside, and assume that temporal
wellbeings are measured in a way that is comparable across people.
10
This allows us to aggregate different people’s lifetime wellbeings
through a social welfare function (SWF) to arrive at an overall value or
‘social welfare’.
11
9
Examples are: Robbins (1937), Archibald (1959), Arrow (1963). A survey and
discussion of this sceptical view appears in Hammond (1993).
10
Potential bases of interpersonal comparisons are examined in: Fleurbaey and
Hammond (2004); Sen (1982); Elster and Roemer (1993); Mirrlees (1982);
Broome, (2004); Arrow (1977); Harsanyi (1977); Adler (2011).
11
A recent major study is Adler (2011).
We shall first consider SWFs under the simplifying but unrealistic
assumption that the decisions that are to be made do not affect how
many people exist or which people exist: all the options contain the
same people. A theorem of Harsanyi’s (1955) gives some grounds for
thinking that, given this assumption, the SWF is additively separable
between people. This means it has the form:
Equation 3�4�1 V = v
1
(w
1
) + v
2
(w
2
) + … + v
J
(w
J
)
Here w
i
is person is lifetime wellbeing. This formula says that each
person’s wellbeing can be assigned a value v
i
(w
i
), and all these val-
ues one for each person are added up to determine the social
value of the distribution.
The proof of Harsanyi’s Theorem depends on assumptions that can
be challenged (Diamond, 1967; Broome, 2004; Fleurbaey, 2010). So,
although the additively separable form shown in Equation 3.4.1 is
commonly assumed in economic valuations, it is not entirely secure.
In particular, this form makes it impossible to give any value to equal-
ity except indirectly through prioritarianism, which was introduced in
Section 3.3.2 and is defined below. The value of inequality cannot be
measured by the Gini coefficient, for example, since this measure is not
additively separable (Sen, 1973).
It is often assumed that the functions v
i
() all have the same form,
which means that each person’s wellbeing is valued in the same way:
Equation 3�4�2 V = v (w
1
) + v (w
2
) + … + v (w
J
)
Alternatively, the wellbeing of people who live later is sometimes
discounted relative to the wellbeing of people who live earlier; this
implies that the functional form of v
i
() varies according to the date
when people live. Discounting of later wellbeing is often called ‘pure’
discounting. It is discussed in Section 3.6.2.
Even if we accept Equation 3.4.2, different ethical theories imply dif-
ferent SWFs. Utilitarianism values only the total of people’s wellbeing.
The SWF may be written:
Equation 3�4�3 V = w
1
+ w
2
+ … + w
J
Utilitarianism gives no value to equality in the distribution of wellbe-
ing: a given total of wellbeing has the same value however unequally
it is distributed among people.
But the idea of distributive justice mentioned in Section 3.3.3 sug-
gests that equality of wellbeing does have value. Equation 3.4.2 will
give value to equality if the function v() is strictly concave. This
means the graph of v() curves downwards, as Figure 3.1 illustrates.
(Section 3.6.1.1 explains that a person’s wellbeing w
i
is commonly
assumed to be a strictly concave function of her consumption, but
this is a different point.) The resulting ethical theory is called priori-
tarianism. As Figure 3.1 shows, according to prioritarianism, improv-
Figure 3�1 | The prioritarian view of social welfare. The figure compares the social val-
ues of increases in wellbeing for a better-off and a worse-off person.
Wellbeing
Social Value of Wellbeing
Equal Increase
in Wellbeing
Increase in
Social Value
Lesser
Greater
223223
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
ing a person’s wellbeing contributes more to social welfare if the
person is badly off than if they are well off. The prioritarian slogan is
“priority to the worse off”. Prioritarianism indirectly gives value to
equality: it implies that a given total of wellbeing is more valuable
the more equally it is distributed (Sen, 1973; Weirich, 1983; Parfit,
1997). In judgements about climate change, a prioritarian function
will give relatively more importance to the interests of poorer people
and poorer countries.
3�4�7 Valuing population
The next problem in aggregating wellbeing is to take account of
changes in population. Climate change can be expected to affect the
world’s human population. Severe climate change might even lead to a
catastrophic collapse of the population (Weitzman, 2009), and even to
the extinction of human beings. Any valuation of the impact of climate
change and of policies to mitigate climate change should therefore
take changes in population into account.
The utilitarian and prioritarian SWFs for a fixed population may be
extended in a variety of ways to a variable population. For example,
the utilitarian function may be extended to ‘average utilitarianism’
(Hurka, 1982), whose SWF is the average of people’s wellbeing. Aver-
age utilitarianism gives no value to increasing numbers of people. The
implicit or explicit goal of a great deal of policy-making is to promote
per capita wellbeing (Hardin, 1968). This is to adopt average utilitari-
anism. This goal tends to favour anti-natalist policies, aimed at limiting
population. It would strongly favour population control as a means of
mitigating climate change, and it would not take a collapse of popula-
tion to be, in itself, a bad thing.
The utilitarian function may alternatively be extended to ‘critical-level
utilitarianism’, whose SWF is the total of the amount by which each
person’s wellbeing exceeds some fixed critical level. It is
Equation 3�4�4 V = (w
1
– c) + (w
2
c) + … + (w
J
c)
where c is the critical level (Broome, 2004; Blackorby et al., 2005).
Other things being equal, critical-level utilitarianism favours adding
people to the population if their wellbeing is above the critical level.
‘Total utilitarianism’ (Sidgwick, 1907) is critical-level utilitarianism
with the critical level set to zero. Its SWF is the total of people’s well-
being. Total utilitarianism is implicit in many Integrated Assessment
Models (IAMs) of climate change (e. g., Nordhaus, 2008). Its mean-
ing is indeterminate until it is settled which level of lifetime wellbeing
to count as zero. Many total utilitarians set the zero at the level of
a life that has no good or bad experiences that is lived in a coma
throughout, for instance (Arrhenius, forthcoming). Since people on
average lead better lives than this, total utilitarianism with this zero
tends to be less anti-natalist than average utilitarianism. However, it
does not necessarily favour increasing population. Each new person
damages the wellbeing of existing people, through their emissions of
GHG, their other demands on Earth’s limited resources, and the emis-
sions of their progeny. If the damage an average person does to others
in total exceeds their own wellbeing, total utilitarianism, like average
utilitarianism, favours population control as a means of mitigating cli-
mate change.
12
Each of the existing ethical theories about the value of population has
intuitively unattractive implications (Parfit, 1986). Average utilitarian-
ism is subject to particularly severe objections. Arrhenius (forthcoming)
crystallizes the problems of population ethics in the form of impos-
sibility theorems. So far, no consensus has emerged about the value of
population. Yet climate change policies are expected to affect the size
of the world’s population, and different theories of value imply very
different conclusions about the value of these policies. This is a serious
difficulty for evaluating policies aimed at mitigating climate change,
which has largely been ignored in the literature (Broome, 2012).
3.5 Economics, rights,
and duties
Sections 3.2, 3.3 and 3.4 have outlined some of the ethical principles
that can guide decision making for climate change. The remainder of
this chapter is largely concerned with the concepts and methods of
12
Harford (1998) shows that an additional person causes damage from her own
emissions and the emissions of her children (and of their children, etc.). Kelly and
Kolstad (2001) examine this issue in the specific context of climate change.
We shall first consider SWFs under the simplifying but unrealistic
assumption that the decisions that are to be made do not affect how
many people exist or which people exist: all the options contain the
same people. A theorem of Harsanyi’s (1955) gives some grounds for
thinking that, given this assumption, the SWF is additively separable
between people. This means it has the form:
Equation 3�4�1 V = v
1
(w
1
) + v
2
(w
2
) + … + v
J
(w
J
)
Here w
i
is person is lifetime wellbeing. This formula says that each
person’s wellbeing can be assigned a value v
i
(w
i
), and all these val-
ues one for each person are added up to determine the social
value of the distribution.
The proof of Harsanyi’s Theorem depends on assumptions that can
be challenged (Diamond, 1967; Broome, 2004; Fleurbaey, 2010). So,
although the additively separable form shown in Equation 3.4.1 is
commonly assumed in economic valuations, it is not entirely secure.
In particular, this form makes it impossible to give any value to equal-
ity except indirectly through prioritarianism, which was introduced in
Section 3.3.2 and is defined below. The value of inequality cannot be
measured by the Gini coefficient, for example, since this measure is not
additively separable (Sen, 1973).
It is often assumed that the functions v
i
() all have the same form,
which means that each person’s wellbeing is valued in the same way:
Equation 3�4�2 V = v (w
1
) + v (w
2
) + … + v (w
J
)
Alternatively, the wellbeing of people who live later is sometimes
discounted relative to the wellbeing of people who live earlier; this
implies that the functional form of v
i
() varies according to the date
when people live. Discounting of later wellbeing is often called ‘pure’
discounting. It is discussed in Section 3.6.2.
Even if we accept Equation 3.4.2, different ethical theories imply dif-
ferent SWFs. Utilitarianism values only the total of people’s wellbeing.
The SWF may be written:
Equation 3�4�3 V = w
1
+ w
2
+ … + w
J
Utilitarianism gives no value to equality in the distribution of wellbe-
ing: a given total of wellbeing has the same value however unequally
it is distributed among people.
But the idea of distributive justice mentioned in Section 3.3.3 sug-
gests that equality of wellbeing does have value. Equation 3.4.2 will
give value to equality if the function v() is strictly concave. This
means the graph of v() curves downwards, as Figure 3.1 illustrates.
(Section 3.6.1.1 explains that a person’s wellbeing w
i
is commonly
assumed to be a strictly concave function of her consumption, but
this is a different point.) The resulting ethical theory is called priori-
tarianism. As Figure 3.1 shows, according to prioritarianism, improv-
Figure 3�1 | The prioritarian view of social welfare. The figure compares the social val-
ues of increases in wellbeing for a better-off and a worse-off person.
Wellbeing
Social Value of Wellbeing
Equal Increase
in Wellbeing
Increase in
Social Value
Lesser
Greater
224224
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
economics. They can be used to aggregate values at different times and
places, and weigh aggregate value for different policy actions. They can
also be used to draw information about value from the data provided
by prices and markets. Economics can measure diverse benefits and
harms, taking account of uncertainty, to arrive at overall judgements of
value. It also has much to contribute to the choice and design of policy
mechanisms, as Section 3.8 and later chapters show.
Valuations provided by economics can be used on a large scale: IAMs
can be used to simulate the evolution of the world’s economy under
different climate regimes and determine an economically efficient
reduction in GHG emissions. On a smaller scale, economic methods of
CBA can be used in choosing between particular policies and technolo-
gies for mitigation.
Economics is much more than a method of valuation. For example,
it shows how decision making can be decentralized through market
mechanisms. This has important applications in policy instruments for
mitigation with potential for cost-effectiveness and efficiency (Chap-
ters 6 and 15). Economic analysis can also give guidance on how
policy mechanisms for international cooperation on mitigation can
be designed to overcome free-rider problems (Chapters 13 and 14).
However, the methods of economics are limited in what they can do.
They can be based on ethical principles, as Section 3.6 explains. But
they cannot take account of every ethical principle. They are suited
to measuring and aggregating the wellbeing of humans, but not to
taking account of justice and rights (with the exception of distribu-
tive justice see below), or other values apart from human wellbeing.
Moreover, even in measuring and aggregating wellbeing, they depend
on certain specific ethical assumptions. This section describes the limits
of economic methods.
Because of their limitations, economic valuations are often not on their
own a good basis for decision making. They frequently need to be sup-
plemented by other ethical considerations. It may then be appropriate
to apply techniques of multi-criteria analysis (MCA), discussed in Sec-
tion 3.7.2.1 (Zeleny and Cochrane, 1982; Keeney and Raiffa, 1993; De
Montis etal., 2005).
3�5�1 Limits of economics in guiding decision
making
Economics can measure and aggregate human wellbeing, but Sections
3.2, 3.3 and 3.4 explain that wellbeing may be only one of several
criteria for choosing among alternative mitigation policies. Other ethi-
cal considerations are not reflected in economic valuations, and those
considerations may be extremely important for particular decisions
that have to be made. For example, some have contended that coun-
tries that have emitted a great deal of GHG in the past owe restitution
to countries that have been harmed by their emissions. If so, this is an
important consideration in determining how much finance rich coun-
tries should provide to poorer countries to help with their mitigation
efforts. It suggests that economics alone cannot be used to determine
who should bear the burden of mitigation (also see Box 3.2).
What ethical considerations can economics cover satisfactorily? Since
the methods of economics are concerned with value, they do not take
account of justice and rights in general. However, distributive justice
can be accommodated within economics, because it can be under-
stood as a value: specifically the value of equality. The theory of fair-
ness within economics (Fleurbaey, 2008) is an account of distributive
justice. It assumes that the level of distributive justice within a soci-
ety is a function of the wellbeings of individuals, which means it can
be reflected in the aggregation of wellbeing. In particular, it may be
measured by the degree of inequality in wellbeing, using one of the
standard measures of inequality such as the Gini coefficient (Gini,
1912), as discussed in the previous section. The Atkinson measure of
inequality (Atkinson, 1970) is based on an additively separable SWF,
and is therefore particularly appropriate for representing the prioritar-
ian theory described in Section 3.4.6. Furthermore, distributive justice
can be reflected in weights incorporated into economic evaluations as
Section 3.6 explains.
Economics is not well suited to taking into account many other aspects
of justice, including compensatory justice. For example, a CBA might
not show the drowning of a Pacific island as a big loss, since the island
has few inhabitants and relatively little economic activity. It might con-
clude that more good would be done in total by allowing the island
to drown: the cost of the radical action that would be required to
save the island by mitigating climate change globally would be much
greater than the benefit of saving the island. This might be the correct
conclusion in terms of overall aggregation of costs and benefits. But
the island’s inhabitants might have a right not to have their homes
and livelihoods destroyed as a result of the GHG emissions of richer
nations far away. If that is so, their right may override the conclusions
of CBA. It may give those nations who emit GHG a duty to protect the
people who suffer from it, or at least to make restitution to them for
any harms they suffer.
Even in areas where the methods of economics can be applied in princi-
ple, they cannot be accepted without question (Jamieson, 1992; Sagoff,
2008). Particular simplifying assumptions are always required, as shown
throughout this chapter. These assumptions are not always accurate
or appropriate, and decision-makers need to keep in mind the result-
ing limitations of the economic analyses. For example, climate change
will shorten many people’s lives. This harm may in principle be included
within a CBA, but it remains highly contentious how that should be
done. Another problem is that, because economics can provide con-
crete, quantitative estimates of some but not all values, less quantifi-
able considerations may receive less attention than they deserve.
The extraordinary scope and scale of climate change raises particular
difficulties for economic methods (Stern, forthcoming). First, many of
the common methods of valuation in economics are best designed for
marginal changes, whereas some of the impacts of climate change and
225225
Social, Economic, and Ethical Concepts and Methods
3
Chapter 3
efforts at mitigation are not marginal (Howarth and Norgaard, 1992).
Second, the very long time scale of climate change makes the discount
rate crucial at the same time as it makes it highly controversial (see
Section 3.6.2). Third, the scope of the problem means it encompasses
the world’s extremes of wealth and poverty, so questions of distribu-
tion become especially important and especially difficult. Fourth, mea-
suring non-market values such as the existence of species, natural
environments, or traditional ways of life of local societies is fraught
with difficulty. Fifth, the uncertainty that surrounds climate change is
very great. It includes the likelihood of irreversible changes to societies
and to nature, and even a small chance of catastrophe. This degree of
uncertainty sets special problems for economics (Nelson, 2013).
3.6 Aggregation of costs
and benefits
3�6�1 Aggregating individual wellbeing
Policies that respond to climate change almost always have some good
and some bad effects; we say they have ‘benefits’ and ‘costs’. In choos-
ing a policy, we may treat one of the available options as a standard
of comparison for instance, the status quo. Other options will have
costs and benefits relative to this standard. Most mitigation strategies
have costs in the present and yield benefits in the future. Policy-making
involves assessing the values of these benefits and costs and weigh-
ing them against each other. Chapter 6 contains an example in which
different mitigation strategies yielding different temporal allocations