Transfers, self-enforcing agreements and climate cooperation

Printer-friendly version
From theory to application
Suzi Kerr, Steffen Lippert and Edmund Lou
Issue number: 
From Theory to Application
International agreements addressing climate change must overcome the difficulties implied by the absence of an institution with the power to ensure compliance. They have to be self-enforcing: the threat of future punishment must give participants sufficient incentives to comply with the agreed reductions in emissions voluntarily. The structure of the underlying game of choosing greenhouse gas emissions is what Albert Tucker named a Prisoner’s Dilemma (Rapoport and Chemmah, 1965). Every country has an incentive to increase their emissions unilaterally, to produce higher economic output, benefitting the individual country. The environmental costs or the increase in greenhouse gas emissions, however, are shared by the community of countries, leading to inefficiently high individual incentives to emit. In contrast, efficient global mitigation – low emissions by all countries – generates the greatest joint gains.

From folk theorems, we know that sufficiently patient players in a prisoner’s dilemma are able to sustain the first-best outcome as a subgame perfect Nash equilibrium. In the context of climate change, this means that, for patient countries, the loss of future cooperation is so large that short-run opportunistic increases in emissions today do not pay. Unfortunately, as we have learned from 30 years of climate negotiations, this theoretical insight does not easily transfer into reality.

Because we do not like where the rules of the game take us, we need to change them. The question is how. A first change of the rules of the game exploits the insight that a country’s payoffs from low or high emissions are critically dependent on its investments in compliance technology. In a recent paper, Harstad, Lancia and Russo (2018) show that, where countries face too high costs or too low benefits to sustain efficient climate cooperation, second-best strategies to sustain low emissions in the repeated climate cooperation game overinvest in green and underinvest in brown or adaptation technologies. This change in the investment increases the benefit and lowers the cost of mitigation and thereby reduces the country’s incentive to increase emissions. As for the first-best, sustaining these second-best strategies also require countries to be sufficiently patient.

An important challenge to the nearly thirty years of climate negotiations has been that the joint gains from efficient climate cooperation are unequally distributed across countries. Without resource transfers, many countries would not agree to a globally efficient agreement, and only partial agreements or agreements with lower than efficient mitigation levels can be sustained. Hence, a second rule of the game we can change concerns resource transfers among countries. Indeed, Fong and Surti (2009) show how side payments affect the optimal degree of cooperation in repeated prisoner’s dilemmas. In the case of climate cooperation, resource transfers relax the emissions compliance constraint of recipient countries with low benefits or high costs from low emissions. And if these transfers increase the payoffs of countries with high benefits from low emissions by sustaining low emissions in more countries, then it would clearly be in their interest to offer them.

Because, however, a country’s investment in compliance technology affects its costs and benefits from mitigation, its investment decisions need to be considered in the design of climate change agreements with resource transfers. In our paper, we build on Harstad et al.’s work and explore how to sustain higher levels of climate cooperation when not all countries have sufficiently high gains from low emissions to sustain Harstad et al.’s second-best equilibrium. In particular, we investigate side payments from countries with high benefits – we call them members of a climate club – to countries with low benefits – which we call applicant countries – with the aim of achieving a commitment to the efficient level of emissions by the applicants.

We model three types of side payment agreements: (1) upfront payment agreements, where members transfer resources to applicants before they decide their investments and emissions; (2) investment-based payment agreements, where transfers to an applicant country occur contingent on having observed its investments but before it decides about its emissions; and (3) results-based payment agreements, where transfers occur after the applicant country’s emissions have been observed. For each agreement type, we study how the transfer needed to induce the efficient level of emissions depends on the recipient country’s investment in compliance technology. In all three agreements, the applicant and member countries need to comply with investment and emissions levels, and members additionally need to comply with the transfers.

Two – realistic – features of our model are crucial to our results. First, emissions are typically observed with a lag, leading to the free-rider problem to start with, and second, some length of time normally elapses between an investment decision and the time at which the compliance technology is fully operational. These lags imply that the size of the transfer needed for compliance and the incentives for investment and abatement differ across our three agreements.

Our first result is that, for each applicant, there is an investment level that minimizes the transfer level it needs to receive to comply with low emissions, while for member countries, there is an investment level that maximizes the transfer level they are willing to make and still comply with low emissions. At these investment levels, countries overinvest in green technology by even more than in Harstad et al.’s (2018) second-best solution.

For member countries, the levels of transfers and investment that satisfy compliance at the emissions stage automatically also satisfy compliance at the investment and the transfer stages in all three types of agreements. Furthermore, in investment-based transfer agreements, these levels of transfers and investment also automatically satisfy compliance at the investment stage for applicants. In contrast, the automatic compliance with the investment level can fail to hold for applicant countries in upfront and results-based transfer agreements. If, for example, the cost of green investment is high for an applicant country, then the investment level that minimizes the transfer necessary to comply with low emissions will satisfy the emissions constraint, but the country will defect by investing less than necessary in the first place, causing a defection in the emissions stage as well.

Absent the need to incentivize overinvestment in green technology, all three types of agreements would have the same scope for climate cooperation. Therefore, when applicants that fulfill their emissions compliance constraints automatically also fulfill their investment compliance constraints, then all three schemes maximize that scope and policymakers can choose either one of them to implement low emissions. In this case, member countries would want to choose a results-based agreement because it minimizes the overall transfer needed for compliance of an applicant country, whereas applicants would prefer an investment-based agreement, and the final choice of the agreement depends on the negotiations between countries.

In contrast, we find that, when – in upfront and results-based payment agreements – applicants that fulfill the emissions compliance constraint would have an incentive to violate the investment constraint, then investment-based agreements can implement low emissions when upfront and results-based payment agreements cannot. In this case, low emissions could be achieved with higher overall transfers to the applicants than those necessary for compliance with low emissions only. This case, in which investment-based agreements implement more climate cooperation than results-based agreements, is empirically relevant. It applies to situations in which an applicant is small relative to the size of the world, has a low stock of green technology implying high gains from investing in green tech, has high costs for investment in green technology, and would be required to reduce emissions considerably.

Resource transfers can increase the number of countries that commit to low emissions. Our work shows that results-based agreements achieve this goal with the lowest levels of transfers. Hence, as long as results-based agreements can ensure that low-benefit countries comply with high investments in green technology and low emissions, existing member countries of a climate club would prefer them. However, in empirically relevant situations, in which such results-based agreements do not provide sufficient incentives to invest in green technology, investment-based payment agreements may. It is worth giving them a look.



  • Fong, Y. and Surti, J. (2009). The Optimal Degree of Cooperation in the Repeated Prisoners’ Dilemma with Side Payments. Games and Economic Behavior, 67, 277-291.
  • Harstad, B., Lancia, F., and Russo, A. (2018). Compliance Technology and Self-enforcing Agreements. Working Paper, Universities of Oslo, Salerno, and Bergen.
  • Kerr, S., Lippert, S., and Lou, E. (2018). Financial Transfers, Compliance Technology and Climate Cooperation. Mimeo. Motu Economic and Public Policy Research and University of Auckland.
  • MacKay, D.J.C., Cramton, P., Ockenfels, A., and Stoft, S. (2015). Price carbon – I will if you will. Nature 526, 315–316.
  • Rapoport, A. and Chammah, A.M. (1965) Prisoner’s Dilemma: A Study in Conflict and Cooperation. University of Michigan Press.


About the Authors

Suzi Kerr is currently a Senior Fellow at Motu, a Principal Investigator in Te Punaha Matatini, and an Adjunct Professor at Victoria University. She graduated from Harvard University in 1995 with a PhD in Economics. From 1998 to 2009 she was the founding Director of Motu. She has also worked at the University of Maryland at College Park, Resources for the Future (USA), the Joint Center for the Science and Policy of Global Change at MIT, and has been a Visiting Professor at Stanford University and at the University of the Andes in Bogotá, Colombia.  She is a member of the Air New Zealand Sustainability Panel, President of the Australasian Agricultural and Resource Economics Society, and a member of New Zealand’s Interim Climate Change Committee. She has led and co-led several large international projects including the writing of the World Bank and ICAP handbook on emissions trading ‘Emissions Trading in Practice’.
Steffen Lippert is a Senior Lecturer in the Department of Economics at the University of Auckland and an Associate Investigator in Te Punaha Matatini. He studied Economics at Mannheim, Berkeley, and Toulouse and received his Ph.D. in Economics in 2005 jointly from the Toulouse School of Economics and the University of Mannheim. His research applies microeconomic theory to topics in Industrial Economics, Innovation, and the Economics of Social Interaction and Networks. Steffen is a co-founder of the Asia-Pacific Industrial Organisation Society and member of its Scientific and Executive Committees.

Edmund Lou is currently a research analyst at Motu Economic and Public Policy Research, Wellington, New Zealand. He will commence his PhD in Economics at Northwestern University in September. His research focuses on environmental economics and game theory.





Developed by Paolo Gittoi