P REFACE
Rising sea levels, prevailing droughts, increased occurrence of flooding and hurricanes: the future prospects are not that bright. Scientific reports, articles and scientists themselves indicate that we have to do something to prevent this scenario from becoming reality. As a political response, the Kyoto protocol was set up to reduce the human induced effects on climate. However, some countries rejected the Kyoto protocol. So, why do countries decide to reject the Kyoto protocol, if ratification is in our collective interest? This question puzzled me and made me decide to search for an explanation. Therefore, I started this research project in which I hypothesize that a country’s decision whether to ratify or reject the Kyoto protocol, resembles a social dilemma.
Hopefully, this study will aid in understanding the real world policy decisions in case of the Kyoto protocol and maybe even in international treaties in general. Moreover, I sincerely hope to contribute to a solution that might bring all countries together in achieving an effective climate policy as well as an effective policy climate. I am convinced that no matter how small my contribution might be, it can make a difference..
I owe some thanks to a number of people, especially my supervisor Laurie Hendrickx. He was able to keep me on the right track. Moreover, he was always willing to give advice regarding this report and the process of writing. Therefore, I am grateful for his coaching. Furthermore, I would like to thank all the people that showed their interest and inspired me by inquiring about the status of my project. This very much increased my own motivation for this project.
Last but not least, I owe some special thanks to my fellow students who supported me with advice along the way.
S AMENVATTING
De mate van klimaatsverandering is afhankelijk van het aantal landen dat het Kyoto protocol ratificeert en navolgt. Het collectieve belang van alle landen is het meest gediend bij het scenario waarin alle landen het Kyoto protocol ratificeren. Desalniettemin zijn er landen die er voor gekozen hebben het Kyoto protocol te verwerpen. Blijkbaar geldt dat voor bepaalde landen het nationaal belang niet gediend is bij ratificatie. De beslissing om het verdrag al dan niet te ratificeren lijkt voor landen overeen te komen met een sociaal dilemma. Sociale dilemma’s zijn keuzesituaties waarin de uitkomsten van de keuze van een land mede afhangen van de keuze die overige landen nemen, en waarin het collectieve belang strijdig is met het individuele belang. Er zijn meerdere typen sociale dilemma’s die essentieel van elkaar verschillen. Zo heeft een land dat geconfronteerd wordt met een zogenaamd
“Prisoner’s Dilemma” een voorkeur voor het verwerpen van het Kyoto verdrag, onafhankelijk van het gedrag van andere landen. Bij een land dat geconfronteerd wordt met een zogenaamd
“Chicken Dilemma” hangt de voorkeur voor ratificatie of verwerpen af van wat alle andere landen besluiten: er is dan een voorkeur om precies het tegenovergestelde te doen als alle andere landen.
Doel van dit onderzoek is het opstellen van oplossingen die er mogelijk aan kunnen bijdragen dat meer landen het Kyoto protocol ratificeren. Aangezien de effectiviteit van de mogelijke oplossingen afhankelijk kan zijn van het type sociale dilemma waar een land mee geconfronteerd wordt stellen we ons de volgende vragen:
Worden landen geconfronteerd met een sociaal dilemma als ze moeten kiezen tussen het ratificeren of verwerpen van het Kyoto protocol?
Worden verschillende landen met verschillende typen sociale dilemma’s geconfronteerd?
Welke landkarakteristieken bepalen of en met welk type sociaal dilemma een land geconfronteerd wordt?
Zijn de oplossingen die voor sociale dilemma’s op individueel niveau effectief zijn, ook effectief en toepasbaar m.b.t. bevordering van ratificatie van het Kyoto protocol?
Om de eerste drie vragen te beantwoorden is er een model opgesteld waarbij wordt aangenomen dat het type keuze waarmee een land geconfronteerd wordt afhankelijk is van vier landkarakteristieken: het ontwikkelingsniveau, de kosten voor het reduceren van broeikasgasemissies, de kwetsbaarheid voor de gevolgen van klimaatsverandering en het aandeel in de mondiale uitstoot van broeikasgassen. Dit model is toegepast op 10 landen die systematisch verschillen ten opzichte van de vier landkarakteristieken. Uit het model blijkt dat de keuze voor het al dan niet ratificeren van het Kyoto protocol voor de meeste maar niet alle landen een sociaal dilemma inhoudt. Voor twee landen betekent de situatie geen sociaal dilemma. Voor acht van de landen is dat wel het geval: zeven landen worden geconfronteerd met een Prisoner’s Dilemma en één met een Chicken Dilemma. In het algemeen blijkt dat een land niet geconfronteerd zal worden met een sociaal dilemma indien zijn kwetsbaarheid voor de gevolgen van klimaatsverandering lager of gelijk is aan de kosten voor het reduceren van broeikasgasemissies. Als de kwetsbaarheid voor de gevolgen van klimaatsverandering van een land veel hoger is dan de kosten voor het reduceren van broeikasgasemissies, zal het geconfronteerd worden met een Chicken Dilemma. Een land dat een middelmatige verhouding tussen de kwetsbaarheid voor de gevolgen van klimaatsverandering en kosten voor reductie van broeikasgassen kent, zal afhankelijk van zijn aandeel in de mondiale uitstoot van broeikasgassen geconfronteerd worden met een Prisoner’s Dilemma of een Chicken Dilemma.
Op basis van bovenstaande bevindingen en literatuur met betrekking tot sociale dilemma’s op individueel niveau, hebben we enkele oplossingen geformuleerd die er voor kunnen zorgen dat meer landen het Kyoto verdrag zullen ratificeren. Enkele van deze oplossingen oefenen een verschillende invloed uit op de diverse typen sociale dilemma’s. Echter, uit een analyse blijkt dat deze oplossingen niet tot de meest effectieve oplossingen behoren. Er zijn drie oplossingen die daarentegen wel een hoge potentie hebben, voor alle landen. Ten eerste het
tweede effectieve oplossing, die op meerdere factoren een gunstige invloed heeft, is het onderverdelen van landen in subgroepen. Dit verhoogt de doeltreffendheid van deelname van een land alsook de groepsidentiteit en het geloof in toekomstige (wederkerige) confrontaties.
Ten derde kunnen deze factoren verder gestimuleerd worden door het uitbreiden van het Kyoto verdrag met andere (milieu)verdragen. Een combinatie van deze drie oplossingen (extra onderzoek, onderverdelen van landen in groepen en uitbreiden van het Kyoto protocol), zal op basis van onze bevindingen het meest kansrijk zijn om meer landen te laten ratificeren.
S UMMARY
The seriousness of the negative consequences of climate change depends on the number of countries that decide to ratify and implement the Kyoto protocol. The collective interest is best served by the scenario in which all countries decide to ratify. However, there are some countries that decided to reject the Kyoto protocol. Apparently, the national interest of these countries is not served by ratification. Therefore, it seems that the decision whether or not to ratify the Kyoto protocol constitutes a social dilemma for some countries. Social dilemmas are decision situations in which the outcome of the decision also depends on what other people do and in which individual and collective interest are in conflict. There are different types of social dilemmas, that differ fundamentally. For instance, a country that is confronted with a so-called Prisoner’s Dilemma will prefer to reject the Kyoto protocol, independent of the behaviour of other countries. The optimal strategy for a country that is confronted with a so-called Chicken Dilemma depends on the decisions by other countries. Such a country prefers to do the opposite of what all other countries decide to do.
Goal of this research is to identify solutions to promote ratification of the Kyoto protocol.
Since the effectiveness of such solutions may depend on the type of social dilemma countries are confronted with, we formulated the following research questions:
Are countries confronted with a social dilemma when they have to decide between ratification or rejection of the Kyoto treaty?
If countries are confronted with a social dilemma, what type of social dilemma would that be?
Which country characteristics determine if and what type of social dilemma a country is confronted with?
Are the solutions to promote cooperation in social dilemmas, that have been shown to be effective at the level of individuals, also applicable and effective for promoting ratification of the Kyoto protocol?
To answer the first three questions, we created a model. This model implies that the choice whether or not to ratify the Kyoto protocol is dependent on four country characteristics: the country’s developmental status, its costs for abating greenhouse gas emissions, its vulnerability to climate change and its fraction in the global greenhouse gas emissions. The model is applied to 10 countries that vary systematically on these four country characteristics.
It turns out that the choice whether or not to ratify the Kyoto protocol constitutes a social dilemma for most, but not all countries studied. For two countries the choice does not constitute a social dilemma at all. Seven of the other eight countries are confronted with a Prisoner’s Dilemma and only one country with a Chicken Dilemma. In general it turns out that a country will not be confronted with a social dilemma if its vulnerability to climate change is lower or equal to the costs for abating greenhouse gas emissions. If a country’s vulnerability to climate change is much higher than the costs for abating greenhouse gas emissions, it will be confronted with a Chicken Dilemma. A country that has an intermediate ratio between the vulnerability to climate change and the costs for abating greenhouse gas emissions will be confronted with a Prisoner’s Dilemma or a Chicken Dilemma, depending on the country’s fraction in the global greenhouse gas emissions. On the basis of these findings and literature regarding social dilemmas at an individual scale, we formulated solutions that can possibly promote ratification of the Kyoto protocol. Some of these solutions affect decisions in the different types of social dilemmas differently. Our analysis shows that these solutions are not amongst the most effective solutions to promote ratification of the Kyoto protocol. There are three solutions that do show a high potential to promote ratification, for all countries. Firstly, additional research may decrease the environmental uncertainty. A second potentially effective solution is dividing the countries into subgroups, which may have a favourable effect on the country’s efficacy, group identity and the believe in future reciprocal exchanges. Thirdly, these factors can be stimulated even more by extending the Kyoto protocol with other (environmental) treaties. Therefore, a combination of these three solutions probably will have the highest potential to promote ratification of the
T ABLE OF CONTENTS
Samenvatting...3
Summary ...5
1. Introduction ...9
1.1. The Kyoto decision: a social dilemma? ...9
1.2. Modelling the Kyoto decision... 10
1.3. Solutions to promote ratification of the Kyoto protocol ... 11
2. Social Dilemma Theory and Solutions to Social Dilemmas... 13
2.1. Social dilemmas & Games ... 13
2.1.1. The Prisoner’s Dilemma Game... 13
2.1.2. The Deadlock Game... 15
2.1.3. The Chicken Dilemma Game... 17
2.1.4. The Assurance Game... 18
2.1.5. Summary ... 19
2.2. Solutions to Social Dilemmas on individual scale... 20
2.2.1. Changing the payoff structure... 20
2.2.2. Reducing environmental uncertainty ... 20
2.2.3. Influencing expectations about others’ behaviour... 20
2.2.4. Increasing efficacy ... 21
2.2.5. Communicational solutions... 21
2.2.6. Improving group identity ... 22
2.2.7. Increase identifiability of the behaviour... 22
3. Modelling the Kyoto-choice as a social dilemma... 23
3.1. The Kyoto protocol as a social dilemma... 23
3.2. Country Characteristics & Country Choice ... 25
3.3. Values assigned to categories... 26
3.4. Results ... 27
3.5. Country characteristics that determine the decision structure ... 32
4. Applying individual level Social Dilemma solutions to promote ratification of the Kyoto protocol ... 35
4.1. Solutions to promote ratification of the Kyoto protocol ... 35
4.1.1. Changing the payoff structure... 35
4.1.2. Reducing environmental uncertainty ... 35
4.1.3. Influencing expectations about others’ behaviour... 36
4.1.4. Increasing Efficacy... 36
4.1.5. Communicational solutions... 37
4.1.6. Improving group identity ... 37
4.1.7. Increasing identifiability of the behaviour ... 38
4.2. Assessment of the most effective solutions... 38
5. Conclusion & Discussion... 41
5.1. Summary of findings... 41
5.2. Critical Evaluation... 42
5.3. Implications & Suggestions ... 44 Appendix 1 : Annex I countries ...I Appendix 2 : Abatement Targets ...I Appendix 3 : Sensitivity Analysis ...I
1. I NTRODUCTION
The environmental assessments of the Intergovernmental Panel on Climate Change (IPCC) provide proof that global warming is occurring (IPCC, 2001a; IPCC, 1995; IPCC, 1990).
Nowadays there is general agreement that the average surface temperature on earth has increased by about 0.6 °C over the last century (IPCC, 2001b). Data obtained from ice cores, tree rings and mountain glaciers showed that this is part of an anomalous, abrupt change in climate (IPCC, 2001b; Mann, Bradley, & Hughes, 1999). There is increasing evidence that the abrupt change is mainly caused by human activities (Alley et al., 2003; IPCC, 2001b;
Crowley, 2000). The main reason for this anthropogenic induced climate change are emissions of GHG, which have been increasing since 1750 and continue to do so (Tett, Stott, Ingram, Mitchell, & Allen, 1999). The consequences of these increases are quite substantial, see e.g. Barret (2003), IPCC (2001b).
In order to reduce global warming the Kyoto protocol was established. The participating countries agreed to a cut-back of the GHG emissions by the developed countries for the period between 2008-2012. These developed (Annex I1) countries had to decrease their emissions in such a way that the total annual emissions over the period 2008 to 2012 would be 5% lower than they were in 1990. Future prospects about how to carry on after the first stage of the Kyoto protocol in 2012, are still subject to negotiations and studying (UNFCCC, 2005).
Although the importance and urgency of the Kyoto protocol is unmistakable, several countries decided to reject the treaty. Presently, about 40% of global GHG emissions is still not subdue to emission abatement in the Kyoto protocol. This low participation in the Kyoto protocol is in contrast with a similar environmental treaty, the Montreal treaty. This treaty successfully achieved the aim of reducing the emissions of CFC’s (Newchurch et al., 2003) and was characterized by a high participation rate. This leads to the question why there is such a low participation rate in case of the Kyoto protocol.
One of the reasons why countries reject the Kyoto protocol, is that the decision whether to ratify or reject it constitutes a social dilemma. Basically, social dilemmas are situations in which the outcome of one’s decision also depends on what other people do, and in which individual and collective interest are in conflict (Messick & Brewer, 1983). Social dilemmas have been studied intensively, mainly in the fields of Psychology and Economy, to create understanding of the decisions people make. Therefore the lion’s share of research on social dilemmas concerns decisions by individuals. Our study will use this knowledge of social dilemmas on individual scale to create understanding of the decisions countries make regarding the Kyoto protocol.
1.1. The Kyoto decision: a social dilemma?
That the choice whether to ratify or reject the Kyoto protocol can constitute a social dilemma will be demonstrated by the following example of the USA. Basically, the USA has two options: to ratify or to reject the Kyoto protocol. To simplify the decision situation, let us assume that all other countries have the same two options and act together as a group. This results in four possible outcomes, which will be discussed here briefly.
One possible outcome is that the USA ratifies the Kyoto protocol and the other countries act likewise. Against the loss to the US economy due to the interventions required to reduce GHG emissions (US Department of Energy, 1998), stand the relative gains for the global environment since ratification by the USA and all countries will probably lead to a decreasing chance of climate change. When the USA decides to ratify the Kyoto protocol and the other countries reject it, USA both encounters a loss to the economy and a loss to the global environment since the other countries continue polluting. If the USA rejects the protocol and
the other countries ratify it, the US economy will be spared. However the other countries do commit to cut back their emissions and decrease the chance on climate change. Would the USA and the other countries reject the Kyoto protocol, then all economies will be spared. But then the environment will have to pay the price: the chance on climate change will increase by not abating the GHG emissions.
The choice the USA is confronted with shows the correspondence with a social dilemma, being a situation in which:
a. the outcome of one’s decision also depends on what other people do
The example of the USA shows the interdependency between countries. The consequences of the country’s decision whether to ratify or reject the Kyoto protocol depend on the actions of the other countries.
b. individual and collective interest are in conflict
From the collective perspective the best scenario is the one in which all countries choose to ratify the Kyoto protocol. The worst outcome is when all countries reject the treaty. From an individual perspective the country’s best outcome is to reject when the other countries ratify2. Because in this scenario the country itself does not have to pay the price for its own GHG abatement but it actually benefits of the fact that all the other countries do abate their emissions. The worst outcome occurs if the USA abates its emissions when the other countries do not act likewise. In that case the particular country has both the costs of abating and the costs by climate change.
This example of the USA also shows that there are country characteristics that influence the choice whether or not to ratify the protocol. For instance, the costs for abatement of GHG seems to be an important determinant of the choice of ratifying the treaty (see e.g. US Department of Energy (1998)). Another influential country characteristic is the vulnerability to the consequences of climate change: a country with a high vulnerability will probably have a stronger incentive to ratify. Since such country characteristics differ between countries (IPCC, 2001b) and since there are different types of social dilemmas (see next chapter), it is possible that different countries will be confronted with different types of social dilemmas.
1.2. Modelling the Kyoto decision
Although the large majority of studies on social dilemmas focused on individual behaviour in social dilemmas (Kollock, 1998; Messick & Brewer, 1983), there are some studies that explicitly addressed social dilemma situations at the scale of countries. For instance, Barrett (2003) created a model to capture the social dilemma structure of environmental treaties.
Barret’s “Treaty Participation Game” splits a two-player dilemma game in three independent stages. In stage 1 the countries choose whether or not to sign a treaty. Then in stage 2 the signatories choose whether to pollute or abate. The last stage is stage 3 in which the non signatories have to make a choice between polluting and abating. This implies that countries that choose to participate in a treaty in the first stage can collectively choose to continue polluting in the second stage. This is not the case for the Kyoto protocol, since the goal of emission-reduction was already settled before countries decided whether or not to ratify.
Barrett’s model is therefore not suitable for modelling a country’s choice whether or not to ratify the Kyoto protocol. Research by Lise & Tol (2004) focussed at coalition making within the Kyoto protocol. They created a model to analyze which coalitions between 4 main players (USA, EU, Japan and Russia) would be stable in the case of the Kyoto protocol.
Both studies illustrate the importance of social dilemma theory to environmental treaty
Basically, they assume that these countries are all confronted with the same dilemma at one time. Contrarily, in our study it will both be investigated if the decision whether or not to ratify the Kyoto protocol constitutes a social dilemma, and, if so, whether different countries are being confronted with different types of social dilemmas. Therefore a model will be created, that incorporates a number of country characteristics. This model will be presented in Chapter 3.
1.3. Solutions to promote ratification of the Kyoto protocol
The disappointing participation in the Kyoto protocol asks for solutions to promote ratification. The knowledge on individual social dilemmas might help in this case. Several studies showed that social dilemma theory can provide solutions to real world (environmental) problems, e.g. Joireman et al. (2001); Van Lange, Van Vugt, & De Cremer (2000); Van Vugt & Samuelson (1999). It is conceivable that the solutions to social dilemmas that are effective at the level of individuals are also applicable and effective at the level of countries. Question is whether these solutions can be used to promote ratification in case of the Kyoto protocol, which is our concern. To answer this question, the solutions on individual scale will be inventoried and translated into solutions applicable to the Kyoto protocol.
Research also indicated that the effectiveness of solutions can differ for different types of dilemmas (Messick & Brewer, 1983). So the proposed solutions will also be judged according to their effectiveness to these different types of social dilemmas.
In sum, the main issues discussed in the introduction result in the following four questions:
Are countries confronted with a social dilemma when they have to decide between ratification or rejection of the Kyoto treaty?
The choice of a country whether to ratify the Kyoto protocol has both characteristics of social dilemmas: a) the outcome of a country’s decision also depends on what other countries do and b) individual and collective interest are in conflict. The first aim of this study is to analyze whether this choice of ratifying the Kyoto protocol truly is a social dilemma or not.
If countries are confronted with a social dilemma, what type of social dilemma would that be?
When the model depicts a country as being confronted with a social dilemma, the next question to ask is what kind of dilemma this is for the country concerned. What type of dilemma countries are confronted with depends on several national characteristics that may differ across countries. We will develop a model that represents a country’s choice whether or not to ratify the Kyoto protocol. The model will include several characteristics that may affect the countries’ decisions, for instance the costs of complying to the Kyoto protocol, the vulnerability to climate change and the national contribution to global GHG emissions.
Which country characteristics determine if and what type of social dilemma a country is confronted with?
The next step is to look how country characteristics (such as the costs of complying and vulnerability to climate change) determine the dilemma structure.
Are the solutions to promote cooperation in social dilemmas, that have been shown to be effective at the level of individuals, also applicable and effective for promoting ratification in the Kyoto protocol?
Finally, solutions to promote cooperation in social dilemmas have been mainly researched on the level of individuals but these may possibly be relevant at the level of countries as well.
The solutions to promote cooperation in social dilemmas, as described for individuals, will be analyzed on their effectiveness and feasibility at the level of countries. The feasibility and effectiveness of solutions may vary across different types of social dilemmas and will be discussed also.
The next chapter will present an overview of relevant key elements of social dilemma theory and several types of social dilemmas that may be relevant to Kyoto decisions. Furthermore, relevant research on social dilemmas and proposed solutions to individual social dilemma situations in literature will be discussed. Results of this literature search will be presented in Chapter 2.
Next, we will develop a model that represents the choice of a country whether or not to ratify the Kyoto protocol. The model will be applied to several cases, consisting of a number of countries that vary on four specific country characteristics. The findings derived from the model will be used to describe which countries are confronted with a social dilemma and if so what type of a dilemma this is. Next, the results will be used to analyze how the different country characteristics affect the payoff structure, and therefore the type of social dilemma, for these countries. The model and results of these analyses will be presented in Chapter 3.
In Chapter 4, we will describe to what extend the solutions described for social dilemmas on individual scale are applicable to the case of the Kyoto protocol. This will be followed by an assessment of the most feasible and effective social dilemma solutions to promote ratification in the Kyoto protocol.
Finally, the conclusions and discussion of the results of this study will follow in Chapter 5.
In this chapter the results as well as their implications will be discussed.
2. S OCIAL D ILEMMA T HEORY AND S OLUTIONS TO S OCIAL
D ILEMMAS
There are different types of social dilemmas, see e.g. Liebarand (1983). As we will show in Chapter 3, the decision whether or not to ratify the Kyoto protocol may have four different decision structures. In the first part of this section we will introduce and explain these four different types in detail. In the second part of this chapter we will present an overview of key factors and solutions to social dilemmas on individual scale, as described in the literature.
This overview will aid us in the search for solutions to overcome the social dilemmas as found for the Kyoto protocol, which will follow in Chapter 4.
2.1. Social dilemmas & Games
Messick & Brewer (1983) describe social dilemmas as situations in which collective and individual interest are in conflict and in which the outcome of one’s actions depends on the actions of others. Social dilemma situations are often studied by “games” representing the decision, e.g. Colman (1995); Kollock (1998). These games are simplified examples of social dilemmas and are used in order to study mechanisms that influence social dilemmas. In general, game theory is a branch of applied mathematics that studies the logic of decision making in social interactions. Based on payoff structures we can classify many different games, e.g. Fraser & Kilgour (1986). Given an interactive situation between two players that both have two options, there would already be 576 different games possible, of which are 78 strategically different (Rapoport & Guyer, 1966). A structural classification of these games is given by Harris (1969). However, only four of these 78 games will prove to be applicable to the case of the Kyoto protocol (see Chapters 3 and 4): these are the Prisoner’s Dilemma Game, Deadlock Game, Chicken Dilemma Game and Assurance Game. As will be explained later on only three of these games have a social dilemma structure. The Deadlock Game has no social dilemma structure but is an example of a payoff structure that is well represented in real life and will play an important role in the remainder of this study. All four games will be explicitly discussed in the following sections.
2.1.1. The Prisoner’s Dilemma Game
A Prisoner’s Dilemma Game is a social dilemma situation in which all players have the incentive not to cooperate, but collectively they would be better off by cooperating. An example of a 2-Player Prisoner’s Dilemma Game can be described as follows: A quizmaster gives two players one euro and the choice of putting it in an envelope (cooperate) or keeping it (defect). Next, each envelope is handed over to the other person after the content has been doubled by the Quizmaster. There are four possible outcomes of this game and each outcome has a particular payoff to both players. The total collection of outcomes with the accompanying payoffs for both players is called the payoff structure, which is demonstrated in Table 2.1. The numbers in the cells indicate the players preference on an ordinal scale: 3 is most preferable, 0 is least preferable. The number in a cell’s bottom left corner represents the payoff to Player 1. Similarly, the number in a cell’s upper right corner represents Player 2’s payoff.
Table 2.1: Payoff structure of a 2-Player Prisoner’s Dilemma Game Player 2
Cooperate Defect
Cooperate 2
2 0
3
Player 1
Defect 3
0 1
1
The outcomes of the Prisoner’s Dilemma Game can be ranked according to the individual interest of the players:
• The highest payoff for an individual player is for the outcome in which the player him/herself defects while the other cooperates. The behaviour of two players, and thus the outcomes for each player, are summarized as follows: the first letter indicates a player’s own decision (C=cooperate, D=Defect), while the second letter indicates the other player’s choice (C or D). So, DC means that Player 1 defects while the other player cooperates. In this case DC=3, the player earns €3.
• The second best outcome is when both players cooperate, referred to as mutual cooperation (CC=2). The player earns €2.
• The next best outcome is when a player defects and the other player also defects, i.e.
mutual defection (DD=1). The player earns €1.
• The least desirable option for each single player is to cooperate when the other defects (CD=0). The player earns nothing.
In sum the players’ order of favourability is DC>CC>DD>CD. This order of favourability characterises a social dilemma structure. When Player 2 cooperates the best option for Player 1 is to defect, resulting in a €3 euro payoff for Player 1. When Player 2 defects, Player 1 is still better of defecting (with a €1 payoff) than by cooperating (which will leave him a €0 payoff). Consequently Player 1 is better of defecting, no matter what the other player does.
Thus, defecting is the optimal strategy for Player 1 in this game. Because the same reasoning applies to Player 2, both players will defect if they want to maximize their individual outcome. This outcome (mutual defection) will leave both players with a payoff of €1. If, instead, both of them would have cooperated, they would have received €2 each. This outcome is better for both players and optimal from the collective perspective since there is no other outcome that allocates a higher payoff to both players. That is what makes the Prisoner’s Dilemma Game such an interesting social dilemma: individually rational behaviour leads to an outcome that is not the collectively optimal outcome.
Whereas Table 2.1 depicts a two player Prisoner’s Dilemma Game, Figure 2.1 represents the multiplayer version of this type of social dilemma. This figure shows the attractiveness of the different outcomes for a player as a function of all other players’ actions. The horizontal axis represents the percentage of cooperating players. The left-most point represents a situation in which none of the other players cooperates (0% cooperation), whereas the right-most point represents a situation in which all other players cooperate (100% cooperation). The two lines represent Player 1’s payoff for defection and cooperation, assuming a linear dependency between cooperation by others and the personal payoff. Firstly, this figure shows that the individual attractiveness is always higher for defection than for cooperation (i.e. the line for defection is always higher than the line for cooperation). Secondly it demonstrates that the outcome of complete defection (line D, 0% cooperation) is less attractive than the outcome of full cooperation (line C, 100% cooperation).
% of other players that cooperate
0% 100%
Attractiveness to player (payoff)
0 1 2 3
Cooperate Defect
Figure 2.1: Multiplayer Prisoner’s Dilemma Game
In sum, the most important characteristics of a Prisoner’s Dilemma Game are the following:
• Defection is the dominant strategy. The individually optimal choice does not depend on the other player’s actions. Defection will always result in a higher individual payoff independent of what the other players do. This is shown in Figure 2.1 where the line for cooperation is always higher than the line for defection.
• When Player 1 chooses to defect and all other players do so (i.e. 0% of other players cooperate), the individual payoff will be lower than it would have been if all players had cooperated.
• The Prisoner’s Dilemma Game is a social dilemma since: a) the personal payoff of one’s decision also depends on what other people do, and b) individual and collective interest are in conflict: the optimal individual choice is defection but the collective interest is best served by mutual cooperation.
2.1.2. The Deadlock Game
A Deadlock Game is a situation in which the dominant strategy is to defect, leading to the optimal collective outcome. An example of a 2-Player Deadlock Game can be described as follows: Suppose there are two adjacent neighbours each with one tree in their garden. The trees represent a relatively high value to their property, since it provides a comfortable shade during the summers. But when neighbour A would cut down his tree (cooperate), this would result in a better view for neighbour B. So the personal loss of A’s tree would result in a relatively small benefit to his neighbour B, since B’s property increases in value by having a better view. The same reasoning applies for the case in which B cuts down his tree. When both of them decide to cut down their trees (cooperate), both of their properties loose in their values by lacking any trees even though they get a better view in return. When both of them refuse to cut down the trees (defect) the values of their properties will remain the same as before. The payoff structure of a 2-Player Deadlock Game is demonstrated in Table 2.2.
Table 2.2: Payoff structure of a 2-Player Deadlock Game
Player 2
Cooperate Defect
Cooperate 1
1 0
3
Player 1
Defect 3
0 2
2
The ranking of the outcomes according to the players’ individual interest within a Deadlock Game is as follows:
• A player obtains the highest payoff by defecting (D: leave own tree standing) when the other cooperates (C: cut down his tree). This outcome is labelled (DC).
• The second best payoff is obtained when both players defect, i.e. leave their trees standing (DD).
• The next best outcome is the one in which both cooperate, i.e. cut down their trees (CC).
• A player’s worst outcome is cooperating when the other player defects, i.e. I cut down my tree but my neighbour leaves his tree standing (CD).
So the individual order of favourability to both players is: DC>DD>CC>CD.
The Deadlock Game does not constitute a social dilemma because there is no conflict between the individual and collective interest. Just like in the Prisoner’s Dilemma Game a player’s optimal strategy is to defect, since defection will always result in a higher payoff than cooperation (no matter what the other player does). So when both players aim to maximize their individual payoff they will choose to defect, which results in the equilibrium of mutual defection (DD). But in contrast to the Prisoner’s Dilemma Game this outcome also yields the highest collective payoff (2+2=4). This means that there is no conflict between the individual and collective interest.
The multiplayer version of the Deadlock Game is presented in Figure 2.2. The absence of conflict between the individual and collective interest can also be derived graphically from this figure: there is no single point on the cooperation line that is higher than any other point on the defection line.
% of other players that cooperate
0% 100%
Attractiveness to player (payoff)
0 1 2 3
Cooperate Defect
Figure 2.2: Multiplayer Deadlock Game
In sum, the most important characteristics of a Deadlock Game are:
• The individual best strategy is to defect, since this will result in the highest personal payoff for all possible outcomes.
• The Deadlock Game is not a social dilemma since individual and collective interests are not in conflict: the individual’s best strategy is to defect and mutual defection also results in the optimal collective outcome.
2.1.3. The Chicken Dilemma Game
A Chicken Dilemma Game is a situation in which the individual interest is best served by doing the opposite of what the other player(s) is/are doing. The following example will illustrate this. Suppose there are two car drivers that play a game: they drive towards each other with full speed. Both players have the option of swerving in order to avoid a collision which is considered as “being chicken” (C, which stands for cooperation) or to drive ahead (D, which stands for defection) with the possibility of a collision. This game provides the following possible outcomes:
1. Both players resolutely drive ahead (DD) and crash into each other. This is the worst outcome of the game and consequently both players obtain the lowest payoff (0,0).
2. One of the players swerves and the other player holds on to his line of driving (CD or DC). The player that swerved experiences a low payoff (1) for being considered
“chicken”. However, an accident is avoided, so he obtains a higher payoff than for the first outcome. Player 2 on the other hand holds on to his line of driving and experiences the highest payoff (3) since he is now considered to be brave.
3. Both players chicken out by swerving (CC), which is considered a draw with intermediate payoffs for both players (2,2)
The payoff structure of a Chicken Dilemma Game is shown in Table 2.3.
Table 2.3: Payoff structure of a 2-Player Chicken Dilemma Game
Player 2
Cooperate Defect
Cooperate 2
2 1
3
Player 1
Defect 3
1 0
0
The ranking of the outcomes according to the players’ individual interest within a Chicken Dilemma Game is as follows:
• The best outcome is the one in which the player defects and the other player cooperates (DC).
• The second best outcome is when both cooperate (CC).
• The next best outcome is when the player cooperates and the other defects (CD).
• The worst outcome is when both players defect (DD).
So the individual order of favourability within a Chicken Dilemma Game is DC>CC>CD>DD.
The Multiplayer representation of the Chicken Dilemma Game is presented in Figure 2.3.
This figure demonstrates that when all other players cooperate player 1 is better off defecting When all players defect (i.e. 0% cooperates) he is better off cooperating. So in a Multiplayer Chicken Dilemma Game the relatively highest payoff can be obtained by doing the opposite of what the majority of the other players does.
% of other players that cooperate
0% 100%
Attractiveness to player (payoff)
0 1 2 3
Cooperate Defect
Figure 2.3: Multiplayer Chicken Dilemma Game
In sum, the Chicken Dilemma Game has the following characteristics:
• Defection is not a dominant strategy. As we can see from Figure 2.3, the line for defection is not always higher than the line for cooperation.
• Defection is a risky strategy since it may result in a very low or very high individual payoff.
• It is a social dilemma since it meets both criteria: individual and collective interests are in conflict and a player’s payoff depends on the actions of all other players.
2.1.4. The Assurance Game
An Assurance Game3 can be summarized as a situation in which individual interest is best served by acting the same as the other player(s). An illustrative example of an Assurance Game is a situation in which two athletes participate in a marathon to settle which one of them is the strongest runner. Both prefer not to use doping (cooperate) instead of using doping (defect). However, when one athlete uses doping, the other athlete is better of by using doping too. In that case the athlete keeps a chance of winning instead of certain loss against a drugged athlete. The Assurance Game payoff structure is demonstrated in Table 2.4:
Table 2.4: Payoff structure of a 2-Player Assurance Game
Player 2
Cooperate Defect
Cooperate 3
3 0
2
Player 1
Defect 2
0 1
1
The ranking of the outcomes according to the players’ individual interest within an Assurance Game is as follows:
• The best individual outcome occurs if both players cooperate, i.e. do not use doping (CC).
• The second best outcome occurs if the player himself defects and the other player
• The worst outcome occurs if the player himself cooperates but the other player defects (CD).
The Assurance Game can be summarized as a game with the following order of favourability:
CC>DC>DD>CD. Lack of trust can result in mutual defection as we saw for the athletes that end up with both using doping. Lack of trust thus creates a situation in which individual and collective interests are in conflict and hence a social dilemma.
The Multiplayer representation of this game is shown in Figure 2.4. This Figure demonstrates that a player is better off defecting when the majority of other players defects. When the majority of other players cooperates the player is better of cooperating. So a player can obtain the highest payoff by doing the same as the majority of people is doing.
% of other players that cooperates
0% 100%
Attractiveness to player (payoff)
0 1 2 3
Cooperate
Defect
Figure 2.4: Multiplayer Assurance Game
In sum, the Assurance Game has the following characteristics:
• Defection is not a dominant strategy in an Assurance Game.
• Cooperation in an Assurance Game is a risky strategy since it can lead to a very high or very low individual payoff.
• A player’s best strategy depends on what the other players do. The best individual strategy is to do whatever most other players do.
• Trust is a key concept in the Assurance Game, since in this case lack of trust creates a social dilemma.
2.1.5. Summary
A player’s optimal individual choice strategy in the four games discussed in the previous sections can be summarized as follows. In a Prisoner’s Dilemma Game defection is the optimal individual choice, independent of the others’ actions, but it leads to a non-optimal collective outcome. In a Deadlock Game defection is the optimal individual choice and also leads to the optimal collective outcome. In a Chicken Dilemma Game the optimal individual choice depends on the others’ actions: the individual interest is best served by doing the opposite of what most others are doing. In an Assurance Game the optimal choice also depends on the others’ actions: individual interest is best served by acting the same as most others.
Of these four games only three constitute real social dilemmas: The Prisoner’s Dilemma Game, the Chicken Dilemma Game and the Assurance Game. With respect to the Kyoto protocol it is important to know whether or not a country is confronted with a social dilemma in order to come up with relevant solutions. If a country is confronted with a Prisoner’s Dilemma, Chicken Dilemma or Assurance Game there could be general solutions to overcome these social dilemmas. However, the strategic differences between these different dilemmas may affect the efficiency of different solutions. The several solutions that have
2.2. Those solutions on individual scale may be of significant importance on solving social dilemmas on national scale in case of the Kyoto protocol, which will be discussed in Chapter 4.
2.2. Solutions to Social Dilemmas on individual scale
Several studies have investigated the factors affecting behaviour in social dilemmas. The majority of these studies concern the behaviour of individuals (from a psychological perspective), groups (sociological perspective) or companies and institutions (economical perspective). A brief overview of these studies can be found in Jager (2000), extensive reviews by Kollock (1998), Liebrand (1989) and Messick & Brewer (1983) provide additional information. The research discussed in these studies has identified several factors that affect social dilemma behaviour and various possible solutions for social dilemmas. These solutions are structural interventions that promote cooperative behaviour in social dilemma situations.
2.2.1. Changing the payoff structure
The greater the personal return from cooperation and the lower the return from defecting, the higher the levels of cooperation will be, e.g. Bonacich, Shure, Kahan & Meeker (1976). In order to promote cooperation the payoff may be changed in two ways. Firstly, by rewarding cooperative behaviour, e.g. subsidizing toxic waste treatment. Secondly, by punishing defective behaviour, e.g. inducing a tax for polluting companies.
This kind of solutions requires the existence of an authority that regulates the payoffs for defection and cooperation. Moreover it is important to keep in mind that the costs for such a sanctioning or rewarding system should remain lower than the benefits that result from cooperation levels. Messick & Brewer (1983) describe several studies that found that when individuals think others are overusing the common resource, they are more willing to change to a system with a superordinate authority.
Another experimental finding is that when the benefits to others from one’s cooperation increase, the levels of cooperation also become higher, e.g. Bonacich et al. (1976). So we somehow do care about others’ payoffs. Few people would accept 500€ with nothing for their friend instead of 495€ for both of them. The individual sacrifice of 5€ somehow counterbalances the other’s individual benefits.
Summarized, changing a game’s payoff structure can promote cooperation. Cooperation has been shown to increase if one’s own or others’ benefits of cooperation increase and if one’s costs or others’ costs of defection increase.
2.2.2. Reducing environmental uncertainty
Wit & Wilke (1998) found that if individuals are more uncertain about the size, quality and growth of a collective resource (environmental uncertainty), cooperation levels decrease. So when people are more uncertain about the scale and growth of a collective resource, they are more likely to overexploit the resource. So cooperation can be promoted by reducing environmental uncertainty. However, Wit & Wilke also suggest that in the case of a public good dilemma, reducing environmental uncertainty will only promote cooperation if the provision point (the amount of contributions that are needed to provide or sustain the public good) is not so high that people lose their motivation to cooperate at all.
2.2.3. Influencing expectations about others’ behaviour
According to Liebrand (1989) most experiments on social dilemmas point out that cooperative players mainly expect cooperative behaviour by the other player(s). Likewise defective players expect the other player(s) to defect, e.g. Tyszka & Grzelak (1976). Work by
likely to be more effective in case of dilemmas in which the optimal choice depends on the decisions by others, e.g. in the Chicken Dilemma or Assurance Game.
An important factor with respect to this solution is the trust one has in the cooperativeness of other group members (Messick & Brewer, 1983). A person may only be willing to cooperate as long as others will act alike. Thus, in this case the trust in other group members will influence the decision whether or not to cooperate. Trust can be increased by the adoption of a
“grim trigger” strategy: each individual only agrees to cooperate when all others will do so (Kollock, 1998). This eliminates the possibility of free riding by individuals but it is risky since it can result in the best or worst collective outcome (all or nothing). In some occasions it may be a solution but on the other hand it can also result in a worse collective outcome.
Moreover this solution strongly depends on the possibility of making conditional choices.
2.2.4. Increasing efficacy
One of the important factors influencing people’s behaviour in social dilemmas is the concept of efficacy (Kerr & Kaufman-Gilliland, 1997). The incentive to defect is higher if individual cooperation does not (or hardly) contribute to achieving the collective goals. For instance, when your free-riding behaviour would have a discernable, immediate effect on public transport you would probably rather pay a bus ticket. Individual contributions for the collective goal often only matter when a substantial amount of others will act accordingly. If an individual has the feeling he can make a difference he will be more encouraged to cooperate. A way to increase efficacy is by “matching grants” (Kollock, 1998). This means that an outsider offers to multiply the contribution of an individual. An example is when a company agrees to double the contribution by others in a fund raising action for a charity.
Another way to increase efficacy is to decrease group sizes. Several studies found that when the number of players increases, relatively more players tend to defect (Bonacich, Shure, Kahan, & Meeker, 1976; Fox, 1977). This effect is attributed to the lower efficacy in larger groups. Decreasing the group sizes or keeping the group sizes as low as possible can thus positively influence cooperation.
2.2.5. Communicational solutions
Research cited in Messick & Brewer (1983) proves that individuals with the opportunity to discuss the dilemma within their group prior to the decision making, make fewer defective choices than individuals without this opportunity. Open communication within groups also increases the probability that individuals will cooperate in a social dilemma. However, communication can only increase cooperation when it concerns what decision to make. In social dilemmas concerning a shared resource pool (resource dilemmas) communication also proved to be an effective solution (Edney & Harper, 1978). Communication can lead to a designation of a set of group members that contribute towards the commons, which has as many members as is necessary to receive the optimal payoff to the group. Research by van de Kragt (1983) shows that when individuals get the chance to communicate in a public goods dilemmas, this often results in designation of a minimal number of people that is necessary to ascertain a collective goal (minimal contributing set). By doing so, the collective interest is maximized.
Moralizing can also be a solution to overcome social dilemmas, according to (literature cited in) Liebrand (1989): Groups that before making their decision to cooperate or defect, discuss the morality of cooperation and immorality of defection show higher levels of cooperation than groups without this possibility.
Another factor that positively influences cooperation is education about the consequences of individual behaviour in social dilemmas. Studies in which the players were informed and educated about what the long term consequences of defection and cooperation in social dilemmas could be, e.g. Van Vugt & Samuelson (1999), showed that education positively influences the willingness to cooperate and leads to a longer maintenance of the common resource pool.
campaigns aimed at long-term decision consequences. These have all been shown to increase cooperation levels.
2.2.6. Improving group identity
Group identity has been found to significantly influence people’s behaviour in social dilemmas. Dividing individuals into groups can be enough to increase an individual’s altruistic behaviour and hence increase the levels of cooperation (Brewer & Kramer, 1986).
Other studies suggest that it is the belief in future reciprocal exchanges that evokes higher cooperation (Kollock, 1998). This belief in future reciprocal exchanges inside the group of players weakens the temptation to defect and encourages cooperation. The expectation of reciprocity might even be so strong that it manifests itself in situations without any possibility of reciprocity, i.e. when the players will probably never meet again. When the group identity and/or the belief in future reciprocal exchanges is increased this might lead to higher levels of cooperation.
2.2.7. Increase identifiability of the behaviour
In social dilemma situations where individuals cannot communicate it is possible to increase cooperative behaviour by increasing the identifiability of the players’ behaviour. When their choices are disclosed publicly, people act more cooperative than when their decisions remain unidentifiable (Fox & Guyer, 1978). However, Kerr (1999) argues that there are several conditions to be met in order to effectively influence individuals’ behaviour by increasing the identifiability. For instance, the likelihood that a defector gets punished for his defective behaviour has to be high.
Summarizing, there are different types of decision structures. The four decision structures of main importance to this study were explained in Section 2.1. These are: the Prisoner’s Dilemma, Deadlock Game, Chicken Dilemma and Assurance Game. In the following chapter it will be investigated whether countries are confronted with a social dilemma, and if so with what type. The type of dilemma a country is confronted with is important to come up with effective and feasible solutions to promote ratification of the Kyoto protocol, which will follow in Chapter 4. The solutions that proved to be effective at individual level were discussed in Section 2.2. It was demonstrated that in order to promote cooperation in social dilemmas there are several solutions. One way to promote cooperation is by changing the payoff structure, i.e. by punishing defective behaviour or by rewarding cooperative behaviour.
Another way to promote cooperation is by reducing the environmental uncertainty. Increasing the efficacy of an individual or the identifiability of his/her behaviour have also been shown to promote cooperation. Influencing the expectations about the behaviour of others, can also increase the levels of cooperation. Furthermore, the literature describes several communicational solutions to social dilemmas, such as better communication within the group, moralizing and education. Moreover, improving the group identity has also been shown to promote cooperation within social dilemmas. These solutions will provide the basis for solutions to promote ratification of the Kyoto protocol in Chapter 4. However, in order to come up with appropriate solutions we first need to know which countries are confronted with social dilemmas, and if so, with what type of social dilemma. In the next chapter a model will be introduced, that will be used to identify the decision structures for a number of countries.
3. M ODELLING THE K YOTO - CHOICE AS A SOCIAL DILEMMA
In Section 3.1 we will introduce a model that describes the decision structure of countries with respect to the ratification of the Kyoto protocol. In Section 3.2 four country characteristics important for decision outcomes will be determined. On the basis of these four country characteristics a set of 10 countries will be selected in such a way that the most important and interesting differences between countries are included. In Section 3.3 the methods for assessing the scores of the countries on the four key characteristics will be discussed. Then, in Section 3.4, the model is used to identify the decision structure for each one of the 10 countries. Finally, in Section 3.5 the model will be used to analyse which of the selected country characteristics are most important for the type of decision structure a country is confronted with.
3.1. The Kyoto protocol as a social dilemma
The Kyoto protocol presents countries with a choice between two options:
1. Reject the Kyoto protocol (defect): the country does not ratify the Kyoto protocol and thus will not abate its GHG emissions. We assume that if a country rejects the Kyoto protocol, the chance of climate change will increase proportionally to the country’s share in the global GHG emissions.
2. Ratify the Kyoto protocol (cooperate): The country ratifies the Kyoto protocol and decreases its GHG emissions. Reducing its GHG emissions will result in a certain amount of abatement costs for the country (see below). By cooperating, the country will decrease the chance of climate change and, again, we assume that this probability decrease is proportional to the country’s share in the global emissions.
In our model, we assume that the outcome of a country’s choice to ratify or not, depends on four key factors, namely:
1. Developmental status of a country (S): The developmental status of a country will serve as a basis level that describes the present status of a country. This present status may change by the choice whether or not to ratify the Kyoto: the country’s costs or benefits resulting from (not) participating in the Kyoto protocol will be subtracted or added to this basis level.
2. Costs involved with ratification of the Kyoto protocol (C): When a country ratifies the Kyoto protocol it will have to abate its emissions. The costs for emission abatement will be subtracted from the basis level given by the developmental status.
3. Consequences of global warming: The consequences of global warming for a particular country can be formulated as the sum of all positive effects minus the negative effects by global warming. We assume that the consequences of global warming are a function of two factors: 1) the intensity of climate change, which in turn is assumed to depend on the total GHG emissions and 2) the country’s vulnerability to climate change. The former factor mainly depends on the extent to which all countries in the world will reduce their GHG emissions, which we call the global GHG abatement. Global GHG abatement depends on the abatement by the country itself and the abatement by other countries. So, basically the consequences of global warming are assumed to depend on the following interactive factors:
1. Intensity of climate change, which depends on the total GHG emissions, which, in turn, depend on:
a) the country’s own decision whether or not to ratify the Kyoto protocol, and the country’s fraction in the total global GHG emissions (O).
b) the fraction of other countries that cooperate (A). This will be operationalized as the emissions by other countries that cooperate divided by the total emissions by other countries. So if all other countries cooperate A equals 1, whereas if no other countries cooperate A equals 0.
2. The country’s Vulnerability to climate change (V)
In the case of cooperation (ratification of the Kyoto protocol), the negative consequences of climate change will be higher if: a) Vulnerability (V) is higher, b) if own emission fraction (O) is lower, and c) if the fraction of other countries that cooperate (A) is lower. If we assume that the consequences of climate change are linearly related to each of these three factors, then the consequences of climate change can be modelled as:
Consequences of climate change = V × (1 – O) × (1 – A)
By combining the above three elements, i.e. developmental status (S), abatement costs (C), and consequences of climate change ( V × (1 – O) × (1 – A) ), a country’s payoff for cooperation can be modelled as in equation 1:
Equation 1:
Payoff (Cooperation) = Developmental Status – Costs for own GHG abatement – Consequences of climate change to the particular country
= Developmental Status – Costs for own GHG abatement – Vulnerability × ( 1 – Own share in global emissions) × (1 – Fraction of other countries that cooperate (A) )
= S – C – V × (1 – O) × (1 – A)
A country’s payoff for defection (rejection of the Kyoto protocol) can be calculated similarly.
We assume that when a country rejects the Kyoto protocol it will not abate its emissions. This implies that a country does not have to pay the costs (C) for abating its emissions. Secondly, if a country decides to defect the consequences of climate change will not depend on it’s own emission fraction. Hence, the country’s costs due to climate change will entirely depend on the fraction of other countries that cooperate (A) and the country’s own vulnerability to climate change (V). Thus the payoff of a country that defects is:
Equation 2:
Payoff (Defection) = Developmental Status – Vulnerability × (1 – Fraction of other countries that cooperate (A) )
= S – V × (1 – A)
In sum: A country’s payoff for cooperation depends on the country’s developmental status, the country’s vulnerability to the consequences of climate change, the country’s own share in global GHG emissions and the fraction of other countries that cooperate. On the other hand, the payoff for defection only depends on the country’s developmental status, its vulnerability and the fraction of other countries that cooperate. Thus, a country’s payoff depends on four country characteristics: the developmental status (S), vulnerability (V), the country’s own share in global GHG emissions (O) and the Costs for own GHG Abatement (C). In addition, a country’s payoff depends on the choice between cooperation and defection, and on the fraction of other countries that cooperate (A). In section 3.4 the model will be used to identify the decision structure a country will be confronted with when deciding whether or not to ratify the Kyoto protocol.
