Essay: Overview of solutions to CAPs in general / climate change in particular

1 Introduction
Climate change is often considered to be one of the most pressing and difficult to solve problems of our time. Broadly speaking, there are two, sometimes competing, ways of combating climate change, namely through technological advances and behaviour change. Proponents of the belief that only technology can save the planet from a climate catastrophe are sometimes known as ‘wizards’ whereas people that believe in the possibility of effecting global behaviour change are ‘prophets’ (Mann, 2018). This essay is concerned with the questions ‘prophets’ face in their attempts to battle global climate change. One key challenge of effecting behaviour change is the existence of so-called collective action problems (CAPs), and climate change is considered to be one of the fiercest (e.g. Esty & Moffa, 2012, p. 777). CAPs were originally theorised by economists who, naturally, assumed humans to be propelled by rational action. The notion that humans are rational actors is called rational choice theory (RCT) or simply rational choice (RC).
In this essay I wish to provide an overview of the solutions to CAPs in general, and climate change in particular, as offered by RCT and discuss potential solutions provided by other approaches, namely bounded rationality (BR) and constructivism. To do so, I will first tackle key issues revolving around the definitions of RCT and CAPs. Secondly, I will present the solutions to CAPs as offered by RCT, namely selective incentives, and argue that whether we can derive solutions to CAPs from RCT depends on the school of RCT we employ. Third and fourth, I will provide definitions of BR and constructivism and describe their solutions to CAPs which are, for instance, nudges and norms in BR and persuasion in constructivism. In a last step, I will summarise my arguments. To illustrate the presented theory, I will, where it is possible, employ the CAP of climate change, and German climate protection policies and actions as a guiding theme for my discussion.
2 Rational choice theory
RC is far from being a unified theory (Eriksson, 2001, p. 6). In the following sections I shall provide a definition of RCT and point to a key theoretical divide within RCT that has major implications for the conceptualisation of solutions to CAPs.
2.1 Shared assumptions
At the heart of RCT lies the universal (e.g. Lichbach, 2003, p. 30) assumption that all humans are rational actors (Eriksson, 2001, p. 16). There are two aspects of rational action all RC theorists agree on. First, rational action denotes the maximisation of expected utility. Utility is the relative satisfaction individuals receive from choosing a certain course of action (Gwartney et al., 2009 p. 10) while its maximisation is achieved if actors, when confronted with a set of alternatives, efficiently and effectively opt for the alternative that best serves their objectives (Olson, 1965, p. 65). Secondly, agents are rational if their preferences in regard to what yields the highest utility are consistent. Consistency of preferences refers to two distinct requirements. First, the requirement of connectedness, meaning that preferences are concisely rank-ordered. Second, the requirement of transitivity, meaning that if option A is preferred to option B and option B preferred to option C, then option A must be preferred to option C (Green & Shapiro, 1994, p. 14).
RCT’s assumption that agents are able to concisely rank their preferences has an important implication. Given the great number of options and combinations humans must posses unlimited computational abilities and complete information (Krugman & Wells, 2013, p. 259). Another implication of RCT is that as rational agents opt for what they believe yields the greatest utility, they will react to incentives. To alter human behaviour incentives can thus be employed, catering to the fact that agents are willing to exploit them to make themselves better off (p. 9).
2.2 Thick vs thin rationality
As stated above, RC is not a unified theory. Here, I will not RCT’s different strands en detail. However, the question of whether all human objectives can be seen as rational marks a shift in RCT itself and has, as we shall see, great implications for this essay’s headlining discussion.
Shedding light on whether all human goals can be rational, Ferejohn (2000) provides a two-part framework. The first is ‘thick’ rationality which is characterised by agents that act rational (in the sense that they maximise utility) but that are also endowed with shared preferences and beliefs that define their goals. This means that agents generally have the same sort of goals (Green & Shapiro, 1994, p. 17), namely “wealth, income, power or the perquisites of office” (Ferejohn, 2000, p. 396). The second part of Ferejohn’s framework is ‘thin’ rationality and it refers to rational behaviour whose goals are not defined by shared beliefs. In the sense of thin rationality, agents are rational when they simply “efficiently employ the means available to pursue their ends” (Ferejohn, 2000, p. 396). Such a liberal definition of rational behaviour denotes, in essence, that the pursuit of all goals and objectives can be regarded as rational and that humans are theoretically able to derive utility from all courses of action. In line with thin rationality, Becker and Murphy (1998), for instance, argue that satisfying the urges of a heroin addiction is rational. Similarly, Riker (1990, p. 173) states that committing suicide is a rational act. This denotes a major shift within RCT as the rationality of human objectives no longer depends on a belief system that defines their rationality.
3 Collective action problems
People have to often act collectively in order to achieve outcomes that lie in their common interest (Olson, 2008, p. 877). The goal of collective action is to provide an economic good. There are four types of economic goods and they can be systemised under consideration of two variables, namely whether a good is excludable and whether it is rivalrous. Excludability refers to whether a good’s consumption is restricted or open to all, while rivalry refers to whether the good’s consumption of one party diminishes a second party’s ability of consumption. Excludable and non-rivalrous goods are private goods whereas excludable and rivalrous ones are club goods (Hindmoor & Taylor, 2015, p. 141). There are two cases in which collective action is compromised or completely fails (Hindmoor, 2006, p. 102), put differently, in which CAPs occur. First, in the case of public goods which are non-excludable and non-rivalrous, so, for instance, national defence and solar energy. Second, common-pool resources are non-excludable but rivalrous. These are, for example, forests and fisheries. Non-excludable goods are subsumed under the term common goods. The line between public goods and common-pool resources is, however, blurred (Hindmoor & Taylor, 2015, p. 142).
Consider stable global temperatures as a public good. When the global population is small, carbon emissions, in other words the consumption of clean air, may be sufficiently absorbed by the environment so that global temperatures stay unaffected. As the population grows, however, more clean air is consumed and larger amounts of emissions are no longer naturally retained. Carbon emissions diminish the amount of stable temperature available, turning them into a common pool whose depletion denotes an increase in temperatures.
When individuals are confronted with the option of whether to contribute to the provision of common goods such as stable temperatures, they may realise that their contribution makes little or no difference at all to the overall amount provided (Hindmoor & Taylor, 2018, p. 43). In such instances, individuals predictively choose to free-ride, a behaviour of non-contribution and defection that relies on others to achieve goals that are in the common, collective interest (Pasour, 1981, p. 453). As individuals can enjoy the benefits from the good being provided without contributing, it is rational for them to defect. Rationality, then, leading people to react to incentives of not providing common goods on an individual level, may result in an undesirable outcome of a good’s non-provision on the collective level (Olson, 1965, p. 21).
There are two main factors determining the success of collective action. First, group size. The larger the group is, the less likely it is for collective action to occur as the benefit captured by each individual group member relatively decreases (Hindmoor & Taylor, 2015, p. 149). Secondly, if group members value the good to different extents, individuals who capture larger amounts good’s benefits have larger incentives to provide the good. Olson identifies three types of groups, each of which are endowed with different probabilities of overcoming CAPs. First, privileged groups that are characterised by at least one group member who values the good to such an extent that he or she is willing to provide the good alone and tolerate free-riding. Privileged groups are expected, by definition, to overcome CAPs. Second, intermediate groups. Defined by members who, whilst not having an incentive to provide the good by their own, are significantly affected by the behaviour of others, are likely to marshal collective action as they have an incentive to monitor and sanction behaviour that is not in the common interest. Third, latent groups are unlikely act collectively. In such groups no member has an incentive to provide the good by their own and the behaviour of one member has no significant impact on others. The group members are unlikely to react and thus marshal collective action (p. 150).
The world’s population can be described as a latent group, which makes climate change a particularly difficult to resolve CAP. As the size of the group, the world population, is over seven billion people, the contribution an individual makes towards the provision of stable temperatures by not emitting green house gases is very small. Therefore, the carbon emissions of one person have a relatively small impact on others which means that the incentives to monitor and sanction others are low but the incentives to free-ride are high. Moreover, as no country can solve climate change on its own (e.g. Ostrom, 2010, p. 1) no country can even hypothetically have an incentive to bear the full costs and provide stable temperatures alone (Hindmoor & Taylor, 2018, p. 47). Olson does not deny the possibility of collective action under such circumstances but states that it can only be effected through, what he calls, ‘selective incentives’ (1965, p. 51).
4 Overcoming collective action problems under the assumption of rational choice
In the next sections I will lead the discussion of what solutions RCT offers in regard to CAPs. Furthermore, I shall demonstrate that whether solutions to CAPs can be found at all, depends on whether we assume thick or thin rationality.
4.1 Overcoming collective action problems in thick rational choice accounts
There are two main roads to resolving CAPs in accordance with the assumptions of thick rationality.
4.1.1 Intervention, regulation and coercion
First, through governance that is either enacted by the state or sometimes non-state actors such as the mafia (e.g. Skarbek, 2011). To overcome CAPs in latent groups, governing structures must, according to Olson (1965), provide selective incentives though intervention, regulation or coercion that induce the provision of common goods. Incentives, potentially altering what agents expect to be their highest utility yielding option, are means through which behaviour change can be effected. In essence, selective incentives are non-common goods in exchange for the provision of common goods. Selective incentives can be either positive or negative. A positive selective incentive is rewarded to those who contribute to the provision of common goods (Hindmoor & Taylor, 2015, p. 152). The German government, for instance, aims to, by offering monetary incentives to buyers, boost sales of e-cars (BAFA, 2018), which are environmentally friendly an, thus, contribute to the proven of stable temperatures. Similarly, firms and private households are guaranteed money by most states and cities in Germany when installing charging ports for e-cars (ADAC, 2019). Negative selective incentives, on the other hand, sanction those who fail to contribute to common goods (Hindmoor & Taylor, 2015, p. 153). In Germany, again, an ecological tax is imposed on gasoline and other forms of fossil energy to enhance energy efficiency and disincentivise energy consumption (BMF, 2018). Also, Germany, having established low-emissions zones (Umweltzonen) in major cities, provides negative incentives in the form of coercion. Cars that surpass a certain threshold of pollution have restricted or deterred access to these designated areas (UBA, 2018). As trespassing is punished, low-emissions zones, by raising the costs of their use, clearly disincentivise people from driving, for instance, older vehicles that are less environmentally friendly.
4.1.2 Privatisation
Second, the state can also do the opposite. Hardin (1965) who theorised the ‘tragedy of the commons’, a depletion of common pools by widespread free-riding on other’s restraint to use a common pool, proposed privatisation as a way of overcoming CAPs. Privatisation turns common-pool resources into either private or club goods, meaning that access to their consumption is rendered restricted. Privatisation, in other words, putting a price on common-pool resources, provides negative selective incentives to those who are unwilling to contribute to a common good. By doing so, CAPs may be solved as restricting access decreases group size. As stated above, smaller groups will predictively marshal collective action more easily.
The solutions to CAPs that Olson and Hardin propose, mostly monetary incentives, are clearly formulated through the perspective of thick rationality. Thick rationality is characterised by goals that, as stated above, concern wealth, income and power. Germany’s climate protection policy also reflects thick rationality as it mostly offers monetary incentives.
4.2 Overcoming collective action problems in thin rational accounts and Popper’s science as falsification
Olson’s assumption of humans being thick-rational is not only supported by the solutions he describes for CAPs but, furthermore, is underlined by his opinion on the effectiveness of selective social incentive. Selective social incentives provide benefits or costs in regard to social standing (Hindmoor & Taylor, 2015, p. 153). Whilst not denying their efficacy entirely, he voices his scepticism regarding their power to resolve CAPs, especially in large groups (Olson, 1965, p. 62). Olson’s stance is coherent with thick rationality as social status as it not subsumed under its assumptions about rational goals. Rather, selective social incentives fall into the domain of thin rationality.
As stated above, due to humans being rational actors, large and latent groups are likely to fail in acting collectively. However, instances of large latent groups marshalling collective action can be observed. One famous example is that even though it is irrational to vote due to the fact that single votes barely contribute to the election of the preferred party or candidate, turnout is usually relatively high, meaning that collective action, paradoxically, and contrary to what we expect from RCT, takes place (e.g. Funk, 2010, p. 1077). One possible explanation for why people vote, even in the absence of laws that make voting mandatory and monetary incentives, are selective social incentives that induce people to vote because of underlying desires of wanting to belong to a ‘community of voters’ which they believe to increase their social status. As established above, thin rationality does not rely on on a shared belief system of what defines rational goals. As long as maximises the agent’s utility, any human goal can be rational, not only the maximisation of wealth, income and power but also, returning to the paradox of voter turnout, social status, the wish to fulfil one’s “civic duty” (p. 1078) or the health benefits that come with waking to the polling station. The assumption of thin rationality, at first glance, can explain a greater number of phenomena than thick RC.
The shift from thick to thin RC comes with a price. By adding “auxiliary assumptions” (Popper, 1963, p. 44) to RCT, thin RC theorists walk in to the trap of engaging in “conventionalists stratagems” (1963, p. 44) or “post hoc theorising” (Green & Shapiro, 1994, p. 34), meaning that theory is expanded in order to accommodate inexplicable phenomena without changing the core of the theory itself. Ultimately, the theory then runs the risk of becoming a ‘theory of everything’. The problem with such theories is that because they are so vague and all events are theoretically explicable (Popper, 1963. p. 44), they are “not refutable by any conceivable event” (p. 48). Thus, they, according to Popper, do not contribute to scientific knowledge which is characterised by systemically approaching truth (p. 44). In order to be scientific, however, theories must be falsifiable as mere verifications are easy to find and thus tell us very little about whether a theory is true or not (Caldwell, 1994, p. 42).
Thin RC, then, as a vague and unscientific theory of everything has a major implication: We loose the ability to both generate explanations and make predictions about human behaviour. This means that, ironically, what we have described above as greater explanatory power (at first glance) denotes the complete loss of such. Conceptualising solutions to CAPs that are successful under the assumption of thin rationality is therefore an impossible task as we no longer posses knowledge about which incentives people will predictably react to. In contrast, thick-rational theorists, by stating that there are just two rational goals, namely money and power, make well-defined assumptions about the drivers of human behaviour. The range of incentives humans will predictively react to is therefore very small, making it easier to select and design them. So, whether we can derive solutions to CAPs from RCT at all, depends on whether we assume thick or thin RCT.
5 Resolving collective action problems outside of rational choice theory
By stating that humans universally act rationally, both thick and thin RCT make an a priori assumption about human behaviour (Bruni & Sugden, 2007, p. 146). Thick rational choice, the stream of RCT from which solutions to CAPs can actually be derived, has, however, a poor empirical record, especially in non-market situations (e.g. Green & Shapiro, 1994). As demonstrated above, this has to do with the assumption of rationality itself which forbids the occurrence of certain events such as high voter turnout. Thus, the question is why people vote even in the absence of incentives to do so. Furthermore, RCT can, of course, not explain all phenomena in market situations either. For instance, the German government, providing monetary incentives for buyers of e-cars, is very unsuccessful in doing so. The government has made subsidies available for 600.000 cars but since the program’s inception not even 20 percent of all subsidies have been given out (Süddeutsche Zeitung [dpa], 2018). Are the incentives too weak or does the consumers’ lack of interest have different reasons? By answering these and related questions scholars and policymakers may discover alternative solutions to CAPs that lie outside RCT. Defending their camp against alternative approaches, RC theorists will invoke the ‘first law of wing walking’ which can be defined as “there is no falsification of theory before the emergence of a better one” (Lakatos, 1978, p. 35). Put differently, before abandoning a theory one must come with a better alternative.
To achieve this task and provide alternatives to RCT, it is necessary to capture the true drivers of human behaviour. Scholars who undertake this challenge have, so far, not come up with a law of human behaviour that possesses the same universality RC does. Rather, their answers are particularistic. This means that they, to use Weber’s terms, move from explanation to understanding. The former is concerned with prediction (Parsons, 2018, p. 83). Thin RC has high predictive power as it stipulates very few goals humans always strive for. The latter, on the other hand, is concerned with the ability to correctly interpret and understand actors’ behaviour (p. 83). The approaches to human behaviour presented in the following sections try to understand rather than explain. This does not mean that no generalised predictions can be derived from their findings. However, to a much lesser extent, as generalising claims can only be made for their particularistic observations (Parsons, 2018, p. 89).
5.1 Bounded rationality
For the past decades RCT has been criticised harshly, not only on the grounds of having a poor empirical record and being partially unscientific (e.g. Green & Shapiro, 1994) but for making unrealistic assumptions about human abilities (e.g. Kanazawa, 1998 [cites Hechter (1992) and Petersen (1994)]) which are unlimited computational capabilities (p. 195) and the possession of complete information (p. 194). Simon (1990), therefore, conceptualised BR, a set of more realistic assumptions of how humans form decisions and therefore behave. Simon showed that humans have, in fact, not unlimited computational capabilities (p. 7) and have incomplete information (North, 1990, p. 23). Rather than maximising utility and following an optimal strategy to meet their objectives, actors satisfice, meaning that they choose options that are ‘good enough’ (Eriksson, 2001, p. 47). BR, however, does not denote stupid or irrational behaviour (Selten, 1999, p. 3).
Being rationally bounded has major implications for trying to find solutions to CAPs. For instance, there is room for human error to be made, meaning that humans can miscalculate the contributions they make towards the provision of a common good and also miscalculate the benefits from pursuing an alternative course of action. Also, North, emphasising the role of institutions which he defines as the rules actors generally live by (1990, p. 3) points out that they, though purposefully designed, shape human thought patterns, rendering humans susceptible to ideas, ideology and norms (p. 76). Furthermore, they have the power to limit knowledge and condition what humans believe to be true (p. 77). The ‘Northian’ model of BR comes with both chances and new challenges for resolving CAPs. On the one hand, norms and ideas, as demonstrated below, can help overcome agents’ rational responses to incentives that compromise collective action. On the other hand, the alteration of what agents believe to be true might lead to not recognising the consequences certain CAPs entail, as can be observed from the phenomenon of the so-called climate change deniers who are often misled by institutions with “vested interests” (e.g. Bain et al., 2012, p. 600).
5.1.1 Behavioural economics
At the heart of behavioural economics lies the assumption of agents’ BR (Jolls, 2015, p. 60). Insights are won through psychological experiments (e.g. Bruni & Sugden, 2007, p. 146). A prominent stream of behavioural economics is concerned with behaviour change. Just recently the Nobel prize for economics was awarded to Thaler for his work on so-called nudges (The Nobel Prize, 2018). Nudges can be defined as “public or private interventions that steer people in particular directions but that also allow them to go their own way” (Sunstein, Reisch & Kaiser, 2018, p. 1). Nudges exploit humans’ bounded rationality to ‘nudge’ them into changing behaviour without having to resort to coercive forces. A famous example is the default nudge. It is employed when agents are offered two or more options, one of them, usually the offering party’s preferred option, being the default setting (Bruttel & Stolley, 2014, p. 767). As agents face costs of acquiring information about alternatives or miscalculate the costs of opting for them, they will often continue with their default option. Having to gather information or miscalculating costs are phenomena which RCT does not accommodate as it assumes agents to have unlimited computational capabilities and complete information.
Behavioural economics in general and nudging in particular have attracted great attention by policy makers in recent years. It is therefore not surprising to see that it has been used to overcome prominent CAPs, such as climate change and global warming (Leggett, 2014, p. 3). The German town of Schönau, for example, has established green energy as the default option for its inhabitants (Sunstein, 2015). The idea is that while it is still possible opt for fossil, potentially cheaper energy, people, displaying a, by RCT inexplicable, tendency towards continuing with default options, will be more likely to consume green energy (Pichert & Katsikopoulos, 2008, p. 66). So, even though consumers have an incentive to defect from providing the common good of stable temperatures by consuming cheaper fossil energy, they, in fact, do contribute due to being bounded in their rationality. Thick rationality cannot accommodate the efficacy of such policies as it predicts perfectly informed agents who are driven by monetary incentives and, therefore, always opt for the least costly option regardless of default settings.
5.1.2 Common-pool governance
Ostrom developed an entirely different way of overcoming CAPs under the assumption of BR (Singleton, 2017, p. 997). She doubted that state intervention and privatisation, as suggested by Olson and Hardin, will necessarily be effective (Olson, 1990, p. 14) by critically asking, for instance, how some common pools such as stable temperatures can even be privatised (p. 13). In her work she, in effect, elaborates on how intermediate groups overcome CAPs to provide common-pool resources even in the face of individual incentives of defection (Hindmoor & Taylor, 2015, p. 155). Much like North, she highlights how the establishment of social norms can help resolve CAPs. Norms, which Ostrom (2005) sees as agents’ culturally embedded prescriptions (p. 17) are not outcome oriented but simply tell agents what to do under certain circumstances. Social norms act, in a sense, as selective social incentives whose compliance is an important determinant of cooperation (Hindmoor & Taylor, 2015, p. 157). Ostrom identifies several properties groups might display that boost or diminish the efficacy of norms to overcome CAPs such as whether monitoring costs are low, whether a group is socially cohesive, whether it is costly to leave the group and whether members believe that their cooperation with each other last for a longer period of time. To understand how CAPs can be resolved it is therefore not only necessary to look at the group’s social traits but also the traits of the economic good and the environment in which it is placed (Hindmoor, 2006, p. 121).
5.2 Constructivism and persuasion
As stated above, thick RCT has, due to its universal assumption about human goals and behaviour, a poor empirical record, especially in non-market situations. RC theorists essentially claim that rationality is human nature (Lichbach, 2003, p. 30). BR, similarly, while stating that humans are not perfectly rational, still adheres to the notion that humans are in pursuit what is ‘good enough’.
Constructivists, in contrast, reject the idea that human nature exists (Lichbach, 2003, p. 30) and make no a priori assumptions about humans behaviour at all (e.g. Dessler & Owen, 2005). Several cultural theorists have aimed to demonstrate that rationality, for instance, is not innate to humans. Polanyi (1992 [1944]) in The Great Transformation shows that human decision making has not always been driven by monetary gains and provides an explanation for how this behaviour was purposefully ‘created’ by political elites in 19th century England. Similarly, Weber (1992 [1905]) explains in his seminal work The Protestant Ethic and the Spirit of Capitalism, how capitalist, that is, rational behaviour, was the byproduct of Calvinists’ search for determining their whereabouts, either paradise or hell, in the afterlife.
Rationality, from a constructivist perspective, is a social construct (e.g. Cabantous, Gond & Johnson-Cramer, p. 400) and constructivists claim that human behaviour is determined by them. Ideas, beliefs and norms can all be defined as social constructions (Parsons, 2018, p. 75). As they make no a priori assumptions about human behaviour, constructivists, when aiming to solve CAPs, need to identify why collective action is compromised in the first place. This requires an analysis of the intersubjectively shared social constructs agents adhere to (Barkin, 2015, p. 158). If these can be found, constructivist scholars are faced with two options. Either, they can cater to the existing constructs. For instance, if scholars find that certain people under certain circumstances display rational behaviour, they may employ solutions to CAPs that are offered by RCT. However, behaviour change might also be effected by changing the constructs agents adhere to altogether. This process is called persuasion (Payne, 2001, p. 38). Much like the British elites in the 19th century, agents who engage in persuasion are entrepreneurial people who invent and pass on ideas and norms to others (Parsons, 2018, p. 86). The German government, for example, ran the 790 million Euro National Climate Protection Initiative from 2008 to 2017 (BMU, 2018). The Together it’s Climate Protection campaign whose aim was to promote climate-friendly behaviour was part of it. As can be seen on the website, the campaign acknowledges that it appears impossible for individuals to contribute to climate protection and that one’s involvement does not matter. This is a classic statement formulated from the perspective of RCT. However, by subsequently stating that “many individuals” can make a difference as a group, the campaign tries to persuade people that they are, in fact, not individuals but rather part of a collective (BMU, 2015). The view that humans act collectively rather than individually is clearly not formulated for the perspective of RCT. Instead, it can be seen as an appeal to a shared culture or belief system that induces people to provide the common good of stable temperatures.
6 Conclusion
In this essay, I have described the solutions RCT offers in regard to resolving CAPs. For this matter, I have made a distinction between thick and thin rationality. I have argued that the former, stipulating that humans are exclusively propelled by money and power, predicts that humans will react to selective incentives. To resolve CAPs, governing structures can provide incentives trough intervention, regulation, coercion and privatisation that alter the costs and benefits of the agents’ preferred options and diminish group size which facilitates collective action. This way, incentives effect the provision of common goods. The latter, being a ‘theory of everything’ that assumes that human goals are not restricted to money and power, has no power to predict which incentives agents will react to due to its extreme ambiguity. Therefore, no predictively working solutions to CAPs can be derived from thin rationality. Whether we can derive solutions to CAPs from RCT at all, therefore, depends on whether we employ thin or thick rationality.
There are instances, especially in the non-market realm, in which RCT falsely predicts failed collective action. In order to explain these phenomena and potentially devise new solutions to CAPs RC does not provide, it is necessary to question the assumptions of RCT. This is what both BR and constructivism do. BR is characterised by the idea that RCT’s implications, namely that humans posses unlimited computational powers and complete information, are unrealistic. Two prominent streams have come out of BR. First, behavioural economics. To resolve CAPs and effect behaviour change behavioural economists exploit humans’ bounded rationality by employing, for instance, default nudges. Because humans may miscalculate the costs of opting for alternatives or face costs of gathering information, they often stick with default options even if they are costlier than alternatives. If policymakers set the provision of common goods as the default, more people, at least in theory, contribute. A second stream of BR emphasises norms as ways to overcome CAPs. Ostrom, for instance, argues that under very specific circumstances communities have developed norms to better cooperate and provide common pools even in the presence of incentives to defect. Constructivists go further than BR theorists who assume that agents at least strive for what is ‘good enough’. Constructivists make no a priori assumptions about human behaviour. When devising solutions to CAPs, constructivists will, therefore, examine why collective action is compromised in the first place. To resolve CAPs constructivists employ, for instance, the technique of persuasion which is spreading ideas that agents adopt and subsequently live by.
All three approaches to human behaviour show strengths and weaknesses. Thick RC is concerned with explanation, which means that it has high predictive power due to the fact that it assumes a fixed set of rational goals all humans will always strive to achieve. These fixed assumptions make RCT, however, vulnerable in non-market situations were humans do not necessarily seem to follow these goals. To solve CAPs in such instances, scholars must understand the true drivers of human behaviour. This is what BR and constructivism aim to do but since they can no longer make universal claims about human behaviour their efforts denote a loss in predictive power. If anything, scholars of BR and constructivists can only make predictive claims about small numbers of people under certain circumstances in certain situations. Therefore, they, especially constructivists who make no a priori assumptions about human behaviour at all, can only devise solutions to CAPs for smaller groups. In order to effectively fight CAPs, especially ones that are extremely hard to tackle such as climate change ‘prophets’, the people that believe that climate change can be solved through behaviour change, must attack the problem from all possible angles to exploit all approaches’ strengths and mitigate the other approaches’ weaknesses. The wide variety of tools the German government has adopted to fight climate change, namely positive and negative incentives, nudges and persuasive campaigns, demonstrates that this strategy is being adopted by real world policy makers.