Demmers-Rational choice theory PDF

Preview

Summary

The professor is Jean Francois Rattelle....

Description

5

Rational choice theory The costs and benefits of war

As students of violence, we prioritize and connect certain aspects of violent conflict and disregard others. Undeniably, our analysis is selective and influenced by ideological and scientific paradigms and interpretative frames. No matter how systematic, our analysis is always shaped by our ‘knowledge environment’ as well as our personal and political positions and preferences. Academic thinking is about the ability to articulate and situate one’s own theoretical position and to critically review the dialogue between ideas and evidence. It is also about the responsibility to critically think through the political consequences of the knowledge one produces. Evidently, the way in which we frame violent conflict has implications outside academia. Framing also always involves claiming. By framing one not only ‘simplifies and condenses the “world out there” by selectively encoding objects, situations, events, experiences, and sequences of actions within one’s present or past environment’ as Snow and Benford (1992: 137) argue, but one also puts moral claims on, for instance, the (il)legitimacy of violent conflict. One of the core arguments of this book is that the selection of an interpretative frame is not merely an isolated academic act, it is also political. The selection of a form and level of explanation for contemporary violent conflict is a serious political act in the sense that representations have political implications. The way in which violent incidents and conflicts are coded and categorized will play – intentionally, or not – a role in casting blame and responsibility. (Brass 1996: 4) Taking this idea of the politics of portrayal as a point of departure, I aim to do two things in this chapter. First and foremost, the chapter reviews the rational choice approach to violent conflict. It will do so by highlighting the work of Paul Collier and his ‘greed theory of conflict’. This theory starts from the proposition that people will conduct civil war if the perceived benefits outweigh the costs of rebellion. We will look into critiques of the greed model and discuss approaches which aim to ‘rescue’ the study of economic interests in war from Collier’s narrow model. Wars are very often not about winning but serve complex economic and political functions. Consequently, we will discuss the idea of rationality and look into ‘thin’ and ‘thick’ theories of rationality. The second aim of this chapter, or rather its underlying agenda, is to contextualize Collier’s greed theory politically. This is where the discussion on the politics of portrayal comes back in. Why is it that despite the fact that in academia ‘greed theory’ is now rather passé and largely dismissed as flawed (even to a certain extent by Collier (2007) himself ), donor policy makers continue to hold the approach in such high regard?1 This chapter tentatively concludes that, together with ‘good governance’, the ‘greed theory of conflict’ offered policy makers in the aid and development industry a way out

108 Rational choice theory of the ‘crisis of neoliberalism’. By the early 2000s, institutions such as the World Bank were struggling with the disappointing results of two decades of neoliberal restructuring programmes in the developing world. The greed theory offered a way to counter mounting criticism by casting blame and responsibility on local (greedy) actors and holding at bay more complicated discussions on the connections between free market reforms and local violent conflicts. The idea of the ‘evil local’ as the source of war fits the paradigm shift of post-9/11, the War on Terror and the securitization of development. In sum, this chapter is about rational choice theory and the impact of the greed–grievance controversy on donor policy-based thinking on conflict. It uses the rise of the greed theory to illustrate the importance of studying ‘conflict theory in context’.

Rational choice Although the academic field of Conflict Studies includes many different schools of thought, which have operated in isolation, competition and cooperation, alternatively the one or the other gained more – or less – recognition among policy makers. In the 1970s, structuralism and human needs theory were en vogue. In the 1990s, theories of identity and conflict gained prominence, in particular those related to ‘ethnic conflict’. And since the turn of the century, the trend is to explain violent conflict in terms of greed and calculation. This shift to individual calculation and opportunity, with a strong emphasis on material profitability (loot), is part of an influential paradigm in Conflict Studies stemming from neo-classical economic theories. We have made repeated references to rationality and rational choice throughout the book; it is about time we outline the theory. Basically, rational choice theories of conflict start from the proposition that individuals will conduct civil war if the perceived benefits outweigh the costs of rebellion. Subsequently, war will be sustained as long as it is profitable. The idea is that people are rational agents, and have choice, also in settings of war. Violence is thus seen as a product of individual rational action, not collective resentment. Underlying this perspective are a number of axioms and assumptions. First of all, the individual decision-maker is regarded as the primary unit of analysis. This is called methodological individualism: the individual is the starting point for all social action and, therefore, explanation. Second, in formal (or thin) rational choice theory, the individual is seen as an objective utility maximizer. We continuously set goals and preferences, and calculate the probability of achieving each of these goals, by weighing costs and benefits and the utility of each outcome. In the end, we choose that which has the highest expected utility, hence the term rational choice. In some cases, the choice is to go to war, that is, to take the risk of dying, or having to kill in combat. Underlying the idea of utility maximizing is neo-classical economic theory and its view of the human being as homo economicus. ‘Economic man’ is the rational human being, who acts out of self-interest and the desire for wealth that economists use when deriving, explaining and verifying their theories and models. The idea is that all social action can in the end be explained as responses to signals from often imperfect markets. As Wallace and Wolf put it: ‘there is a prize for everything and everything has its prize’ (1999: 294). Rational action approaches underlie the realist school in the field of international relations, referring to the grand design of states engaged in power politics, but also guide game theoretic approaches to ethnic violence such as Fearon’s (1994) model of credible commitments and the work of Lake and Rothchild (1997) on inter-group and intra-group strategic dilemmas that produce ethnic violence. Deep down, rational action approaches find their roots in the work of Machiavelli and Hobbes and the underlying idea of the human being as driven by an

Rational choice theory

109

instinct for self-preservation and a will to dominate. They also draw on eighteenth-century economists such as Adam Smith and his work on the unintended beneficial effects of the individual’s profit maximizing tendency. It is beyond the scope of this chapter to discuss the theoretical foundations of rational action (for an excellent overview, see Jabri 1996: chapter two). Our focus is on the sub-field of rational choice theories of violent conflict. One of the advocates of the utility maximization theory of violent conflict is Bruce Bueno de Mesquita. In The War Trap (1981), he argues that war is likely to emerge if actors (assumed to be unitary) on both sides expect it to be profitable. In summarizing his theory, he writes: ‘The broadest and seemingly obvious generalization that emerges from the theory is the expectation that wars (or other conflicts) will be initiated only when the initiator believes that war will yield positive expected utility’ (1981: 127). Within this line of thought, the origins of the decision-maker’s ‘beliefs’ are not subject to inquiry; they are accepted as given. We will return to this later. Stripped of nuance, Bueno de Mesquita’s theory holds that actors (and these can be individuals, groups or states) are unitary objective utility maximizers whose actions are consistent and goal-directed. More than two decades after The War Trap, Paul Collier and colleagues produced an influential World Bank report named Breaking the Conflict Trap (World Bank 2003). The report contains very much the same message as the book it refers to: it is utility maximization that drives wars, in this case, civil wars and rebellions. Much of the report’s policy recommendations are based on Collier’s earlier work with Hoeffler (Collier and Hoeffler 2001) on what has become known as the ‘greed theory of war’, and which marked the beginning of the proliferation of neo-classical economic theories of violent conflict. Paul Collier’s theory of greed In his quest to find the ‘true motivation for rebellion’, Collier takes an explanatory epistemological stance. That is, he studies social action from without and infers motivation from patterns of observed behaviour. In this view, interpretation, meaning or sense-making are merely seen as distortions. Rebel movements, for instance, have a tendency to embed their actions in ‘narratives of grievance’, for this is simply more acceptable, and hence profitable. To discover the truth, in the words of Collier, is not to study what people say but what they do. The argumentation that leads him to conclude that ‘the true cause of much civil war is not the loud discourse of grievance but the silent force of greed’ is based first and foremost on the construction of a dichotomy. Collier breaks down the motivation for civil war into two distinct extremes. On the one hand, ‘rebellions may arise because rebels aspire to wealth by capturing resources extra-legally’: this type of motivation is labelled greed. On the other, ‘rebels aspire to rid the nation, or the group of people with which they identify, of an unjust regime’: in this case the rebellion is grievance-based (2000: 91). Now, how do we know whether a violent conflict is driven by greed or grievance? For a good explanation of Collier’s work, I need to spend a few words on proxies. Building an empirical model of civil war with the tools of neo-classical economic analysis and based on rational choice assumptions implies – as is the case in all quantitative statistical large-N analyses – having to quantify possible causal factors, such as greed or grievance. Since it is not possible to quantify greed or grievance directly, scholars need to define the kind of evidence that would indicate whether the one or the other is at play. In more technical terms, scholars have to define quantifiable proxies, that is, ‘indirect measures to stand in for the actual theoretical postulate’. Subsequently, data can be gathered on these quantifiable variables across the sample and findings can be produced. We will return to the use of proxy variables later. Collier builds his

110

Rational choice theory

model of greed versus grievance with three proxies for greed, and four for grievance, tested against available data for a large sample of countries. Greed is reflected in the following variables: the share of primary commodity exports in gross domestic product (GDP); the proportion of young males in the total population; and average years of schooling. Simply put, where there is a large proportion of young men in a society, where these men have few income-earning opportunities (expressed in years of education) and where there is plenty of opportunity for economic gain by looting primary commodities, a country is much more at risk of conflict than one with opposite characteristics. Consequently, Collier contrasts his greed factors with those that proxy grievance. First, Collier measures the tendency to ethnic grievances or religious hatreds by the extent to which a society is factionalized by ethnicity and by religion. Second, grievances focused on economic inequality are measured by land ownership. Third, grievances over a lack of political rights are measured by levels of repression and autocracy. Finally, grievances on government economic incompetence are expressed in the proxy variable of the growth rate in the five years prior to the outbreak of civil war. Collier puts the above hypotheses to the test and concludes that the results overwhelmingly point to the importance of greed agendas as opposed to grievance. Indeed, he concludes that the grievance factors are so unimportant or perverse that they cannot be anything but ‘seriously wrong’ (2000: 96). This is, in a nutshell, the empirical part of the model. Collier’s choice of proxies has been heavily criticized, and there is a long and ongoing debate about this in the subfield of quantitative conflict analysis (Ron 2005; Fearon 2005; Humphreys 2005). Fundamental criticism is expressed by Cramer (2002, 2006), who states that: ‘The use of empirical surrogates for theoretical hunches is patchwork through and through. Thus, it is not clear that the proxy is capturing the thing it is meant to stand for in the abstract model’ (2006: 131). He argues, as have many others, that Collier’s proxies to measure greed might just as well be used to measure grievance. For example, a lack of education might block the way to gainful employment and, therefore, reveal a low opportunity cost for conflict but it might just as well be a source of social anger and grievance. The same could be said for the share of primary commodities in GDP, as a high dependence on such exports often characterizes a country’s general economic underdevelopment. Many others have critiqued Collier’s representation of greed versus grievance as a misleading opposition and stressed how the two motivations for civil war are in fact highly interrelated (Ballentine and Sherman 2003; Murshed and Tadjoeddin 2007). Empirically, critiques pertain to a general lack of reliable data, as well as to how the focus on aggregate countrylevels hides the variation in violence (in line with the micro-political turn in chapter four). It is, however, not the intention of this chapter to extensively review this discussion; others have done this very well before us (see Cramer 2002, 2006; Goodhand 2003; Fearon and Laitin 2003; Humphreys 2005). We focus here not so much on the methodology of Collier’s work but on the theory that is used to support his findings. Violent conflict as market Rational choice theories of violent conflict revolve around the question of ‘why do people follow’ – one that continues to pop up in this book. Why would people want to join an insurgency or a rebel movement? Or, in rational choice language: ‘Why would they want to risk joining?’ As Kalyvas and Kocher (2007b) point out, it is indeed the joining (or ‘following’) that needs to be explained and not so much the ‘first movers’, that is, the political entrepreneurs with strong motivations who are willing to assume high risks:

Rational choice theory

111

The world does not lack Che Guevaras ready to launch insurgencies. What it does lack, however, is a mass of followers willing to take the necessary risks: it is the success of entrepreneurs in recruiting followers that results in insurgencies and that calls for investigation. (2007b: 182) Collier’s explanation of why it is greed and not grievance that drives civil war is derived from the idea of the individual as utility maximizer. For the rational, utility maximizing actor the costs of taking part in a grievance-based rebellion simply outweigh the benefits. This is explained by means of Mancur Olson’s ‘collective action problem’ (1965). The collective action problem is still canonically applied to the analysis of political groups and group action, including those engaged in civil war. Simply put, groups struggling for certain public goods (e.g. the overthrow of a dictator, freedom of speech) need to overcome three collective action problems: free riding, coordination and time-consistency. Collective action problems are seen as particularly large for rebel movements. In Gould’s words (1995: 204) ‘While activists, might have little trouble persuading a casual acquaintance to sign a petition, they would have great difficulty convincing such a person to risk injury, death, or imprisonment’. Rational, well-informed individuals may well feel very strongly about an injustice, but they will be unwilling to put their lives on the line for a cause unless it is likely to benefit them. No matter how badly oppressed a group is, or how sharp their hatred for a rival group, the three-fold problem of free riding, coordination and time-consistency will foil any rebellion attempted on the basis of grievance alone. Only by appealing to individual greed can these collective action problems be overcome. Collier explains the three collective action problems as follows. Enjoyment of a public good (such as equality, stability, cleaner air, defence) cannot be exclusive to an individual, it is always collective. So, if I am suffering under a repressive government’s rule, I may very well want to join a movement that aims to overthrow this government. However, it is most likely that the fall of the regime does not depend on whether I personally join the rebellion. Hence, it is more profitable for me to have others fight, while I benefit from the justice that their rebellion achieves. This standard free-rider problem, according to Collier, will often be enough to prevent the possibility of grievancemotivated rebellions. The second problem that plagues grievance-based rebellions is that of coordination. Rebellions have to start somewhere. The problem is that individuals would rather join large rebellions that stand a chance of military victory than small ones. Again, the costs and risks of punishment faced by individuals in a small rebel group are calculated to be high, without much prospect of achieving justice any time soon. And then there is the third, time-consistency problem: rebels have to bear the cost of fighting before they can enjoy the benefit of achieving justice. For the utility maximizer, this is calculated as too uncertain an investment. In stark contrast, Collier argues, the above three collective action problems can be overcome by ‘the logic of greed’. By engaging in violence and looting, greed-motivated rebels can benefit instantly from war. There is no free-rider problem because the benefits of the rebellion can be confined to those who take part in it. If one can benefit from looting, it can also pay to be part of a small rebellion group. The time-inconsistency problem is also overcome: for rebels do not have to patiently await the rewards of justice, or equality, or whatever public good they were fighting for, but can instantly gain individually. The underlying view here is that of violent conflicts as markets, where individuals make investments in violence and reap private returns. The market of grievance-based conflict shows important ‘imperfections’: investments but no returns, costs without benefits, risks

112

Rational choice theory

without compensations. Hence, although real enough, grievances cannot ‘produce sufficient violence to clear the market for conflict’. Greed motivations very easily can. According to Cramer, Collier’s way of thinking is representative of a tremendously influential trend in the social sciences – what some have called ‘economic imperialism’, in which a certain brand of economic theory colonizes other social sciences, including the study of civil war. The idea is that it might be possible to explain ‘virtually everything that otherwise is inarticulately garbled in the efforts of historians, sociologists, political theorists’ as – not partial, but complete – responses to signals from often imperfect markets (Cramer 2006: 125). Critiques At a fundamental level, scholars simply reject the assumptions underlying rational choice theory. Tarrow, for instance –...