Moving beyond the rational choice debate via social capital: The study of illegal private protection

Abstract

In the context of the epistemological and methodological debates about rational choice (RC) sparked by Green and Shapiro and the subsequent rejoinders, Hindmoor made the theoretical claim that there is now an opportunity for RC to be enhanced and for past debates to be set aside by incorporating different theoretical perspectives into RC. This article presents an example of one context in which this can be done: the study of illegal private protection (IPP) produced by organised criminals and corrupt officials. It extends a rational theory of criminal protection to enable the simultaneous analysis of corrupt protection and augments this approach with theories of social capital (SC). Both positive and negative forms and effects of SC are considered. This approach highlights a middle ground between advocates and critics of RC, demonstrating a capacity to explain patterns in the production of IPP between organisations and across countries.

针对格林、沙皮罗与弗里德曼的否定者所发起的有关理性选择的认识论及方法论论战，辛德摩尔提出一个理论主张：现在正可以通过吸纳不同的视角而弘扬理性选择，以往的辩论可置之一旁。本文讨论了一个可行的案例：有组织犯罪分子造成的非法私人保护。作者应用理性理论于犯罪保护，对腐败保护做了分析，并用社会资本理论扩展了这一方法。作者对社会资本理论的正负短长做了分析。这样的研究强调了介于理性选择理论赞同方与反对方的中间地带，展示对于组织间及跨国非法私人保护的产生的解释能力。

Keywords:

• rational choice theory
• social capital theory
• organised crime
• corruption

Notes

¹The application of RC is segmented-universalist if it applies RC in contexts where behaviour can be observed as rational in a sense that fits the theory. Such contexts are those with competitive environments and institutional structures that eliminate agents who do not act rationally, such as firms in competitive markets or political parties in competitive political systems (Ostrom 1998; see also Ferejohn 2002). Clark (1996) explored the way such structures constrain choice. RC does less well at explaining, say, the behaviour of voters in political systems (Satz and Ferejohn 1994: 72). Applying RC to explain voter behaviour would not, therefore, adhere to segmented-universalism.

²This article's approach is also instrumentalist, in that IPPMs emerge in conditions that constrain individuals such that they behave as if they were rational. MacDonald (2003) provides an introduction to instrumentalist empiricism in contrast with scientific realism.

³For more on the distinction between thick and thin rationality, see Ferejohn (1991, 2002) and Lichbach (2003).

⁴Hobbes acknowledged cooperation as a possible (though unlikely) solution to this dilemma (Gallarotti 2010: 78–93). That RC theory itself allows for agents to overcome problems of collective action is often not well understood (Dowding 2001).

⁵There are other possible solutions. Good examples of the independent formation of such institutions among merchants (rather than the exogenous imposition of institutional rules) are provided by Milgrom et al. (1990) and Greif et al. (1994).

⁶There is a subtle but important difference between running a protection racket, in which individuals are extorted to pay protection money or suffer the consequences, and between the production of genuine protection as a service through illegal means. The line between genuine protection and extortion can be blurry. The tendency for one to mutate into the other is acknowledged but not explored here. This article considers IPP to be solely genuine protection, as the factors affecting extortion differ from those affecting genuine protection.

⁷Gambetta (1988) attributed this to historical geopolitical factors such as the divisive actions of Spanish Hapsburgs in southern Italy and the coercive pressure applied by landed lords upon peasants in the 19th century. For other accounts of the origins of the Sicilian Mafia, see also Hess (1973), Servadio (1976), Arlacchi (1986), Nicaso and Lamothe (1995) and Jamieson and Violante (2000).

⁸State protection may even come to be seen cynically as part of a broader predatory regime. The idea of predatory state protection was outlined by Tilly (1985), who compared early European states to criminal organisations, arguing that their emergence had more to do with running profitable protection rackets and controlling power than with providing genuine protection. For the reverse comparison (of gangs with states) see Grossman (1995) and Skaperdas and Syropoulos (1995). Problems trusting the state are related to what Ostrom (1998: 6) termed second-order dilemmas and to Leeson's (2009: 27) definition of the paradox of power as ‘the need for an authority and the fact that the very introduction of such an authority generates strong incentives for him to abuse his power’. Olson's (1993) comparison of autocratic and democratic states and his subsequent theory of dictatorships (Olson 2000) identify two key factors that are useful in determining the presence (or degree) of the paradox of power: the encompassing interest and time horizons of those in power.

⁹On the problems business faced and the range of services provided by gangs, see Varese (2001: 102–20) and Volkov (2002: 17, 53). For more on the sorry state of Russian security services in this period, see Galeotti (1998), Frye and Zhuravskaya (2000) and Frye (2002).

¹⁰For an example of the importance of self-protection and its role in the initial formation of a group that later specialises in producing protection for others, see Zhang and Chin (2008: 186–87) on the United Bamboo gang in Taiwan.

¹¹Gambetta maintained this was distinct from being a violent entrepreneur, who pursues other business interests through the use of violence. He also rejected the argument that organised crime was concerned with managing violence per se (Blok 1974), asserting instead that violence is a factor of production in protection. Chu (2000), Varese (2001), Volkov (2002) and Hill (2003) have all found supporting evidence for this distinction in the contexts of China, Russia and Japan (even though the title of Volkov's Violent entrepreneurs suggests the contrary).

¹²Note also that officials may possess the private capacity for violence, but if they use it to provide protection then they do so as just another private individual entrepreneur of violence, distinct from their public office.

¹³Gong (1997: 282) gives an example of the former, in which a public official offers protection in the form of facilitating tax evasion. Stewart (2007: 156–7) provides an example of the latter, detailing circumstances in which officials in the Australian passport office colluded to provide a major illegal narcotics importer with the passports that facilitated their criminal enterprise.

¹⁴This is what Morselli (2009: 63) refers to as the efficiency–security trade-off.

¹⁵The crudest form of trust created in this way is the guarantee of compliance through fear, but loyalty to a group may be (or become) internalised as a social norm (as discussed below) rather than due to a virtual form of mutual blackmail or hostage-taking.

¹⁶It is worth noting, however, that the structure of Italian organised crime in Sicily and elsewhere has moved beyond older, more traditional family-based networks (Paoli 2003).

¹⁷On the vory v zakone, see Rawlinson (1997). For an analysis about how and why professional sport in Russia helped forge relations and skills which lent themselves naturally to IPP, see Varese (2001: 57–59) and Volkov (2002: 8–10). The role played by wrestlers in providing IPP in Bulgaria (Petrunov 2006; Tzetkova 2008) is a good example of the similar trend in other post-communist states.

¹⁸For a discussion of different kinds of networks and their relations to theories of social capital and trust, see Schuller et al. (2000).

¹⁹There is a reasonably well-formed literature on the role of trust and social capital in crime, but studies of illegal private protection have been essentially (if not wholly) based on an RC approach.

²⁰In developing these concepts, Putnam built on Granovetter's (1973) work on weak and strong ties.

²¹Note the distinction here between the IPP the organisation produces for itself and the IPP it produces for third parties. To the extent that rival organisations pose a threat, high levels of bridging capital reduce the need for self-protection, while simultaneously allowing an organisation to better cooperate and collaborate with other groups to produce protection on a larger scale for third parties.

²²Coleman's (1988) pioneering work on SC was rooted in RC, and his later work (Coleman and Fararo 1992) considered the role of RC in substantial detail, though it has also been claimed that SC lies outside the explanatory range of RC (Goldthorpe 1998: 177).

²³Such distinctions are useful as they allow insights from game theory and experimental economics to be used to better understand the nature of such trust. Ostrom's (1998) work on rationality is one example of an overview of such studies of relations and trust between individuals in strategic interaction.

²⁴This idea is present in Coleman's (1988: 98–100) views on social capital, where he shows that networks tended to form whenever two or more people stood to gain from cooperating.

²⁵A study of Chicago drug gangs (Levitt and Venkatesh 2000: 769), for example, showed that gang members were acutely aware of the social incentives to look after each other. The ‘us versus them mentality’ among Australian police officers is another example (Stewart 2007: 42).

²⁶That norms can be reinforced within a group and strengthened by the joining of new like-minded members helps explain the prevalence of group-specific norms regarding reputation or saving face, or the effectiveness of methods of group censure such as sanctions. These factors play a vital role in resolving problems of collective action in general (Ostrom 1998). Weisfeld and Feldman (1982: 570) note the role of identity in gang membership.

²⁷Nor is it necessary for all individuals to share the same motivation. Some may use SC as an instrument, others may develop it to form an identity – in reality people quite often do both. By being robust to variation in motives, the approach here can accommodate the full spectrum of these possibilities.