Whom Can We Trust?: How Groups, Networks, and Institutions Make Trust Possible

Chapter 3. Social Networks and Trust in Cross-Cultural Economic Experiments

Chapter 3 Social Networks and Trust in Cross-Cultural Economic Experiments ABIGAIL BARR, JEAN ENSMINGER, AND JEFFREY C. JOHNSON I N THIS CHAPTER, we present two datasets from Africa, one rural and one urban, in which we examine the correlates of individual-level demographics and trusting and trustworthy behavior in economic experiments . We use a slightly modified version of the Joyce Berg, John Dickhaut, and Kevin McCabe investment game (1995). Our primary original contribution is to include in these demographics data on each individual’s standing in their social network (compare Alesina and La Ferrara 2002; Anderson, Mellor, and Milyo 2005a, 2005b; Bouckaert and Dhaene 2003; Burns 2004; Chaudhury and Gangadharan 2003; Croson and Buchan 1999; DeBruine 2002; Eckel and Wilson 2003, 2004). We hypothesize that those who are pivotally and centrally located in social networks hold such positions because they have established and maintained reputations as successful social and political entrepreneurs, and that such positions are achieved in part by demonstrating trustworthiness. We define political entrepreneurs as those who strategically cultivate, create, and invest in social relationships to enhance their bargaining power and political brokerage abilities in areas such as conflict management and institutional change (compare Schneider and Teske 1992). This is a quality we believe is well identified by conventional measures of social network centrality (compare Christopoulos 2006). Further, we hypothesize that such entrepreneurs are risk takers by nature, and that this propels them to risk trusting to reap the rewards of cooperation that stem from such behavior. Inclinations to trust are further reinforced by the greater access to information, including 65 information concerning who is worthy of trust, that such pivotal positions in the social network afford. Social networks are central to the concept of social capital as most people use the term, but studies of social capital have suffered from a lack of conceptual clarity. The metaphorical use of the concept and the looseness with which the term social capital has been operationalized, together with the power with which some have endowed an ill-defined version of the concept, has lead to some discounting the entire concept. In this chapter, we seek to differentiate individual from group-level concepts of social capital and attempt to restore some clarity to the concept. We then take the individual component of social capital that exists as social or political entrepreneurship , and use social network analysis to provide precise measures of this individual-level trait. Finally, we derive specific predictions about the relationship between network position and the relative level of trust and trustworthiness exhibited by individuals within a given society. Concepts of Social Capital We are certainly not the first to attempt to make a link between social capital and trust and trustworthiness, nor are we the first to recognize the lack of consistency in providing a clear operational definition of social capital. Over the next few sections we take a brief look at the concept of social capital particularly as it relates to social networks and economic experimental games. There has been considerable discussion in the trust literature concerning the vagueness surrounding the various definitions, measures and applications of the concept of social capital (Durlauf 2002; Carpenter, Daniere, and Takahashi 2004). Further, and more important, is the general recognition of problems stemming from confusion surrounding the analytical levels at which social capital has been theoretically conceptualized. As others have noted, the important distinction among societal, organizational , group- and individual-level notions of social capital has often been either obscured or not well articulated (Cook, Hardin, and Levi 2005; Glaeser, Laibson, and Sacerdote 2002; Carpenter, Daniere, and Takahashi 2004; Guillén et al. 2002; Lin, Cook, and Burt 2001). Yet such a distinction is critical to gaining a more realistic understanding of the role of social capital in accounting for individual-level variation in trust and trustworthiness. Group-Level Social Capital Much of the work examining the relationship between trust and social capital has been at the community or aggregate level (Glaeser, Laibson, and Sacerdote 2002). One general use in the literature has to do with the degree to which an actor is embedded in a dense set of social relations—the denser an actor’s relations, the higher their social capital (Coleman 1990; Portes 66 Whom Can We Trust? and Sensenbrenner 1993). This group-oriented conceptualization stands in stark contrast to the individual-level social capital. Here density provides security for individual group members in that...