Identifying Legitimacy: Experimental Evidence on Compliance with Authority

We consider the extent to which individuals’ perceptions of an authority’s legitimacy affect their intrinsic compliance motivations. Identifying this relationship empirically is challenging because actions and institutions that might affect legitimacy generally also affect citizens’ material incentives. We formalize this identification problem and describe sufficient conditions under which a “personal” legitimacy effect can be separately identified. We then describe results of two public goods experiments that circumvent barriers to inference highlighted by our model. The first reveals a large, highly significant personal legitimacy effect. The second demonstrates that this effect is not reducible to subjects’ motivation to “pay back” authorities. Because unlimited coercive capacity is prohibitively costly or otherwise infeasible, governing authorities who wishes to translate laws and behavioral norms into social order must rely at least in part on the intrinsic motivations of individuals to comply with their edicts. This paper considers the extent to which those motivations are affected by perceptions of the legitimacy of that authority; and the extent to which the choices of that authority, and the institutional context he or she occupies, enhance or diminish those perceptions. The empirical antecedents of legitimate authority in hierarchical settings are a concern across the social sciences, particularly in sociology (Weber, 1947, 1978) and social psychology (Leventhal, 1980; Tyler, 1990), but also political science (Lipset, 1960; Easton, 1965; Dickson, Gordon, and Huber, 2015) and criminal justice (Bottoms and Tankebe, 2012). Applications to law enforcement are especially salient in the contemporary United States, where the issue of police legitimacy, particularly in minority communities, has received prominence following a sequence of highly publicized police shootings. Economists (e.g., Greif, 2006; Dixit, 2009) have also highlighted a government’s legitimacy as an essential component of its success in enforcing the law (see also North, 1981; Levi, 1988). Greif (2006, 148), for example, points to the legitimacy of those who issue rules as central to institutionalization, because the success of any institution established by decree requires that a critical mass of followers believe others will abide by it. But the individuals and organizations that possess said legitimacy vary markedly across social and historical contexts, and numerous equilibria are possible in principle (e.g., Basu, 2015). These issues arise not only when considering the relationship between government and citizens, but also between managers and workers (Kreps, 1990). The notion that legitimacy of authorities is a constituent feature of equilibrium in a class of coordination games necessarily abstracts from the antecedents that may give rise to the beliefs supporting it. To construct a micro-level account of these antecedents, we begin with a definition from social psychology: Tyler (1997) defines legitimacy as a “judgment by group members that they ought to voluntarily obey social rules and authorities irrespective of the likelihood of reward or punishment” (emphasis ours; see also French and Raven 1959). In other words, legitimacy obtains when the features of authoritative institutions, or the choices of individuals in positions of authority, enhance the intrinsic motivations of citizens to carry out certain social duties. This account of the cognitive basis of legitimacy is closely related to Hermalin (1998), who explores how costly actions by leaders (self-sacrifice and leadership by example) can reveal information to subordinates about the marginal return to their efforts; and to Bénabou and Tirole (2003), who explore how performance incentives can de-motivate an agent when they reveal a principal’s private information about the agent’s motivation or ability.1 While these papers offer micro-founded accounts of intrinsic motivations that incorporate the beliefs and actions of both authorities and subordinates, they touch on the notion of legitimacy only indirectly. Empirically identifying legitimacy and its behavioral and procedural antecedents is challenging because of a fundamental problem of causal attribution. The problem arises because procedures or choices that enhance an authority’s legitimacy may simultaneously alter citizens’ beliefs about the authority’s capacity to bestow rewards and punishments. These material factors may, in turn, affect citizen behavior, quite apart from considerations of the authority’s legitimacy.2 To develop a research design that can circumvent this problem, we describe a model in which agents are motivated to comply with an authority not only extrinsically by material rewards, but also intrinsically by their positive perceptions of the authority herself (“personal” legitimacy) and the underlying governance structure (“institutional” legitimacy). The authority can take a costly action that influences those perceptions, but may also affect the agents’ extrinsic reward structure. The model suggests that institutional legitimacy effects are generally not separately identifiable from the effects of extrinsic rewards and punishments, but also reveals sufficient conditions under which identification of a personal legitimacy effect is feasible. These conditions correspond to situations in which costlier actions on the part of the authority correspond to “better” probability distributions over governance structures. Under such conditions, an informative equilibrium exists in which authorities who care more about, e.g., the fairness of governance undertake 1Also related is Akerlof (2017), who considers a moral hazard model in which principals with high legitimacy use orders in lieu of monetary incentives to motivate agents. 2For a related critique of the analytical utility of the concept of legitimacy in the sociology of law, see Hyde (1983).