Exploring the Effects of Social Capital on Police Performance in U.S. Cities

This study explores the relationship between social capital and police performance. Since substantial academic interest in social capital emerged, many studies demonstrate that social capital has a significant influence on social and political outcomes. Despite the renewed interest in examining the role of social capital in the field of public safety, critics have pointed to several limitations: (1) the ways in which social capital variables are defined and measured; (2) the ways in which police performance variables are defined and measured; (3) the lack of specific theoretical propositions linking these two constructs; and (4) the lack of empirical work at the city level. This dissertation addresses these limitations by modeling causal mechanisms linking eleven dimensions of social capital and four types of police performance using two competing explanations: citizen demand and citizen supply. According to citizen demand explanations, communities with high social capital are more likely to demand that their police departments do the job more efficiently and more effectively. Citizen supply explanations, however, suggest that social capital may facilitate informal social control mechanisms to reduce crime and delinquency. The divergence lies in the way the two explanations see citizens' roles in police duties, as demanders or co-producers of public goods and services. Based on these two streams of explanations, this study proposes eight testable hypotheses regarding the relationships between social capital and police performance and employs ordinary regression analysis (OLS). In addition, in order to take into account potential endogeneity problems between social capital and outcome-related police performance, this study also proposes four non-recursive models and employs a two-stage least squares (2SLS) regression analysis. To measure each of the constructs of social capital and police performance, this study uses the data of the Social Capital Community Benchmark Survey by Harvard University, the Uniform Crime Reports of the FBI, and the Law Enforcement Management and Administrative Statistics of the Bureau of Justice Statistics, yielding a cross-sectional data set with 44 U.S. cities in the year 2000. This study presents theoretically and practically significant implications regarding the causal relationships between social capital and police performance. First, the findings support the multi-dimensionality of social capital and show that the results of the analyses are mixed with regard to each of the four types of police performance. These findings are contrary to the traditional wisdom that social capital is equally positively related to any type of government performance in general and police performance in specific. Second, the results of the analyses with the disaggregated four types of police performance variables show that social capital is a useful concept for explaining the different levels of outcome for police performance across local police departments, yet less useful a concept for explaining the efficiency within government bureaucracy. Third, with regard to causal mechanisms, there are no overarching theories which account for every performance practice in police administration. For outcome-related police performance, social capital may work; for output-related police performance, it may not. Lastly, the non-recursive analyses further underscore the complexities of the interrelationships between the dimensions of social capital and police performance. The results of the study also have some important implications. First, the findings, in general, suggest that police administrators should think more about their relationship with citizens and consider citizens as consumers as well as co-producers of public safety. Second, this study also reinforces the basic tenet of community policing strategies, suggesting collaborative partnership between police and citizens. Third, this study provides an opportunity to rethink the criteria for the success of police service effectiveness or performance, and suggests that there is no one explanation for the effect of social capital on police performance.