Do Fringe Banks Create Fringe Neighborhoods? Examining the Spatial Relationship between Fringe Banking and Neighborhood Crime Rates

Abstract

In the aftermath of one of the worst recessions in US history, high unemployment has placed millions of Americans in precarious financial positions. More than ever, Americans are opting out of traditional financial services, relying instead on “fringe lenders” such as check cashers, payday lenders, and pawnshops to manage their finances. Given their tremendous growth and the concern that consumers who are least able to pay for high-cost, high-risk financial products are most likely to use them, fringe lenders have been the subject of controversy and the focus of much research. Largely unknown, however, are the effects of fringe lenders on the communities where they are located. Given their spatial concentration in low-income neighborhoods with greater concentrations of racial and ethnic minorities—areas with typically more crime—of concern is whether fringe lenders themselves are criminogenic. We consider this by examining the impact of several types of fringe lenders on neighborhood crime rates in Los Angeles. Our findings reveal that the presence of fringe banks on a block is related to higher crime levels, even after controlling for a range of factors known to be associated with crime rates. The presence of a fringe bank also impacts crime, particularly robbery, on adjacent blocks. Whereas we find that pawnshops have little impact on crime levels, payday lenders and check cashers have a much stronger impact. Finally, we discover there are moderating effects, as the fringe lender–crime relationship is considerably reduced if the lender is located in a higher population density area.

Keywords:

• fringe lenders
• community crime rates
• land use
• social disorganization
• spatial analysis

Acknowledgments

We thank Tim Wadsworth for comments on an earlier draft of this paper and Adam Boessen, Nick Branic, James Bustamante, and James Wo for research assistance.

Notes

¹ Research finds that criminals generally prefer to sell stolen goods locally (Sutton, 2010, p. 7; Wellsmith & Burrell, 2005, p. 743).

² These arguments help explain why the impact of fringe lenders on crime is likely to be qualitatively different compared to, say, conventional banks or ATMS, both of which provide immediate cash to customers, thus creating criminal opportunity. Although like fringe lenders, conventional banks and ATMs perform similar functions, unlike fringe lenders, banks and ATMS in a community are not associated with reduced informal social control, deterioration, disorder, and incivilities.

³ We are careful to avoid claims of causality in the current study, given our data are not longitudinal.

⁴ We don’t analyze rape, given well-known reporting issues with this crime type (e.g. Jensen & Karpos, 1993).

⁵ We also constructed an index of concentrated disadvantage using other variables sometimes employed in such indices (e.g. percent divorced, per capita income, the poverty rate, the unemployment rate). The correlations among the various factor scores ranged from .94 to .99. Thus, the choice of variables is not crucial.

⁶ Given that only the percent single-parent households variable is available for blocks, we used an aerial interpolation technique utilizing ancillary data based on the technique of Flowerdew, Green, and Kehris (1991). The other variables used in the imputation model were percent owners, racial composition, percent divorced households, percent households with children, percent vacant units, population density, and age structure (percent aged: 0–4, 5–14, 20–24, 25–29, 30–44, 45–64, and 65 and up).

⁷ We also constructed measures based on smaller (2.5 mile) buffers, and the results for our fringe bank measures were unchanged. The use of larger buffers makes little difference, given the inverse distance decay function used to weight the data.

⁸ By using these buffers we avoid the boundary problem. These buffers include information on all blocks within five miles, given that we have Census data for all blocks (both inside and outside the city). If we had instead constructed spatial buffers of crime, we would encounter the boundary problem, given that we do not have crime data outside the city boundaries.

⁹ We tested ancillary models in which we also included interactions between fringe banks and block or block group-concentrated disadvantage. These interactions were never statistically significant, suggesting that the economic context does not constitute an important moderating effect of these relationships.

¹⁰ The coefficients for the instrumented block-level fringe banks were: 8.46 (t-value = 6.76) for robbery; 4.44 (t-value = 3.28) for aggravated assault; 6.18 (t-value = 8.23) for burglary; 9.71 (t-value = 14.67) for larceny; 6.89 (t-value = 9.58) for motor vehicle theft; and 6.78 (t-value = 1.37) for homicide.

¹¹ Because of the completeness of our model, any one individual variable’s share of the variance explained will typically not be very large. Furthermore, pseudo R-squares are not an exact analogue to variance explained. Nonetheless, we assessed that, on average, the inclusion of the fringe bank measure increased the pseudo R-square .24%. With the exception of the population measures which had a larger impact on the pseudo R-square, the inclusion of the other variables in the model one at a time only increased the pseudo R-square between .12% and 3.06%.

Additional information

Funding

Funding. This work was supported by the National Institute of Justice [2012-R2-CX-0010].

Notes on contributors

Charis E. Kubrin

Charis E. Kubrin is a professor of Criminology, Law and Society at the University of California, Irvine. Her research examines neighborhood correlates of crime. She is coauthor of Researching Theories of Crime and Deviance (Oxford University Press 2008) and Privileged Places: Race, Residence, and the Structure of Opportunity (Lynne Rienner 2006) and coeditor of Introduction to Criminal Justice: A Sociological Perspective (Stanford University Press 2013), Punishing Immigrants: Policy, Politics and Injustice (New York University Press), and Crime and Society: Crime (3rd ed.; Sage 2007).

John R. Hipp

John R. Hipp is a professor in the departments of Criminology, Law and Society, and Sociology at the University of California—Irvine. His research interests focus on how neighborhoods change over time, how that change both affects and is affected by neighborhood crime, and the role networks and institutions play in that change. He has published substantive work in such journals as American Sociological Review, Criminology, Journal of Quantitative Criminology, Social Problems, Mobilization, City & Community, Urban Studies, and Journal of Urban Affairs. He has published methodological work in such journals as Sociological Methodology, Psychological Methods, and Structural Equation Modeling.