Racial Bias in Motor Vehicle Searches: Theory and Evidence

Journal of Political Economy (Impact Factor: 2.9). 12/1999; 109(1). DOI: 10.1086/318603
Source: RePEc

ABSTRACT African- American motorist in the United States are much more likely than white motorists to have their car searched by police checking for illegal drugs and other contraband. The courts are faced with the task of deciding on the basis of traffic-search data whether police behavior reflects a rackial bias. We discuss why a simple test for racial bias commonly applied by the courts is inadequate and develop a model of law enforcement that suggests an alternative test. The model assumes a population with two racial types who also differ along other dimensions relevant to criminal behavior. Using the model, we construct a test for whether racial disparities in motor vehicle searches reflect racial prejudice, or instead are consistent with the behavior of non-prejudiced police maximizing drug interdiction. The test is valid even when the set of characteristics observed by the police is only partially observable by the econometrician. We apply the test to traffic-search data from Maryland and find the observed black-white disparities in search rates to be consistent with the hypothesis of no racial prejudice. Finally, we present a simple analysis of the tradeoff between efficiency of drug interdiction and racial fairness in policing. We show that in some circumstances there is no trade-off; constraining the police to be color-blind may achieve greater efficiency in drug interdiction.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In the last decade, models of rational choice have chimed into the discussion on racial profiling, the use of race in stop and search decisions of the police. The models describe the behavior of motorists and the police and provide empirical tests to assess the question whether the police exhibit racial animus. However, existing studies have neglected the effect of spatial and temporal aggregation of the data on the application of the tests. Using data from the Florida Highway Patrol, this paper shows that regional subsets disclose policing behavior which deviates substantially from the aggregate. Broad conclusions on the absence or presence of racial prejudice are thus at risk of being unfounded. The results call for a cautious application of the tests and the interpretation of their conclusions, not least because a disaggregated analysis may fail to verify key assumptions of the models.
    Schweizerische Zeitschrift für Volkswirtschaft und Statistik / herausgegeben von der Schweizerischen Gesellschaft für Statistik und Volkswirtschaft = Revue suisse d'économie politique et de statistique / publiée par la Société suisse de statistique et d'économie politique 01/2013; 149(1):27-56.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The current study explores whether inconsistent findings in published research regarding the relationship between driver race and discretionary searches can be explained by variation in methodology. A review of published research reveals a consistent pattern between the populations of drivers included in analyses—all drivers or searched drivers—and the relationship between driver race and discretionary searches. This methodological explanation is tested using data from a large municipal agency and confirms that variation in the association between driver race and discretionary searches is based on the population used in analyses. The results and implications are further discussed.
    Journal of Ethnicity in Criminal Justice 07/2013; 11(3):133-149. DOI:10.1080/15377938.2013.739455
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In their article “An Alternative Test of Racial Prejudice in Motor Vehicle Searches: Theory and Evidence,” published in the American Economic Review in 2006, Shamena Anwar and Hanming Fang study racial prejudice in motor vehicle searches by Florida Highway Patrol officers (“troopers”). Their data include the race and ethnicity of the trooper and of the motorist stopped and possibly searched. A search is deemed successful if the trooper finds contraband in the vehicle. Using data on troopers and motorists of three race-ethnicity groups (white non-Hispanic, black, and white Hispanic, with others being dropped), Anwar and Fang compute nine trooper-on-motorist search rates and nine search-success rates. They present a model that exploits this information to test whether troopers go beyond statistical discrimination to racial prejudice. Irrespective of whether troopers exhibit racial prejudice, the model has a crucial testable implication, an implication that concerns the rank-order of the search and search-success rates. Anwar and Fang report that their data neatly fit this predicted rank-order implication with high statistical significance across the board, strongly supporting the soundness of the model. In turn, the model is applied to address the question of racial prejudice. They do not find evidence of racial prejudice, and neither do I—so the present critique does not arrive at results about prejudice contrary to their results. The present critique starts by reporting on my effort to replicate Anwar and Fang’s preliminary rank-order findings. I am unable to replicate two of their nine reported search-success rates, nor can I replicate the reported statistical significance of four of the six Z-statistics and one of the three χ2 test statistics for the rankings of the search-success rates. My new results imply that the empirical support for the model’s soundness is not what Anwar and Fang claim it to be. This problem of irreplicability is my primary point, but I then move on to another matter: My replications draw attention to a neglected statistical caveat in Anwar and Fang’s implementation of the empirical tests of racial prejudice. It turns out that the novel resampling procedure they employ does not provide robust results. I pinpoint the empirical source of this issue and, in an appendix, show how a simple extension to their method improves robustness. In another appendix I put forth an alternative randomization test that seems more appropriate when testing such resampled data.
    Econ journal watch 01/2014; 11(3):250-276. · 1.08 Impact Factor


Available from