NBER WORKING PAPER SERIES RACIAL BIAS IN MOTOR VEHICLE SEARCHES: THEORY AND EVIDENCE John Knowles Nicola Persico Petra Todd Working Paper 7449 http://www.nber.org/papers/w7449 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 December 1999 An earlier version of this paper was circulated as CARESS Working Paper #99-06. We thank Hanming Fang, John Ham, George Mailath, Andrew Postlewaite, Raphael Rob, Ken Wolpin and the participants to the University of Pennsylvania Empirical/Applied Theory seminars for helpful comments. We thank the Maryland ACLU for providing us with data and information. Persico and Todd are grateful to the NSF for support under #SBR-9905564 and #SBR-9730688, respectively. The views expressed herein are those of the authors and not necessarily those of the National Bureau of Economic Research. © 1999 by John Knowles, Nicola Persico, and Petra Todd. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
NBER WORKING PAPER SERIES
RACIAL BIAS IN MOTOR VEHICLE SEARCHES:THEORY AND EVIDENCE
John KnowlesNicola Persico
Petra Todd
Working Paper 7449http://www.nber.org/papers/w7449
NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts Avenue
Cambridge, MA 02138December 1999
An earlier version of this paper was circulated as CARESS Working Paper #99-06. We thank Hanming Fang,John Ham, George Mailath, Andrew Postlewaite, Raphael Rob, Ken Wolpin and the participants to theUniversity of Pennsylvania Empirical/Applied Theory seminars for helpful comments. We thank the MarylandACLU for providing us with data and information. Persico and Todd are grateful to the NSF for support under#SBR-9905564 and #SBR-9730688, respectively. The views expressed herein are those of the authors andnot necessarily those of the National Bureau of Economic Research.
© 1999 by John Knowles, Nicola Persico, and Petra Todd. All rights reserved. Short sections of text, not toexceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice,is given to the source.
Racial Bias in Motor Vehicle Searches: Theory and EvidenceJohn Knowles, Nicola Persico, and Petra ToddNBER Working Paper No. 7449December 1999JEL No. J70, K42
ABSTRACT
African-American motorists in the United States are much more likely than white motorists
to have their cars 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
racial 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.
John Knowles Nicola PersicoUniversity of Pennsylvania University of Pennsylvania3718 Locust Walk 3718 Locust WalkPhiladelphia, PA 19104 Philadelphia, PA [email protected] [email protected]
Petra Todd University of Pennsylvania3718 Locust WalkPhiladelphia, PA [email protected]
Table 1Means and Standard Deviations of Variables used in Analysis
(standard deviations in parentheses)
By Race By Sex AllObser-vations Black Hisp. White Other Female Male
Black 0.63(0.01)
1.00(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.64(0.04)
0.63(0.01)
White 0.29(0.01)
0.00(0.00)
0.00(0.00)
1.00(0.00)
0.00(0.00)
0.35(0.04)
0.29(0.02)
Hispanic 0.06(0.01)
0.00(0.00)
1.00(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.07(0.01)
Female 0.07(0.01)
0.07(0.008)
0.00(0.00)
0.09(0.01)
0.22(0.09)
1.00(0.00)
0.00(0.00)
Guilty 0.33(0.01)
0.35(0.02)
0.12(0.03)
0.32(0.02)
0.47(0.11)
0.38(0.05)
0.32(0.01)
Cocaine 0.08(0.01)
0.10(0.01)
0.03(0.02)
0.03(0.01)
0.37(0.11)
0.09(0.03)
0.08(0.007)
Marijuana 0.23(0.01)
0.23(0.01)
0.23(0.01)
0.26(0.02)
0.41(0.11)
0.21(0.04)
0.23(0.01)
Crack Cocaine 0.04(0.005)
0.05(0.01)
0.01(0.01)
0.01(0.004)
0.22(0.09)
0.06(0.02)
0.04(0.005)
Heroine 0.02(0.003)
0.02(0.004)
0.03(0.02)
0.03(0.01)
0.22(0.09)
0.06(0.02)
0.02(0.004)
Morphine 0.001(.001)
0.00(0.00)
0.00(0.00)
0.002(0.002)
0.00(0.00)
0.00(0.00)
0.001(0.001)
Other Drugs 0.01(0.002)
0.00(0.00)
0.00(0.00)
0.01(0.005)
0.00(0.00)
0.01(0.01)
0.02(0.003)
Paraphernalia 0.01(0.002)
0.003(0.002)
0.010(0.010)
0.02(0.006)
0.00(0.00)
0.01(0.01)
0.01(0.002)
Night(12am-6am)
0.43(0.01)
0.46(0.02)
0.44(0.05)
0.35(0.02)
0.51(0.11)
0.47(0.05)
0.43(0.01)
Number ofObservations 1582 1002 97 463 20 117 1465
Table 2aProportion of Vehicles Searched Found with Drugs
by Race/Ethnicity
Not Guilty Guilty
African American 0.66 0.34White 0.68 0.32
Hispanic 0.87 0.11
Table 2bProportion of Vehicles Searched Found with Drugs
by Sex
Not Guilty Guilty
male 0.68 0.32female 0.64 0.36
Table 2cProportion of Vehicles Searched Found with Drugs
by Race/Ethnicity and Sex
Not Guilty Guilty
male African American 0.66 0.34White 0.67 0.33
Hispanic 0.89 0.11Other 0.68 0.32
female African American 0.56 0.44White 0.78 0.22
Hispanic * *Other 100.00 *
Table 3P-values on Pearson Chi-Squared Tests on
Hypothesis that Proportion Guilty is Equal Across Various Groups
Groups χχ2 p-value
race (African American,Hispanic and white)
21.59 <0.001
race (African American,White)
0.97 0.33
sex (male, female) 0.82 0.37
sex and race (AfricanAmerican, Hispanic, white andmale, female)
26.97 <0.001
sex and race (AfricanAmerican, white and male orfemale)
6.29 0.10
Table B.1aProportion of Vehicles Searched Found with Drugs
by Race/Ethnicity
Period 1 Period 2 Period 3Not
GuiltyGuilty Not
GuiltyGuilty Not
GuiltyGuilty
African American 0.66 0.34 0.62 0.38 0.69 0.31White 0.72 0.28 0.71 0.29 0.64 0.36
Hispanic 0.88 0.12 0.87 0.13 0.91 9.00
Table B.1bProportion of Vehicles Searched Found with Drugs
by SexPeriod 1 Period 2 Period 3
NotGuilty
Guilty NotGuilty
Guilty NotGuilty
Guilty
male 0.68 0.32 0.67 0.33 0.69 0.31female 0.63 0.37 0.64 0.36 0.65 0.35
Table B.1cProportion of Vehicles Searched Found with Drugs
by Race/Ethnicity and Sex
Period 1 Period 2 Period 3Not
GuiltyGuilty Not
GuiltyGuilty Not
GuiltyGuilty
male African American 0.66 0.34 0.63 0.37 0.71 0.29White 0.73 0.27 0.69 0.31 0.62 0.38
Hispanic 0.88 0.12 0.87 0.13 0.91 0.09Other 0.70 0.30 0.50 0.50 0.80 0.20
female African American 0.64 0.36 0.52 0.47 0.48 0.52White 0.56 0.44 0.00 100.00 0.79 0.21
Hispanic n/a n/a n/a n/a n/a n/aOther 100.00 0.00 n/a n/a n/a n/a
Table B.2P-values on Pearson Chi-Squared Tests on
Hypothesis that Proportion Guilty is Equal Across Various Groups
Period 1 Period 2 Period 3Groups χχ2 p-value χχ2 p-value χχ2 p-value
race (African American,Hispanic and white)
8.68 0.01 8.90 0.01 9.72 0.01
race (African American,White)
2.41 0.12 2.54 0.11 1.18 0.28
sex (male, female) 0.7225 0.40 0.11 0.74 0.20 0.66
sex and race (AfricanAmerican, Hispanic, white andmale, female)
9.9481 0.04 13.99 0.007 17.00 0.002
sex and race (AfricanAmerican, white and male orfemale)
3.66 0.30 7.47 0.06 8.27 0.04
Table B3Parameter Estimates for Probit Model of Conditional Probability of being ‘Guilty’
Models without Covariates(p-values from Hypothesis Tests shown in footnote)
Variable Model (1)(a) Model (2) (b) Model (3) (c)
Indicator for white -0.46(0.06)
-0.66(0.13)
…
Indicator for black (-0.38(0.04)
-0.32(0.07)
…
Indicator for Hispanic -1.16(0.16)
-1.20(0.32)
…
Indicator for white * time … 0.007(0.004)
…
Indicator for black * time … -0.003(0.004)
…
Indicator for Hispanic * time … 0.002(0.011)
…
indicator for white * period 1 … … -0.58(0.11)
indicator for white * period 2 … … -0.53(0.13)
indicator for white * period 3 … … -0.34(0.09)
indicator for black * period 1 … … -0.39(0.05)
indicator for black * period 2 … … -0.27(0.09)
indicator for black * period 3 … … -0.45(0.09)
indicator for Hispanic * period 1 … … -1.17(0.28)
indicator for Hispanic * period 2 … … -1.13(0.29)
indicator for Hispanic * period 3 … … -1.17(0.28)
(a) P-value from test of hypothesis white = black = Hispanic is 0.0001. P-value from test that white =black is 0.2523.
(b) P-value from test of hypothesis black = white = Hispanic for both intercept and time trend is 0.0001.P-value from test that black = white for both intercept and time trend is 0.0530.
(c) P-value from test of hypothesis that black = white = Hispanic within all time periods is 0.0007. P-value from test that black = white for all time periods is 0.2266.
1995 1996 1997 1998 1999
0.0
0.2
0.4
0.6
0.8
1.0
Proportion of Black Drivers Searched
1995 1996 1997 1998 1999
0.0
0.2
0.4
0.6
0.8
1.0
Proportion of White Drivers Searched
1995 1996 1997 1998 1999
0.0
0.1
0.2
0.3
0.4
0.5
Proportion of Female Drivers Searched
Figure 2
prop
ortio
n
1995 1996 1997 1998 1999
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Proportion of Black Drivers Found with Drugs
prop
ortio
n
1995 1996 1997 1998 1999
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Proportion of White Drivers Found with Drugs
Figure 3
Related Documents
