The Relationship between Racial Residential Segregation and Black-White Disparities in Fatal Police Shootings at the City Level, 2013–2017

Abstract

Background

Although there is a disparity in the rates of fatal police shootings of Black victims compared to White victims throughout the nation, the magnitude of this racial disparity varies widely between cities. Examining the reasons behind this vast variation in the racial disparity in fatal police shootings across different localities could be helpful in identifying the underlying causes of this problem. Therefore, the objective of this study was to assess potential causes of the Black-White racial disparities in police shootings at the city level.

Methods

Using a cross-sectional design, we examined the relationship between the level of racial residential segregation in a city and the ratio of the rates of fatal police shootings of Black people compared to White people within that city during the period 2013–2017, using data from 69 of the largest cities in the United States. Using linear regression analysis, we modeled the ratio of Black to White fatal police shooting rates as a function of the level of racial segregation in each city as measured by the index of dissimilarity calculated at the Census block level. We controlled for several factors at the city level that might be expected to influence the rate of police shootings by race/ethnicity, including the level of deprivation of the Black population in economic status, the degree of Black representation in the city's police force, the relative size of the police force, crime rates, racial composition of the city, and city population.

Results

The level of racial residential segregation was significantly associated with the racial disparity in fatal police shooting rates. For each one standard deviation increase in the index of dissimilarity, the ratio of Black to White fatal police shooting rates increased by 44.4% (95% confidence interval, 17.1%–71.7%).

Conclusions

Racial residential segregation is a significant predictor of the magnitude of the Black-White disparity in fatal police shootings at the city level. Efforts to ameliorate the problem of fatal police violence must move beyond the individual level and consider the interaction between law enforcement officers and the neighborhoods that they police.

Introduction

Between 2013 and 2018, a six-year period, 6,178 people were shot to death by police officers.¹ Of the victims whose race was identified, 28% were Black,¹ which is twice the proportion of people who are Black in the overall U.S. population. Although the ratio of the rates of fatal police shootings of Black victims compared to White victims is 3.1 for the entire nation, the magnitude of this racial disparity varies widely across states, ranging from a low of 0 to a high of 18.² Examining the reasons behind this vast variation in the racial disparity in fatal police shootings across different localities could be helpful in identifying the underlying causes of this problem. What factors explain why the level of disproportionality in the rate of shooting of Black Americans by police depends so heavily on where a person lives?

We are aware of two studies that have attempted to identify variables that are associated with geographic differences in the Black-White racial disparity in police violence.2, 3 Ross examined racial bias in police shootings at the county level during the period 2011–2014 and found that county population, income, percent Black population, and financial inequality were associated with racial differences in the rate of fatal police shootings of unarmed victims.³ Mesic et al. reported that racial residential segregation, along with racial disparities in economic and employment status, were significant predictors of the ratio of the rate of fatal police shootings of Black compared to White victims not known to be armed.²

A major limitation of these previous studies is that they were conducted at the state or county level. States and many counties encompass a diversity of police departments; therefore, these categorizations are too broad to shed light on why individual police departments act differently from one another with respect to their treatment of suspects based on race. Because most, if not all cities are served by a single police department, investigating fatal police shootings at the city or neighborhood level will allow for a more finely-grained analysis.

Recently, Johnson et al. published the first study to examine, at the zip code level, factors related to differences between racial groups in the probability of having a fatal encounter with the police.⁴ They found that higher levels of residential segregation and income inequality increased the relative odds of a Black male being killed by the police compared to a White male. However, it is not possible from this study to identify the racial disparity in fatal police shootings at the city level. Calculating and reporting the city-level differences in racial disparities in police shootings is essential for two reasons: (1) city leaders cannot appreciate if there is a problem and how severe it is unless they know the magnitude of the racial disparity in fatal police shootings in their city and how it compares to other cities; and (2) policies and interventions to ameliorate this problem are implemented at the city level so documenting the causes of racial disparities in police shootings at the city level is necessary to inform the development of effective policies and interventions. While we do not hypothesize that the findings will be different at the city-level than the zip-code level, documenting that factors operating at the neighborhood level explain differences in racial disparities between cities is necessary before these findings become actionable by city officials.

To the best of our knowledge, no previous national study has examined or characterized differences between cities in the racial disparity in fatal police shootings. This study has two aims: (1) to calculate and report differences in the ratio of Black to White fatal police shootings among 62 of the largest U.S. cities in order to identify those cities in which intervention is most urgently needed; and (2) to extend previous work by examining the relationship between racial residential segregation and other city-level factors and racial disparities in fatal police shootings at the city level. Understanding why certain cities have greater racial disparities in fatal police shootings than others may help identify interventions—that can be implemented at the city level—to reduce these disparities.

This paper uses the theoretical framework of the racial threat hypothesis, which was applied by Liska and Yu to the police use of deadly force.⁵ According to this hypothesis, racial residential segregation amplifies the perceived threat and diminishes the restraint of police working in neighborhoods that are predominantly black, leading to an increased rate of police homicides in these neighborhoods.⁵ For this reason, racial segregation is the primary predictor variable examined in our analysis.