Incarceration Gap
About
The Incarceration Gap measures the ratio of incarcerated individuals identified as African American and Black to incarcerated White individuals in a county in the U.S. Individuals from both Hispanic and non-Hispanic ethnicities are included. Incarcerated individuals can include both people in jail and prison. According to the Bureau of Justice Statistics, the imprisonment rate for Black U.S. residents was 911 per 100,000 people in 2022 1. In comparison, the imprisonment rate for White residents was significantly lower, at 188 per 100,000 2. This disparity is present across various stages of the criminal justice system, from arrest and charging to sentencing and parole. Black Americans are incarcerated at roughly 5-6 times the rate of White Americans, with nearly 5 times more at the state prisons and 3 times more in jails 3. The Vera Institute of Justice (Vera) has shown that these disparities persist across different geographic locations, including urban and rural areas.
Why is the Incarceration Gap important to the Structural Racism and Discrimination (SRD) Index?
The Incarceration Gap is one of the most important indicators of structural racism and discrimination within the criminal justice system in the U.S., as it highlights the significant disparities in imprisonment rates between Black and White populations. This gap is driven by historical and ongoing biases in law enforcement practices and sentencing policies, particularly those implemented during the war on drugs era, which disproportionately targeted Black communities 4. Furthermore, socioeconomic inequalities, such as limited access to education and employment opportunities, contribute to higher incarceration rates among marginalized groups 5. The long-term consequences of mass incarceration extend beyond individuals to their families and communities, perpetuating cycles of poverty and social disruption 6. Addressing the Incarceration Gap is essential to dismantling systemic racial biases and promoting fairness and equity within the justice system 7.
What is the expected relation to Structural Racism and Discrimination?
A higher value of the Incarceration Gap between Black and White populations contributes to a higher value or score of the SRD Index.
How is the Incarceration Gap calculated?
We obtained data from the Vera Institute of Justice Prison Policy Initiative 8. The data is publicly available.
Data Source
Data
We used the following two variables at the county level.
| Variables* | Year | Unit |
|---|---|---|
| Percentage of Black Population Incarcerated in Jails or Prisons | 1990 | 2000 | 2010 | 2020 | Number |
| Percentage of White Population Incarcerated in Jails or Prisons | 1990 | 2000 | 2010 | 2020 | Number |
* Individuals from both Hispanic and non-Hispanic ethnicities are included.
*For 2020, we used 2018 jail data and 2016 prison data.
Methodology
We calculated the Incarceration Gap using a ratio formula:
$$
RPrBlWhIncar = \frac{PrBlIncar}{PrWhIncar}
$$
Where:
RPrBlWhIncar: Gap in incarceration rates between Black or African American and White populations
PrBlIncar = BlIncar / BlPop15to64: Proportion of incarcerated Black or African American individuals (BlIncar) to the total Black population from 15 to 64 years (BlPop15to64).
PrWhIncar = WhIncar / WhPop15to64 : Proportion of incarcerated White individuals (WhIncar) to the total White population from 15 to 64 years (WhPop15to64).
Missing Data
We replaced missing values (where applicable) by using functions that were based on Tobler’s First Law of Geography, i.e., “everything is related to everything else, but near things are more related than distant things” 9. This law also informs the concept of Spatial Autocorrelation (SA). The variables were highly spatially autocorrelated, and the missing value in a county was imputed using the median value of the eight nearest neighbors or counties. The eight nearest neighbors were identified using the KDTree algorithm in Python’s scipy.spatial module 10 . After imputing missing data, we have 121 counties (including entire states of Connecticut, Vermont, and Delaware, Hawaii, and island counties) with no data in 2020, 103 counties (Connecticut, Vermont, Delaware, and and island counties) with no data in 2010, 30 counties (entire states of Connecticut, Vermont, Delaware, and Rhode Island) with no data in 2000, and 30 counties (entire states of Connecticut, Vermont, Delaware, and Rhode Islands) with no data in 1990.
Limitations
In states with unified corrections systems, both jail and prison populations are reported as prison data. These states include Connecticut, Delaware, Hawaii, Rhode Island, and Vermont. These states do not have separate local (county or city) jails. Instead, the state government runs facilities that house both individuals awaiting trial and those serving sentences. Because of this unified structure, these states do not report separate jail incarceration rates or data at the county level.
References
1. Bureau of Justice Statistics. (2023). Prisoners in 2022 – Statistical tables (NCJ 305125). U.S. Department of Justice, Office of Justice Programs.
2. Bureau of Justice Statistics. (2023). Prisoners in 2022 – Statistical tables (NCJ 305125). U.S. Department of Justice, Office of Justice Programs.
3. Nellis, A. (2021, October 13). The Color of Justice: Racial and ethnic disparity in state prisons. The Sentencing Project.
4. The Sentencing Project. (2018). Report on racial disparities in arrests.
5. Western, B., & Pettit, B. (2010). Incarceration & social inequality. Daedalus, 139(3), 8-19.
6. Clear, T. R. (2009). Imprisoning communities: How mass incarceration makes disadvantaged neighborhoods worse. Oxford University Press.
7. National Academies of Sciences, Engineering, and Medicine. (2022). Reducing racial inequality in crime and justice: Science, practice, and policy.
8. Kang-Brown, J., O. Hinds, E. Schattner-Elmaleh, J. Wallace-Lee. “Incarceration Trends Project: Data and Methods for Historical Jail Populations, 1970-2018,” Vera Institute of Justice, September 2020.
9. Tobler, W. R. (1970). A Computer Movie Simulating Urban Growth in the Detroit Region. Economic Geography, 46(sup1), 234–240.
10. Virtanen, P., Gommers, R., Burovski, E., Oliphant, T. E., Weckesser, W., Cournapeau, D., … & Feng, Y. (2021). scipy/scipy: SciPy 1.6. 0. Zenodo.