Investigating Federal Economic Incentives for County Jail Involvement in Immigration Detention

Background/Purpose: County jails are indispensable for Immigration and Customs Enforcement (ICE)’s immigration detention operations. Federal law enforcement agency (LEA) grants, such as State Criminal Alien Apprehension Program (SCAAP), may incentivize jails’ participation in immigration detention. SCAAP is not a formal agreement between LEAs and ICE, but it reimburses jails for holding migrants with certain criminal convictions. Investigating such economic incentives may elucidate the persistence of county jail involvement in immigration detention. Although some counties have enacted sanctuary policies that restrict jails’ involvement in immigration detention, research has questioned sanctuary policies’ efficacy. Still, limited research has interrogated economic incentives for immigration detention, focusing instead on analyzing the relationship between LEA political ideology and involvement in immigration detention. In turn, one research question guides this paper: are county receipt of SCAAP in 2014, jail capacity, and presence of a jail sanctuary policy associated with the count of migrants held for ICE in county jails in 2015.

Methods: This project draws on seven publicly available secondary data sources, including Vera Institute of Justice’s Incarceration Trends, Bureau of Justice Assistance’s SCAAP data, Bureau of Labor Statistics for unemployment data, CQ Press for voting data, violent crime data from the Federal Bureau of Investigation, and ICE 287(g) agreement and sanctuary jurisdiction data. The data were merged, cleaned, and analyzed using R software. The dependent variable is a count of migrants held in county jails for ICE. County receipt of SCAAP, presence of a county jail sanctuary, and mean-centered jail capacity—the ratio of the jail’s average daily population to its rated capacity—were independent variables. Covariates included presence of an active 287(g) agreement, unemployment, percent Latinx county residents, percent Latinx residents-squared, violent crime, political majority, and urbanicity. Two negative binomial regression models (NBRM) were developed (N=1,959). Model 1 included all independent variables and covariates, while model 2 added an interaction between receipt of SCAAP and mean-centered jail capacity. Tests of over-dispersion suggested that NBRM was more appropriate than Poisson regression for both models.

Results: Receipt of SCAAP increased the expected number of migrants held for ICE by 189%, B=1.06, SE =.17, p<.001. Neither sanctuary policies nor mean-centered jail capacity significantly predicted the count of migrants detained for ICE. Percent Latinx and Latinx-squared were both significant, B=0.14, SE=0.02, p<.001 and B=-0.01, SE=0.001, p<.001 respectively. The significant, negative quadratic term suggests a concave curvilinear relationship between percent Latinx residents and the count of migrants detained for ICE. In model 2, SCAAP and mean-centered jail capacity interaction was significant and negative suggesting that overcrowded-SCAAP receiving jails had lower rates of immigration detention, B=-1.197, SE =.537, p=.026.

Conclusions/Implications: SCAAP may incentivize jails to hold more migrants for ICE, while sanctuary policies may have minimal efficacy in disrupting such collaboration. The effect of jail capacity on immigration detention rates may depend on receiving SCAAP. By investigating economic structures underpinning immigration detention, this paper offers a novel research contribution and important insight for future policy: policies reducing jails’ use of federal funding incentives may disrupt LEA-ICE collaboration.