Secure Communities: Broad Impacts of Increased Immigration Enforcement

The establishment of the Secure Communities program led to a notable increase in immigration enforcement throughout the United States, as the program spread from border counties to the interior of the nation. The program, which began in 2008, had the goal of increasing information sharing between local law enforcement and Immigration and Customs Enforcement (ICE) in order to identify and remove undocumented immigrants, and this resulted in over 450,000 people deported between 2008 and 2014. There is growing evidence that the impact of increased enforcement and deportations went beyond those immigrants directly affected, and had ripple effects on the employment and safety net participation of citizens.

The impacts of increased immigration enforcement can extend beyond the direct effects on the immigrants deported and their immediate families.

The Facts:

• After rising steadily throughout the 1990s and early 2000s, the undocumented population stabilized and began a gradual decline after 2007. In 2015, an estimated 11 million undocumented individuals lived in the U.S. These overall population numbers reflect trends in new immigrants coming in, as well as the number of individuals leaving the country. A number of policies have been implemented at the local, state and federal level to address the issue of undocumented immigration. One such type of policy is interior immigration enforcement (as opposed to enforcement along the border). Increases in enforcement have in part led to a dramatic rise in immigrant detentions and deportations: between 2003 and the mid 2010s the number of people detained annually increased by 3000% to about 160,000 annual detentions in 2014, and the number of people deported annually increased by 200% to about 300,000 annual deportations in 2014, according to data from the Transactional Records Access Clearinghouse (TRAC). The total number of deportations is the combined result of all enforcement activities — of which Secure Communities was one component. As an example, of the approximately 300,000 deportations that took place in 2014, about 100,000 were interior deportations and about 75,000 of these were due to Secure Communities (see chart.)
• The Secure Communities program automated a system by which, when an individual was arrested and their fingerprints were taken for a background check, the fingerprints would also be sent to ICE to check the individuals’ immigration status. The program was piloted in 2008 under the George W. Bush administration and continued during the Obama administration until late 2014 when it was replaced by the Priority Enforcement Program (PEP). Through executive order, the Trump administration reinstated Secure Communities to replace PEP in January 2017. Interior immigration enforcement resulted in over 80,000 people deported in 2017, the latest data available, and 83% of these deportations were related to the Secure Communities policy (SC).
• Male immigrants from Mexico and from Central American countries were disproportionately represented among those deported through Secure Communities between 2008-2014. 78% of those deported under Secure Communities were from Mexico and 19% were from Central American countries, while, in contrast, among all undocumented immigrants in 2007, before Secure Communities began, 57% were estimated to be from Mexico and 12% were from other Central American countries. Additionally, men were over-represented: 96% of those deported were men, even though men made up only 57% of all undocumented in 2007. A primary goal of the program was reducing crime, however, the share of those deported that had a serious criminal conviction was relatively small: 17% of those deported as part of Secure Communities were not convicted of any crime, and the most serious criminal conviction for 79% of those deported was non-violent— including traffic violations and immigration offenses (such as illegal entry or re-entry or possession of fraudulent documents for 8% of those deported). Researchers have found no evidence that the implementation of Secure Communities had the effect of reducing crime (see here).
• The impacts of increased immigration enforcement can extend beyond the direct effects on the immigrants deported and their immediate families. Qualitative studies show that in addition to direct effects on immigrants detained and deported, Secure Communities had broader “chilling effects” due to fear induced by the policy. For instance, people reported being less likely to interact with authorities for fear of being asked about their immigration status or about the status of people they know. This could be due to mistrust of local law enforcement and government more broadly, possibly due to the lack of targeting serious criminals and to certain demographic groups being over-represented. For example, over 40% of Latino immigrants living in several large cities in the U.S. reported thinking that police officers stop Latin American immigrants without reasonable cause, that they are afraid to leave their home, and that they feel more isolated because local law enforcement is involved with immigration enforcement (see here).
• The rollout of Secure Communities was gradual, as it was implemented county by county over the period 2008 to 2013. This helps researchers to try to isolate the impact of the program by comparing the outcomes in places that adopted Secure Communities early to very similar places that, by virtue of resource constraints and waiting lists, adopted Secure Communities later. There is an emerging literature in economics that uses this approach to evaluate the direct effects of Secure Communities on immigrants, as well as spillover effects onto citizens.
• A policy that aims to remove undocumented individuals could be expected to decrease the participation of these workers in the labor force. Indeed, the implementation of Secure Communities in a location significantly reduced the availability of less-educated immigrant men in the local labor force — and this had negative consequences for high-skilled male citizen workers. Specifically, Annie Hines, Philip Luck, Hani Mansour, Andrea Velasquez and I find that the number of male non-citizens with a high- school degree or less (characteristics that make them most likely to be undocumented) who were employed decreased by an average of 7 percent when Secure Communities was implemented in a local area. This decline could be due to both direct removals, as well as potential “chilling effects” reducing labor force participation among those remaining in the U.S. But, the reduction in employment was not limited to immigrant men. Secure Communities exposure in a local area also reduced high-skilled citizen men’s employment in sectors that have historically relied on undocumented labor. This is likely operating through decreases in demand for high skilled workers when the supply of low skilled workers shrinks: for instance, when the supply of construction workers decreases, there is also less need for construction managers, who are more likely to be high skilled citizens. We calculate that a 1 percent decline in the employment share of likely undocumented male immigrants is associated with a 0.12 percent decline in the employment rate of male citizens. Secure Communities may have increased employment for low-skilled Hispanic citizen men who are closest substitutes for likely undocumented men, however we find that the net effect of Secure Communities on all male workers in a local area is negative.
• Another pathway through which Secure Communities negatively impacted high-skilled citizen workers is through the change in the price of household service workers. Female undocumented immigrants are over-represented in household service work (such as housekeeping and childcare), and my research with Andrea Velasquez shows that Secure Communities also reduced labor supply of female immigrants in household services. Given that women were for the most part not deported under Secure Communities, this effect is potentially due to chilling effects—specifically, immigrant women remain working in the U.S., but reduce their hours worked due to fear of leaving their house and putting themselves or their family at risk of deportation (see here for anecdotal evidence of these responses). By increasing the cost of these household services, increased immigration enforcement also reduces the labor supply of high-skilled citizen mothers, who are most likely to outsource household production. The largest impact we observed was among college-educated citizen mothers with children under age 6 (before children are likely to enter school): mothers of young children experience a 0.8% reduction in the likelihood of working and a reduction of 1.2% in hours worked by as a result of the implementation of Secure Communities in their area. Importantly, being exposed to Secure Communities around a child’s birth has long-run negative effects on mother’s labor supply, with no effects on long-run father’s labor supply, so immigration enforcement may worsen the gender wage and employment gap.
• Secure Communities also reduced safety net participation among citizen households of Hispanic origin. These households are eligible to participate in safety net programs because of their citizenship status, however Marcella Alsan and Crystal Yang have found that the implementation of Secure Communities in a local area reduced the participation of these households in the Supplemental Nutrition Assistance Program (SNAP), otherwise known as food stamps, by 10% and in Supplemental Security Income (SSI) by 30%. These effects are likely due to fear of deportation of friends or family members, which reduce Hispanic citizens’ willingness to interact with the government to apply for or renew public benefits. The effects are larger in local areas with a greater fraction of mixed status households (households with both citizens and non-citizens, the latter of whom are eligible for deportations), in areas where more detainers are issued for non-violent crimes, and in areas with more self-reported deportation fear. Taken with previous findings from my own work, that limiting access to SNAP for children of immigrants harms their long-run outcomes, this suggests Secure Communities could have lasting negative consequences for future generations (see also: "SNAP: Nutrition Aid Can Provide Long-Term Benefits").

What this Means:

Secure Communities resulted in almost half a million deportations during 2008-2014, although there is little evidence that serious criminals were removed and no evidence of meaningful improvements in public safety as a result of the policy. Secure Communities reduced the labor supply of likely undocumented immigrants through both direct removals and chilling effects, which had several negative spillover effects onto labor market outcomes of citizens, and no evidence of benefits to citizen workers on net. Moreover, chilling effects caused Hispanic citizens to reduce their safety net participation. These adverse spillover effects onto citizens and immigrants not directly targeted by Secure Communities should be considered as policy-makers actively change immigration policy.