In 2025 and 2026, the Trump Administration has issued numerous executive actions and initiatives related to the enforcement of federal immigration laws (some are discussed in CRS Legal Sidebar LSB11299, Recent Executive Branch Actions on Immigration (Part 1)). The 2025 budget reconciliation act (P.L. 119-21) provides increased resources for interior immigration enforcement programs.
This In Focus summarizes seven studies of the effects of two previous immigration enforcement programs on domestic labor markets. For research purposes, a key aspect of both of these programs is that they affected different parts of the United States at different times during the period studied.
Section 287(g) of the Immigration and Nationality Act (INA) permits the delegation of certain immigration enforcement functions to state and local law enforcement agencies. Agreements entered pursuant to INA Section 287(g) (referred to as §287(g) agreements) enable specially trained state or local officers to perform certain functions relating to the investigation, apprehension, or detention of noncitizens under federal oversight by the U.S. Department of Homeland Security's (DHS's) Immigration and Customs Enforcement (ICE). The first §287(g) agreement was implemented in 2002. The number of §287(g) agreements fluctuated over time before expanding in 2025. For more information, see CRS In Focus IF11898, The 287(g) Program: State and Local Immigration Enforcement.
Secure Communities was an information sharing program between local law enforcement, DHS, and the Department of Justice that used biometric data to screen for removable aliens as people were being booked into jails. Such people were non-US citizens who had either been convicted of a crime or had violated immigration laws (such as by entering the country illegally or by overstaying a visa). Secure Communities was initiated in about a dozen jurisdictions in late 2008, with more added over time. Secure Communities was replaced in late 2014 with the Priority Enforcement Program (PEP). It was restored during the first Trump Administration; a decision reversed by the Biden Administration. During 2008–2014, about 21% of the 454,000 people deported from the United States through this program had not been convicted of any crime, 19% had been convicted of violent crimes, and 61% had nonviolent criminal convictions.
The authors of the studies presented below argue §287(g) agreements and Secure Communities programs could have reduced the number of people working without work authorization in three possible ways. First, these programs led directly to deportations. Second, evidence suggests they affected where people without work authorization chose to live by influencing locational choices within the United States, international migration to the United States, or return migration to countries of origin. Third, study authors hypothesize that immigration enforcement programs may have had a chilling effect on employment for people in these locations without work authorization because the programs increased the likelihood that any interaction with local police—even being pulled over for a traffic stop on the way to or from work—could have led to deportation.
The fundamental challenge of measuring the labor market impacts of immigration enforcement is untangling these impacts from other factors happening at the same time. For example, in 2025 and 2026 many changes in immigration enforcement were/are taking place at the same time other federal policies and new technologies may also be affecting labor markets. This makes it difficult to attribute current changes in labor markets to individual policy changes.
The §287(g) agreements and the Security Communities program were implemented in different parts of the United States at different times. This allows researchers to compare areas where §287(g) agreements or the Secure Communities program were implemented with areas in which they had not (or had not yet) been implemented, as if they were experiments with treatment and control areas. Studies using such variation, which was not intended to be an experiment, are called quasi-experimental studies.
The reliability of such studies rests on showing that the treatment and control areas were similar before the treatment began. The studies summarized below show that the areas they compare were similar before the implementation of the immigration enforcement programs.
Overall, these studies find immigration enforcement programs implemented during 2004-2014 were associated with increases in wages in low-wage labor markets, and with mixed results on whether employment for U.S.-born people increased or decreased in low-wage labor markets. The studies generally find these immigration enforcement programs were associated with decreases in employment for U.S.-born people in higher-wage labor markets.
This study examines the impact of the rollout of Secure Communities during 2008-2013 on employment patterns of foreign-born people aged 20-64 with a high school or less education (the authors argue many such people did not have work authorization) and U.S.-born people aged 20-64 of all education levels.
For U.S.-born people, the authors find the local implementation of Secure Communities was associated with small declines in employment and wages overall. The impacts they find for U.S.-born workers vary by education level and occupation: employment increased only for U.S.-born workers with a high school degree or less who worked in the very lowest-paid occupations. The authors find no impact or declines in employment for all other U.S.-born workers, with particularly negative employment impacts for U.S.-born workers in higher-paid occupations, even those with low educational requirements such as construction managers and food service and lodging managers. They find no impact or declines in wages for U.S.-born workers of all education levels in all occupations, with the largest declines in wages among U.S.-born workers with a high school degree or less in higher paid occupations.
This study uses employer-reported payroll tax data to examine industry-level employment and average wages by county during 2004-2010 in 55 counties that signed local §287(g) agreements during 2005-2009, compared with patterns in neighboring counties. It finds no overall county-level employment or wage impacts of implementing §287(g) agreements, but some differences for certain industries. Employment was 7%-10% lower and wages were 3%-4% percent higher in services such as landscaping, janitorial work, and maintenance. Employment in manufacturing was 4% higher, with little difference in wages.
This study uses the rollout of Secure Communities during 2008-2013 to examine impacts on long-term home-based eldercare. It finds that the implementation of Secure Communities reduced the size of the home care workforce by 7.5%, largely from geographic areas with larger numbers of foreign-born Hispanic residents with less than a high school education. The number of U.S.-born home care workers did not increase in these areas.
This study uses the rollout of Secure Communities during 2008-2014 to examine impacts on labor market outcomes for Hispanic foreign-born women with less than a high school education (many of whom had no legal work authorization in the United States) and the U.S.-born people who may be most willing to pay for household help: U.S.-born college-educated women with children under age five.
The authors find that after the implementation of Secure Communities in their county of residence, the number of hours per week that Hispanic foreign-born women with less than a high school education worked in household services declined by 0.4 hours per week, on average, relative to similar women in other counties, while the wages earned by women with less than a high school education working in household services increased. Among U.S.-born college-educated mothers, those who lived in counties that were part of Secure Communities were working 0.4 hours fewer per week, on average, than similar mothers in other counties, with declines in the probability of working full-time and increases in the probability of working part-time or not working for pay at all. As further evidence that this change was due to the reduced availability of childcare, the authors find no similar patterns of changes for U.S.-born college educated women without children or U.S.-born college educated fathers of young children.
This study uses measures of pre-existing local unoccupied jail capacity to predict which jurisdictions chose to sign §287(g) agreements during 2007-2012. It finds that §287(g) agreements predicted by unoccupied jail capacity were associated with significant increases in labor expenditures in agriculture, declines in the number of farmworkers employed per acre, declines in the number of farms, and declines in the amount of land being farmed. These effects were driven by the task force-type of §287(g) agreements, which allowed local officials to check immigration status during any police encounter, including routine traffic stops.
This study examines the impact of signing §287(g) agreements on the share of foreign-born noncitizens in larger counties during 2005-2010 and on agricultural outcomes during 2002-2007 in all counties. It finds signing §287(g) agreements was associated with declines in the share of foreign-born non-citizens, particularly three or more years after the agreements were signed, as well as decreases in the number of agricultural workers hired and in the number of acres planted with vegetables (a labor-intensive crop). Expenditures per agricultural worker increased.
This study uses the rollout of Secure Communities during 2008-2013 to examine impacts in the construction sector. It finds the implementation of Secure Communities was associated with no increase in the U.S.-born construction workforce overall (with some evidence of increased U.S.-born construction employment appearing after two years). In lower-skilled construction occupations, employment of U.S.-born workers increased, but U.S.-born workers did not fully replace the declining number of foreign-born workers with less than a high school education. In higher-skilled occupations, employment of U.S.-born workers declined. Neither group of U.S.-born workers reported higher wages.