Skip to content Skip to navigation

How Strict Immigration Enforcement Harms Schoolchildren

Oct 2018
Policy Brief
By  Thomas S. Dee and Mark Murphy

Earlier this year, the Trump administration’s policy of criminalizing undocumented entry into the United States and separating adult immigrants from their children thrust the nation into a political firestorm. Jarring reports, images, and videos of distraught parents and jailed children cast an unforgettable spotlight on how U.S. policy affects the youngest immigrants. But the forcible breakup of families in such a manner is not the only way immigration enforcement can harm young people. The more subtle but still aggressive application of immigration laws by local jurisdictions can also jeopardize a child’s educational success.

To learn how strict application of immigration laws at the local level affects schoolchildren, we examined a federal initiative to enlist state and local police in the enforcement effort (Dee and Murphy, 2018). Immigration enforcement has historically been a responsibility of the national government. But, in 1996, Congress authorized the federal government to delegate certain immigration authorities to state and local police agencies through partnerships with the U.S. Immigration and Customs Enforcement (ICE) division (sometimes referred to as 287(g) agreements). The first county-level ICE partnership was established in 2005.

The Trump administration views these ICE partnerships as one of its prime tools for ramping up enforcement of immigration laws and has more than doubled the number of active partnerships since taking office (Capps, Chishti, Gelatt, Bolter, and Soto, 2018). Just five days after his inauguration, President Trump issued an executive order expressing his intent “to empower state and local law enforcement agencies across the country to perform the functions of an immigration officer ...,” citing ICE partnerships authorized under section 287(g) of the Immigration and Nationality Act as the vehicle for accomplishing this. (Trump, 2017). By June 2018, ICE had active partnerships with 78 agencies in 20 states, mostly county sheriff’s offices (U.S. Department of Homeland Security, 2018). Through these partnerships, ICE has trained 1,514 state and local police officers to administer immigration laws despite the majority of program costs (i.e., officer salaries and benefits) being borne by the adopting jurisdictions (Capps, Rosenblum, Rodriguez, and Chishti, 2011; U.S. Department of Homeland Security, 2018).

Rigorous Enforcement

In jurisdictions with ICE partnerships, immigration enforcement is notably rigorous. Under the “task force” model, police include immigration enforcement among their regular duties (e.g., traffic stops). Under the “jail enforcement” model, police instead screen for immigration violations among those booked in jail. As a practical matter, the ways in which ICE partnerships change the character of local police enforcement are varied. One of the reasons there were fewer ICE partnerships during the Obama administration was due to the concern that these partnerships encouraged racial profiling and unconstitutional policing (Boushey and Luedtke, 2011; Shahani and Greene, 2009; Skinner, 2010).

It stands to reason that undocumented people — and perhaps Hispanic citizens — would avoid localities where immigration laws are enforced zealously. However, previous research investigating the effects of ICE partnerships has yielded inconsistent results (Kostandini, Mykerezi, and Escalante, 2013; O’Neil, 2013; Parrado, 2012; Watson, 2013). Those studies relied on U.S. Census Bureau survey data, which can be misleading because vulnerable populations sometimes avoid individual surveys or do not answer them accurately (Brown, Heggeness, Dorinski, Warren, and Yi, 2018).

Enrollment Drops

Our approach was different. We used administrative data from the National Center for Education Statistics, which annually publishes comprehensive counts of school enrollment. These data are reported by schools and are likely more reliable than individually reported census statistics for tracking undocumented populations. This was the first study to link these comprehensive student enrollment data with immigration enforcement data to investigate the effects of the ICE partnerships established through section 287(g) on educational outcomes. This research strategy uncovered evidence that, in localities where ICE partnerships were adopted between 2005 and 2011, the number of Hispanic children attending public schools dropped significantly.

School enrollment data doesn’t specify the immigration status of schoolchildren or their parents. But we know that about 80 percent of undocumented immigrants in the U.S. during our study window were from Latin American countries (Passel, 2005). Thus, we can get a good proxy for undocumented population trends by looking at enrollment of Hispanic children. We estimate that, in places where these ICE partnerships were in effect, the number of Hispanic children in public school dropped 7.3 percent, strong evidence that many Hispanic families left those areas. The decline was most pronounced at the elementary school level, where Hispanic enrollment fell 9.4 percent. Moreover, the drop in Hispanic enrollment became increasingly severe (i.e., a decline of 4.8 percent in the adoption year, 7.6 one year later, and 9.7 percent two or more years after adoption) the longer the agreements were in place. By contrast, non-Hispanic enrollment was largely unaffected by the adoption of ICE partnerships.

Crucially, during our study period, more than three-quarters of the children of undocumented immigrants were U.S. citizens, having been born in the U.S. It is probable then that a large majority of the children who left schools in localities with ICE partnerships were U.S. citizens themselves.

We don’t know where these young people went, though we can presume their families moved to places within the U.S. where policing of immigration was less strict (Watson, 2013). We can say with confidence that when their families moved to get away from aggressive immigration enforcement, the education of these children was put at risk.

Academic Performance

Extensive research has found that when students change schools because of reactive moves — that is, moves made “to escape a bad situation”— academic performance often suffers (Rumberger, Larson, Ream, and Palardy, 1999; Welsh, 2017). The effects are intensified if students move several times or in the middle of the school year (Xu, Hannaway, and D’Souza, 2009). And minority students register greater achievement losses than white students following reactive moves, especially in mathematics (Hanushek, Kain, and Rivkin, 2004; Xu et al., 2009).

We estimate that at least 320,000 Hispanic students — and probably many more — were displaced following adoption of ICE partnerships in the communities where they lived. These moves are likely to have reduced student achievement in mathematics in a meaningful way. Academic performance probably fared even worse for those students who moved multiple times or during the school year. More research is needed to precisely determine how much harm these reactive moves caused. Nevertheless, it is clear that ICE partnerships had the unintended consequence of damaging the educational achievement of many Hispanic students.

Our study also looked at how ICE partnerships affected pupil-teacher ratios and the percent of students eligible for the National School Lunch Program. We found no evidence that ICE partnerships had any impact on either. The stability of pupil-teacher ratios is important, indicating that the number of teachers employed fell at about the same rate as Hispanic student enrollment. Schools receive various types of funding based on the number of students they enroll. This is evidence that school districts were cutting back as a result of the funding lost as Hispanic enrollment declined. Furthermore, the stability of the percent of students eligible for the National School Lunch Program indicates that the socioeconomic status of the remaining peer group was not substantially altered. To summarize, we found unambiguous evidence that ICE partnerships significantly reduced the number of Hispanic children attending public school in localities where agreements were in place without providing clear benefits to the students who remained.

Policy Implications

This research has a direct bearing on policy in a number of areas. Of course, the most important policy implications concern the ICE partnerships themselves and, by extension, all strict application of immigration laws. In evaluating these partnerships — or any immigration enforcement regime — policymakers should consider not only the direct effects, that is, the number of undocumented immigrants detained and subjected to the enforcement process, but also the indirect effects on families and communities.

Education is a prime case in point. Aggressive policing of immigration can set back the educational achievement of young people — including many U.S. citizens — whose families are forced to move against their will. These children, who in many cases are already at a disadvantage for reasons of language and income, could face the additional obstacle of having to adjust to new schools in new places. Furthermore, extant literature indicates that ICE partnerships do not reduce crime rates and have negative effects on economic outcomes, such as housing foreclosures, employment, and agricultural yields (Forrester and Nowrasteh, 2018; Kostandini et al., 2013; Pham and Van, 2010; Rugh and Hall, 2016).

The fact is that vigorous immigration enforcement increasingly is a reality throughout the United States, especially in communities with ICE partnerships. The effects on public schools are profound and educators must be prepared to respond. Counseling and other forms of intervention may be necessary to support the academic achievement of children who live in mixed-status households. In particular, a community’s adoption of an ICE partnership should be a signal that these children will need special attention.

The effects may also be felt far away from the localities where immigration enforcement is most aggressive. Undocumented families that stay in the U.S. are likely to move to communities where policing is more lenient. Schools in those places may find they are enrolling more children from mixed-status families and may have to bolster language education and other programs designed to meet the needs of these students.

When immigration enforcement becomes stricter in a society in which educational attainment is the single most important factor determining life circumstances, children from mixed-status households may find educational success to be that much harder to obtain. The question of whether we are willing to jeopardize the education of these young people is something that must be considered in setting enforcement policy.

Though the Trump administration remains interested in expanding ICE partnerships, this amplification of the program is not a foregone conclusion. Decision makers in local communities must also authorize these partnerships; before doing so, these local leaders should closely review the likely unintended consequences of program participation. The educational harm, limited impact on crime, negative economic effects, and locally borne program costs should be strongly considered prior to the initiation or renewal of an ICE partnership.


Boushey, G., and Luedtke, A. (2011). Immigrants across the U.S. Federal Laboratory: Explaining State-Level Innovation in Immigration Policy. State Politics and Policy Quarterly, 11(4), 390–414.

Brown, J. D., Heggeness, M. L., Dorinski, S. M., Warren, L., and Yi, M. (2018). Understanding the Quality of Alternative Citizenship Data Sources for the 2020 Census (U.S. Census Bureau, Center for Economic Studies, Discussion Papers Nos. 18–38).

Capps, R., Chishti, M., Gelatt, J., Bolter, J., and Soto, A. G. R. (2018). Revving Up the Deportation Machinery: Enforcement and Pushback under Trump. Washington, D.C.

Capps, R., Rosenblum, M. R., Rodriguez, C., and Chishti, M. (2011). Delegation and Divergence: A Study of 287(g) State and Local Immigraton Enforcement (Vol. 2). Washington, D.C.

Dee, T., and Murphy, M. (2018). Vanished Classmates: The Effects of Local Immigration Enforcement on School Enrollment (NBER Working Paper Series No. 25080). Cambridge, Mass.

Forrester, A., and Nowrasteh, A. (2018). Do Immigration Enforcement Programs Reduce Crime? Evidence from the 287(g) Program in North Carolina (CATO Working Paper No. 52).

Hanushek, E. A., Kain, J. F., and Rivkin, S. G. (2004). Disruption versus Tiebout improvement: The costs and benefits of switching schools. Journal of Public Economics, 88(9–10), 1721–1746. 

Kostandini, G., Mykerezi, E., and Escalante, C. (2013). The impact of immigration enforcement on the U.S. farming sector. American Journal of Agricultural Economics, 96(1), 172–192.

O’Neil, K. S. (2013). Immigration Enforcement by Local Police under 287(g) and Growth of Unauthorized Immigrant and Other Populations (SSRN Working Paper).

Parrado, E. A. (2012). Immigration Enforcement Policies, the Economic Recession, and the Size of Local Mexican Immigrant Populations. The ANNALS of the American Academy of Political and Social Science, 641(1), 16–37.

Passel, J. S. (2005). Estimates of the size and characteristics of the undocumented population. Pew Hispanic Center. Washington, D.C.

Pham, H., and Van, P. H. (2010). The Economic Impact of Local Immigration Regulation: An Empirical Analysis. Cardozo Law Review, 32(2), 485–518.

Rugh, J., and Hall, M. (2016). Deporting the American Dream: Immigration Enforcement and Latino Foreclosures. Sociological Science, 3, 1053–1076.

Rumberger, R. W., Larson, K. A., Ream, R. K., and Palardy, G. J. (1999). The Educational Consequences of Mobility for California Students and Schools. Policy Analysis for California Education. Berkeley, Calif.

Shahani, A., and Greene, J. (2009). Local Democracy on ICE: Why State and Local Governments Have No Business in Federal Immigration Law Enforcement. A Justice Strategies Report.

Skinner, R. (2010). Department of Homeland Security Office of Inspector General: The Performance of 287(g) Agreements. Washington, D.C.

Trump, D. (2017). Executive Order 13767 of January 25, 2017: Border Security and Immigration Enforcement Improvements. Federal Register, 82(18), 8793–8797.

U.S. Department of Homeland Security, I.C.E. (2018). Delegation of Immigration Authority Section 287(g) Immigration and Nationality Act. Washington, D.C. Retrieved from

Watson, T. (2013). Enforcement and Immigrant Location Choice (NBER Working Paper Series No. 19626). Cambridge, Mass.

Welsh, R. O. (2017). School Hopscotch: A Comprehensive Review of K–12 Student Mobility in the United States. Review of Educational Research, 87(3), 475–511.

Xu, Z., Hannaway, J., and D’Souza, S. (2009). Student Transience in North Carolina: The Effect of School Mobility on Student Outcomes Using Longitudinal Data (CALDER Working Paper Series).