Adriana M. Cerón, Ángel A. Escamilla García, Erin R. Hamilton, Stephanie Hong, Agustina Laurito, Tina Law, Benjamin Meadows, and Carolyn Sattin-Bajaj are contributors to RSF: The Russell Sage Foundation Journal of the Social Sciences issue “The U.S. Deportation System and Its Aftermath,” edited by Caitlin Patler (University of California, Berkeley) and Bradford Jones (University of California, Davis). The issue sheds light on various dimensions of the increasingly punitive U.S. deportation system and the many ways it harms individuals and communities. Responses have been edited for length and clarity.

Listen to the interview.

Q. How did Congressmembers, on both sides of the aisle, react on Twitter (now X) to the Trump administration’s implementation and suspension of the family separation policy in 2018? What does this tell us about deportation policymaking in the current political landscape?

Law: You might remember that in 2018, the Trump administration implemented and then subsequently suspended the family separation policy, which authorized federal authorities to prosecute migrants and then separately detain parents and children while they waited for criminal proceedings and deportation. This policy, at the time, was widely publicized and met with immediate and strong public disapproval. For social scientists, like myself, this policy also represents an important case in which the roles and responsibilities of the executive branch in Congress—as it relates to deportation policy making—are really kind of put to the test. In the United States, Congress has primary responsibility over immigration and deportation policy, though the executive branch has become much more active in this area in recent decades. What this means is that Congress is responsible for enacting deportation policies, as well as checking the executive branch when they overreach by, for example, calling for oversight, initiating investigations, and publicly criticizing their actions. So, the question is, when the executive branch does something to assert power on deportation policy, what will Congressmembers do?

In my study, I examined Congressmembers' activities on Twitter, now called X, during this period because social media plays an increasingly important role in contemporary American politics. And what I find is that there was actually no change in Congressmembers' likelihood of tweeting about deportation and immigration after the family separation policy was implemented. In other words, both Democratic and Republican Congressmembers were not more likely to tweet about these topics when the policy was implemented. However, I find that Congressmembers were actually less likely to tweet about these topics after the policy was suspended. When I examined Democratic and Republican Congressmembers separately, I find that it was actually Democratic Congressmembers who were less likely to tweet about these topics when the policy was suspended.

What I think these findings tell us is that Congressmembers, on both sides of the aisle, are unlikely to publicly respond to executive assertions of power when it comes to matters of deportation, even when these actions are highly unpopular among the American public. Why Congressmembers, particularly Democratic Congressmembers, did not respond is still unclear, but it seemed to have to do with more than just political polarization. So, even though the family separation policy itself ultimately failed—it was suspended by the Trump administration after 45 days—the lack of response by Congressmembers to this executive assertion of power may have more lasting consequences. We are likely to see more executive action on deportation policy in the next few years, which will make interventions by Congress both more necessary but also more difficult.

Q. Can you speak about the relationship between home-country natural disasters and deportations of undocumented migrants from areas experiencing those disasters? Why might this be?

Laurito: Thank you for asking this question. In our paper, we show that deportations increase after countries experience salient natural disasters, which we defined as disasters that, over a one-year period, affected or killed more than 1 percent of that country's population. So, this is a cumulative measure of magnitude that accounts for all disasters that happened in a given country over a year. We hypothesize that we may see an increase in deportations or removals through two main channels. On the one hand, if these shocks induce undocumented migration to the United States, we may see an increase in removals because there are more immigrants that could be subject to deportation. On the other hand, if immigrants already living in the United States change their behaviors and actions in response to these shocks, we could see an increase in removals if these changes increase immigrants' vulnerability to detection by immigration enforcement authorities. This last idea is informed by research showing that immigrants live transnational lives and maintain connections with their home countries through remittances, perhaps engaging with hometown associations. They stay in touch with friends, family, among other things.

In the paper, we explore these two mechanisms. We examine the first mechanism by focusing on recent undocumented migration. So, we calculate the share of recent likely undocumented immigrants using the American Community Survey (ACS). Now the ACS does not include information about legal status, so we have to impute this information. After we do that and identify likely undocumented immigrants, we define recent immigrants as those present in the U.S. for less than three years. Using this measure, we do not find that salient home country natural disasters increase the share of likely undocumented immigrants in the U.S. for less than three years. But there are some caveats to these results. Mainly, because we use the ACS, we're focusing on immigrants already in the United States. These are likely undocumented immigrants that left their countries, were able to enter the United States and evade detection. To the extent that—after the saliant home country natural disasters—there are migrants who leave but then perhaps get detained at the border, those migrants would not be captured in our measure. That's important to consider when interpreting this result.

To examine the second mechanism, we focus on employment outcomes. The thinking is that, if these shocks induce immigrants in the United States to increase or start sending financial assistance to their home countries, some of them have to adjust their labor outcomes, for example, by entering the labor force or working more. We also use the ACS to investigate this, and in this case, we find some evidence of increases in employment and labor force participation. We do some supplementary analysis to analyze remittances, and we find some evidence that these also increase after salient natural disasters. However, our data is not limited to remittances or transfers originating from the United States—it includes transfers from all countries, so, that's important to take into account also when interpreting these results.

Overall, our findings suggest that immigrants already in the United States may change or adjust their behaviors in ways that increase the probability of detection, apprehension, and eventual removal, though we cannot specifically identify the ways in which this happens. For example, is this happening while immigrants are going to work, while they're at work, or while they visit a location to send remittances home? We can't speak to that in the paper. That said, I do think this last set of analysis speaks more broadly to the need for more research that tries to understand how home country shocks, in this case, salient natural disasters—but it could be other type of events—change the behaviors of immigrants or induce changes in behaviors that have repercussions for immigrants' lives in the United States.

Q. Can you speak about how local immigration enforcement affects educational outcomes for elementary and middle school students? Do these effects differ by group?

Meadows: Our research looks at the impacts of direct local ICE enforcement upon two subgroups of students that we think are going to be impacted: elementary and middle school students. We do this by using a Freedom of Information Act request that gave us daily, per individual, ICE records of apprehensions in a local field office matched to all the school districts that students were attending. And what we document is this classic finding across education, health, and other outcomes: ICE enforcement chills some of the behaviors in a local population, and, in this case, student test scores. We document that student test scores drop by 0.27 to 0.52 of a standard deviation, which is a decent drop. If you can think about this from like human capital formation, these students are showing less progress in their achievement scores when they're exposed to ICE enforcement. Our study is an addition to the literature showing the negative externalities associated with ICE enforcement at the local level.

Q. Can you briefly speak about how exposure to immigration enforcement—such as workplace raids—impact the educational and career trajectories of students who live nearby?

Sattin-Bajaj: There are a multitude of consequences for students associated with immigration enforcement, and that includes both actual enforcement episodes or events and the threat of them. Today, I'm going to talk about the findings from the paper I co-wrote with Jacob Kirksey that was published in the RSF special issue on deportation, that focused on actual events, in particular, a specific large-scale workplace raid at a trailer manufacturing plant that happened in North Texas in August of 2018. We have powerful evidence showing how exposure to workplace raids—and general ICE activity that results in apprehension and detention—has really negative effects on students' attendance, academic performance, postsecondary enrollment trajectories, and school mobility. Furthermore, students' sense of safety at school and mental health is directly affected by exposure to enforcement. Although we didn't cover those outcomes in our RSF paper, we do have earlier work looking at students in California schools that shows a decline in sense of safety when there's increased ICE activity nearby.

We found that absence rates among Latinx and multilingual learner students enrolled in schools in the four counties located nearest to the trailer plant spiked dramatically in the six-week period following the raid, which also coincided with the start of the school year. We hypothesized that students in schools in these counties were both more likely to have family members, neighbors, or other members of their communities working at the plant and/or directly affected by the ICE event and they were more likely to have heard about it.

The consequences for student achievement due to exposure to immigration enforcement actions are similarly severe. Immigration enforcement can impact student performance on standardized tests, course taking, and college enrollment patterns. Although we didn't focus on those outcomes for the RSF special issue paper, we do have another paper published in the American Behavioral Scientist, based on the same raid, that measured declines in reading achievement among all students in North Texas attending the schools in the counties closest to the site of the workplace raid. For the full sample of students, exposure to the raid was associated with a 1.75 percentage point lower difference in reading scores compared to similar students in other schools. We call the students at other schools the synthetic control group because that's the method we used to develop a comparison group to try to show the causal impact of exposure to the raid. For the Latinx students enrolled in schools in those four nearby counties, there was a five-percentage point difference in reading scores, and math scores for Latinx and multilingual learner students in those schools also declined after the raid.

Significantly, the negative effects of exposure to the workplace raid did not end there for students in those schools. We actually tracked longer range consequences associated with having been in nearby schools at the time of the raid. We found serious disruptions to students' post-secondary pathways. First, we saw a marked decline in the rate of direct enrollment to four-year colleges among students in counties affected by the workplace raid. We didn't have any evidence of decline in enrollment in two-year colleges, but the decline in direct enrollment to four-year colleges was particularly notable and severe among Latinx and emergent multilingual learner students. We also found a notable reduction in college readiness, which was measured by a decline in the percentage of students achieving high scores on readiness indicators, and a decline in the percentage of students completing advanced academic coursework after the large-scale workplace raid. Next, we saw a large decline in student applications for financial aid, both FAFSA and TAFSA, the Texas State Financial Aid Application, starting in 2019.

Finally, we found an increase in the likelihood of students in schools in affected counties working during high school. That increase was in contrast to the either steady holding trend or the declining trend of working during high school among students in the comparison group. We took that to signal increased economic pressure or shifts in priorities among students and families who were most directly affected by the raid. These are some of the most notable results that we found in our paper, looking first at the more immediate effects of achievement and test scores after the raid, and then the middle range outcomes and the effects on students' postsecondary enrollment patterns.

Q. What is the relationship between intensified immigration enforcement and safety net participation among mixed-status families? How does this relationship vary based on the undocumented parent’s length of residency in the U.S.?

Hong: Thank you so much for inviting our team to share our paper and talk about how intensifying immigration enforcement can affect safety net participation and the role of durations in the U.S. in that relationship. In the past few decades, interior immigration enforcement has increased significantly in the U.S., and the enforcement activity has been devolved to states and locales. So, deportation threats have been brought directly into our communities, and undocumented immigrants have been exposed to deportation threats in their daily lives. While undocumented immigrants are generally ineligible for federal safety programs, their citizen children are generally eligible. However, their access to benefits depends on parents' willingness and ability to go through applications, provide documentations, and engage with state agencies. Under intensified immigration enforcement, undocumented parents may avoid engaging with state agencies, particularly if they are recently arrived immigrants who are not familiar with immigration enforcement systems and safety policies in the U.S.

Dr. Marci Barra, Dr. Angela Garcia, and I got together to work on this paper to examine how everyday exposures to deportation threat are related to children's safety net participation, and how this might vary by durations in the U.S. These children were living in Latino mixed-status families, where their mothers were likely undocumented, and they were all living in California. We first conducted our analysis in all Latino mixed-status families, regardless of their durations in the U.S. We found some suggestive evidence, but not definitive evidence, that higher deportation threats are associated with lower participation in early care and education programs like Head Start and state preschool programs—effect sizes were moderate, about 20% reductions in participation rates. We also examined other programs like SNAP and WIC, which are both food assistance programs, Medicaid, public health insurance, and TANF, which is cash assistance, however, these programs did not show significant reductions under higher deportation threats.

But when we looked at this question by durations in the U.S., we found interesting variations in our results. We found that children of undocumented Latina mothers who have arrived in the U.S. in the past four years were most vulnerable to deportation threats. It appeared really consistently across different models that that they show lower participation in TANF and Medicaid under higher deportation threats. However, children of undocumented Latina mothers who have stayed in the U.S. for longer periods did not show such reductions in Medicaid and TANF participation. Combining all of these findings together, we concluded in our paper that deportation threats can have negative spillover effects onto undocumented parents’ citizen children, who are eligible for safety net programs, but that the story varies by types of programs and also durations in the U.S. Our findings particularly demonstrated that recently arrived mothers may be especially vulnerable to higher deportation threats. This might be because recently arrived immigrants are not familiar with U.S. safety net programs and immigration systems. But as they live in the U.S. for a longer period of time, their experience and learning will accumulate over time, and they may be more likely to apply for public programs despite deportation threats.

Another thing to consider when interpreting our study findings is that we conducted our analysis in the context of California. California is pretty accommodating to undocumented immigrants, and so our findings may not generalize to other states, especially those that have more restrictive immigration policies. Those states might potentially show larger reductions in program participation under deportation threats.

Why do we have to care about the lower participation in public programs? There's a large body of literature showing that early care and education programs, cash assistance, and public health insurance, like Medicaid, can have short- and long-term benefits for children's health and development. So, by missing out on the benefits, children would be losing an opportunity to thrive, and as a society as a whole, this could create costs in the short and long term.

A practice and policy implication of our findings is that recently arrived immigrants may particularly benefit from targeted outreach programs and multilingual support services that could help them navigate safety net programs, but only to the extent that these programs do not provide any threat to their safety. There were many things that I could not really cover in this short interview. Our team, including myself, would be more than happy to answer any questions, and we would love to further connect. Thank you so much for listening.

Q. What is de facto deportation? How many people have been de facto deported? Can you talk briefly about the experiences of de facto deportees?

Hamilton: De facto deportation is when a person migrates out of a country to accompany a family member who has been deported. People who are de facto deported are not formally removed by the law, but they are still forced to move if they want to stay together with their deported loved one. When someone is deported, the family faces three possible outcomes. The first is separation. This is when the family stays in the country they immigrated to, and the deported person lives apart from them in another country. The second is reunification in the country of immigration, but that is rarely possible, at least in the United States. Most people who are deported from the United States are legally barred from returning for five to twenty years, and re-entering the United States without permission after deportation is a federal felony. The third outcome is reunification in the country of deportation. This is when the family leaves to join their deported loved one, and this is de facto deportation.

My colleagues and I study de facto deportation from the United States to Mexico. More people are deported from the United States to Mexico than to any other country, and many of them are long-term settled immigrants with families in the United States. Using Mexico's 2020 census, we estimated that more than 11,000 people were de facto deported from the United States to Mexico between 2015 and 2020. Most people who were de facto deported from the United States to Mexico during this period were children who accompanied deported parents, and most of those children were born in the United States.

U.S.-born children who were de facto deported to Mexico experienced significant disadvantages there. In our research, we found that de facto deported U.S.-born children in Mexico in 2020 were less likely to be enrolled in school or to have health insurance. For example, 70% of de facto deported U.S.-born children were uninsured, so they lacked any form of health insurance coverage. This is compared to 53% of U.S.-born children who migrated for other reasons and only 14% of Mexican-born children. In the United States, in 2019, for example, 33% of Mexican-born children were uninsured, and only 10% of U.S.-born children of Mexican-born parents were uninsured. So, de facto deported U.S.-born children in Mexico experience much, much higher rates of uninsurance than any of their counterparts. We also found that de facto deported children in Mexico were more likely to live in homes built with precarious materials or crowding. Research by other scholars has found that de facto deported children in Mexico have higher rates of mental health problems, such as internalizing and externalizing behavioral problems and parent-child conflict. These disadvantages are likely due to the forced nature of the family migration. When a parent is deported, where a child lives—and whether that child moves at all—is inextricably tied to keeping the family together. When facing deportation, families can no longer choose where their child lives based on what's best for the child, at least not without being separated, and when that choice is taken away, children are much less likely to grow up in the best possible circumstances.

Q. How does separation from family in the U.S. influence deported Northern Central American migrants’ motivations to return to the U.S.? How does this impact their experiences migrating back to the U.S.?

Cerón: In 2019, we met Luis, a Salvadoran man transiting alone through southern Mexico, moving along dangerous routes to avoid immigration checkpoints and reach the U.S. border as quickly as possible. His goal was clear: to reunite with his family in the U.S. following his deportation a year and a half earlier. Luis explained that he had three young daughters and a wife waiting for him in South Carolina and that, despite having been deported four additional times while attempting to return, he was not afraid to keep crossing because his life was there. Luis's story reflects the reality of many deported parents. Indeed, one major consequence of deportation is that it divides loved ones across national borders, often creating serious emotional and financial harms. Children, for example, may experience depression and anxiety, while the sudden loss of household income can leave mothers facing housing insecurity. All of this unfolds alongside the pain and uncertainty of not knowing when or whether families will be together again.

In our study, we argue that deportation is not the end, but rather the beginning of a series of challenging decisions for families. Given the legal barriers that make family reunification nearly impossible, deported parents, like Luis, may come to see returning to the U.S. undetected as one of the few viable ways to be with their families again and put an end to the suffering caused by their separation. But doing so carries serious risks. Unauthorized reentry after deportation is a felony under U.S. immigration law, punishable by imprisonment. And for the Central American migrants in our study, returning also means retracing the treacherous journey to the U.S.-Mexico border. Over the years, that journey has become more costly and dangerous as U.S. immigration enforcement has externalized into Mexico.

What we find is that deported Central American parents often navigate two opposing forces: the strong desire to reunite with loved ones and the harsh realities of enforcement and violence along the journey. To examine both sides of this experience, we bring together cross-sectional survey from the Encuesta Sobre Migración en la Frontera Sur de México (EMIF Sur) for demographic and interview data on deported Central Americans from El Salvador, Guatemala, and Honduras, a group with disproportionate exposure to the U.S. deportation system. Our survey data show a clear relationship between family separation and intentions to return to the U.S. We find that deported migrants who were separated from both migrant children and a partner in the U.S. were significantly more likely to say they intended to return in the future than those who had no partner or children there. We also wanted to know how soon they plan to return, specifically whether they intended to do so within the next thirty days. Of those who said they intended to return in the future, single parents with minor children in the U.S. were the most likely to report plans to do so within thirty days.

Escamilla García: Our qualitative findings further help us understand what this urgency looks like during the journey itself. Deported parents transiting north describe the harm their absence was causing their families, especially financially and emotionally. For many, this created pressures to return as quickly as possible, often within days of their deportation. The urgency parents experienced also shaped how they traveled through Mexico. Although migrants were painfully aware of the dangers they faced, they often took calculated risk to keep moving forward, sacrificing rest, traveling through dangerous areas and harsh weather, or relying on people they did not fully trust if it meant reaching the border faster. This urgency was so profound that it persisted even when deported parents experienced violence or were deported again. Luis, for example, had been deported from Mexico three times. Each time meant starting over, losing contact with his family, and asking his wife for more money to continue their journey, but he still saw those deportations as temporary setbacks rather than reasons to give up.

Our study reveals the profound violence that deported migrants face under restrictive immigration regimes. These policies not only separate families across borders but also severely limit legal pathways to reunification, leaving many deported migrants with few alternatives other than taking life-threatening risks in order to reunite with their loved ones. Our study emphasizes the need for immigration policies that move beyond deterrence and punishment and instead place greater emphasis on keeping families together. 

Adriana M. Cerón is a PhD candidate in sociology at Yale University. She is a co-author of the article “Urgent Returns: The Link Between Family and the Remigration Intentions of Deported Central Americans in an Era of Border Externalization” with Ángel A. Escamilla García.

Ángel A. Escamilla García is an assistant professor of sociology at Yale University. He is a co-author of the article “Urgent Returns: The Link Between Family and the Remigration Intentions of Deported Central Americans in an Era of Border Externalization” with Adriana M. Cerón.

Erin R. Hamilton is a professor of sociology at University of California, Davis. She is a co-author of the article, “De Facto Deportation from the United States to Mexico, 2015–2020” with Claudia Masferrer, Angelita Repetto, and Nicole Denier.

Stephanie Hong is an assistant professor in the School of Social Welfare at Stony Brook University. She is a co-author of the article, “Settlement Duration Matters: Deportation Threat and Safety Net Participation Among Mixed-Status Families” with Marci Ybarra and Angela S. García.

Agustina Laurito is an associate professor in the Department of Public Policy, Management, and Analytics at University of Illinois Chicago. She is a co-author of the article “Climate of Exclusion: Spillover Effects of Home-Country Natural Disasters on Immigrant Removals from the United States” with Ashley N. Muchow.

Tina Law is an assistant professor of sociology at the University of California, Davis. She is the author of the article, “Unpacking the Politics of the US Deportation System”.

Benjamin Meadows is an assistant professor of economics at the University of Alabama-Birmingham. He is a co-author of the article, “ICE at the Door, Tests on the Floor: Student Achievement and Local Immigration Enforcement” with Cora Bennett and Virginia Graves.

Carolyn Sattin-Bajaj is an associate professor in the Department of Education at the Gevirtz Graduate School of Education at University of California, Santa Barbara. She is a co-author of the article, “Future, Interrupted: Examining the Impact of a Large Worksite Enforcement Operation on Students’ Educational and Workforce Pathways” with J. Jacob Kirksey.