On July 11, 2019, the Trump Administration released an executive order that requires the Department of Homeland Security, the Department of State, the Social Security Administration and the Department of Health and Human Services to collect citizenship data using administrative records for the federal administration to determine the number of undocumented immigrants in the United States. This executive order was in response to the federal administration’s failed attempt to add a citizenship question to the 2020 census that was blocked by the Supreme Court.

In order to justify the executive order, the Trump Administration cited a 2018 study from three Yale professors who specialize in management studies, not demographics, who estimated the undocumented population to be between 16.2 million and 29.5 million, far higher than the widely accepted estimates of about 11 million.

Researchers interested in the estimated undocumented population in the United States typically look to the reliable and widely accepted estimates from either the Migration Policy Institute (MPI), the Pew Research Center or the Center for Migration Studies (CMS), all three of which estimate the number of undocumented immigrants to be about 11 million in 2017, and the most recent estimate from the Department of Homeland Security is 12 million.[1]

The study by the three professors, Mohammad Fazel-Zarandi, Jonathan Feinstein and Edward Kaplan, used a simulation analysis based on border apprehension rates and voluntary emigration rates, among other data points. That is a fundamentally different approach from the methods CMS, MPI and the Pew Research Center estimates use; their analyses are based on census and survey data, as well as supplemental administrative data.[2] Border apprehension rates are controversial because one person may be apprehended multiple times and the data does not take this into account.

Researchers and demographers have slammed Fazel-Zarandi, Feinstein, and Kaplan study as simply irresponsible. In a commentary by Julia Gelatt, Michael Fix and Jennifer Van Hook from MPI, they note that the professor’s study doesn’t account for circular migration patterns in the 1990s, and there is an error in estimating border crossers in their methods which makes their future estimates inaccurate and it is completely misaligned with the Census.[3]

Robert Warren of CMS, a demographer who specializes in changing human populations, points out that Fazel-Zarandi, Feinstein, and Kaplan’s estimates for the cumulative inflows are too high and the cumulative outflows are too low. He shows that the professors’ study is far off the mark, and that they add an additional and inaccurate 11 million Mexicans to the U.S. population count between 1990 and 2000. Warren points out that the increase of 11 million is completely misaligned with the 5.5 million Mexican emigrants counted in the 2000 U.S. census. Warren also highlights that the professors rely on border apprehension rates to determine the number of Mexican arrivals, which are purely speculative and not based on empirical data. Border apprehensions are a straightforward count, but the rate of apprehension is far less clear, since it is not clear how many people are not apprehended. Complicating their calculations further, many of the apprehensions during the 1990s were migrants returning to their home in the U.S. after visiting Mexico, not new arrivals to the United States. Lastly, Warren notes that professors used the wrong emigration rates to determine the rates of Mexican nationals who left the U.S.[4]

Stepping back from the methodological details, could there really be twice as many undocumented immigrants as other researchers have found? Gelatt, Fix and Van Hook make a clear case that there can’t be that many individuals that are hidden because people leave behind “footprints” in statistical records, such as birth, death, school enrollment, housing, and others. For example, there would be records of their children in school, as well as records of them using state level programs such as obtaining a driver’s license.

Generally speaking, exploring different methodologies for research is a good idea, even if the results run counter to the common wisdom. But in this case, three professors are stepping outside of their field of expertise in order to tread into one of the most politicized issues in the country. Where careful researchers have for decades developed nonpartisan estimates grounded in sound demography, the three professors cast that aside and wind up giving a president who is generally disrespectful of research an inflated number that is double what any other researchers have found. In this political climate, that is not responsible research.

It is important to remember that this population of immigrants contribute to our local economies, many of them have fled hostile conditions, and many have resided in the U.S. for decades and overtime have become deeply embedded in our communities. A 2017 Institute on Taxation and Economic Policy (ITEP) report found that undocumented immigrants contribute an estimated $11.74 billion dollars to local and state taxes in the United States.[5] Another ITEP study found that young immigrants who were eligible for the Deferred Action for Childhood Arrivals Program (DACA) contributed $1.8 billion to state and local taxes. In 2017, the Trump administration eliminated the DACA program and ITEP estimated that the loss to tax contributions was $694 million.[6]

Over half of undocumented immigrants speak English at least well, about one third have at least some college, and they have a high labor force participation rate of 76 percent,[7] which is higher than the national native-born rate of 63 percent.[8] The undocumented population still contributes to the cultural and economic well-being of the United States. Many have been here since a young child and the United States is the only place they know as home.

By: Cyierra Roldan


[1] Center for Migration Studies: http://data.cmsny.org/,  Migration Policy Institute: https://www.migrationpolicy.org/data/unauthorized-immigrant-population/state/US, Pew Research Center: https://www.pewhispanic.org/interactives/unauthorized-trends/, Department of Homeland Security: https://www.dhs.gov/sites/default/files/publications/18_1214_PLCY_pops-est-report.pdf

[2] Fazel-Zarandi, M.M, Feinstein, J.S. and E.H. Kaplan. (2018). The Number of Undocumented Immigrants in the United States: Estimates Based on Demographic Modeling with Data from 1990 to 2016. PLoS One: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0201193

[3] Gelatt, J., Fix, M., and J., Van Hook. (2018). People Leave Footprints: Millions More Unauthorized Immigrants Cannot Be ‘Hidden’ in Data Estimates. Migration Policy Institute: https://www.migrationpolicy.org/news/people-leave-footprints-millions-more-unauthorized-immigrants-cannot-be-hidden

[4] Robert Warren, https://journals.plos.org/plosone/article/comment?id=10.1371/annotation/b53601d7-9b42-450a-b23c-39f4d889fcbd

[5] Institute on Taxation and Economic Policy: https://itep.org/wp-content/uploads/ITEP-2017-Undocumented-Immigrants-State-and-Local-Contributions.pdf

[6] Institute on Taxation and Economic Policy: https://itep.org/wp-content/uploads/2018DACA.pdf

[7] FPI analysis of Center for Migration Studies data: http://data.cmsny.org/

[8] FPI analysis of ACS 2017 5-year estimates.