The number of people living in the U.S. illegally could be more than double the widely accepted estimate of 11 million, according to a new study by researchers at Yale and MIT.
Unlike previous studies that have concentrated on census statistics, the study used fresh federal data on border arrests, visa overstays, illegal crossings and voluntary emigration. The statistics were used to create a new mathematical model for estimating the number of immigrants here illegally.
A revision of such magnitude could have a profound impact on the debate over immigration policy and border security, crime and public resources.
Mohammad Fazel-Zarandi, a senior School of Management lecturer at MIT who co-wrote the study, said the findings initially worried him.
“I didn’t know if I wanted to do this. I didn’t want it to be political,” he said. “Our goal is to provide a better number, not to do a policy debate.”
Researchers not involved in the new study questioned the reliability of the conclusions, saying they were too broad and may have been based on flawed assumptions.
The widely accepted figure of 11 million comes from a 2016 Pew Hispanic Center study that was calculated using U.S. Census figures subtracting the foreign-born population here legally from the total foreign-born population.
The U.S. Department of Homeland Security used the same methodology in 2014 and came up with a similar figure of 12 million.
The authors of the new study wondered whether they could come up with a mathematical model using several types of raw data released by the U.S government.
The researchers said they were suspicious of a method that was based so heavily on one set of survey results from the Census Bureau. So they looked for additional models for making an estimate.
Using a variety of data compiled from 1990 to 2016, they formed a model that estimated how many people come into the U.S. illegally and subtracted it from the number of people in the U.S. illegally who leave.
The model showed that the population living in the U.S. without authorization in 2016 was at least 16.2 million and that it could have been as high as 29.5 million. The average estimate was 22 million.
“When we were testing out our model, we were thinking it’d be lower than 11 million. We even tried to get to that 11 million number and we couldn’t,” Fazel-Zarandi said.
Other researchers question the findings, saying that if the figure of 16.2 million were accurate, it would mean that the the number of immigrants in the U.S. illegally is 42% higher than the census statistics show — a figure that would seem to defy belief.
The problem with the new approach is that it generates errors that compound over time and that the amount of time studied was lengthy and seemsarbitrary, according to Randy Capps, director of research for U.S. programs at the Migration Policy Institute.
“The big issue with this study is that instead of using the U.S. Census Bureau survey and then comparing it to a model, they use a model” they created themselves, he said. “There’s a lot of uncertainty each year, and so those errors get wider and wider.”
The authors of the new study disagree.
Edward Kaplan, a professor of operations research at Yale, said a wide range of time was used to determine if current estimates made sense.
Kaplan said the research included 1 million simulations with varying assumptions to account for errors.
“We tried to deliberately construct a conservative estimate,” Kaplan said. “We underestimated the rate at which people came into the country and overestimated the people who leave.”
The high estimate could prove welcome to President Trump.
During his presidential campaign, he said without evidence that there could be as many as 30 million people in the country illegally, and he has made cracking down on them a centerpiece of his agenda.
Steven Camarota, director of research at the Center for Immigration Studies, which advocates for tougher border enforcement, said he was skeptical of the study. But he added that, if true, it would show that enhanced security would be a good idea.
“If it’s really that easy to come through the border, then a wall and more agents does make more sense,” he said.
Higher estimates could also undermine arguments Trump has used. For example, a doubling of the population of people in the U.S. illegally would mean that they commit crimes at half the rate previously estimated.
Aarti Kohli, executive director at the Asian Law Caucus, said the study suggests that a majority of those living illegally in the U.S. have been here for more 15 years.
“What this study tells us is that we have many more mixed-status families, and it shows folks that they have integrated into our economy,” Kohli said. “They are working and living in our neighborhoods without us being aware.”