Deportations Don’t Lower Crime Rates, Study Says
September 3, 2014
By KIRK SEMPLE, New York Times,
Six years after the federal government opened an immigration enforcement program intended to improve public safety, deporting hundreds of thousands of people, many of them convicted criminals, a new study has concluded that the program has had “no observable effect on the overall crime rate.”
The finding “calls into question the longstanding assumption that deporting noncitizens who commit crimes is an effective crime-control strategy,” said the study, conducted by two law professors at the University of Chicago and New York University.
The analysis, scheduled for publication in the November issue of The Journal of Law and Economics, a journal for peer-review research, coincides with the Obama administration’s internal review of the program, known as Secure Communities. Jeh Johnson, the homeland security secretary, has suggested that he might overhaul the program, saying it needs “a fresh start.”
Secure Communities, which began in 2008 under President George W. Bush, became a cornerstone of the Obama administration’s immigration enforcement strategy, allowing officials in the Department of Homeland Security to more easily compare the fingerprints of suspects booked at local jails with those in its files. If the authorities find that a suspect is a noncitizen who is in the country illegally or has a criminal record, they may seek custody of that suspect and begin deportation proceedings.
Yet the program has been plagued with trouble since it began.
Immigrants’ advocates, joined by some officials, have long questioned the crime-fighting impact of the program, complaining that it sweeps up many immigrants who have committed minor infractions, like traffic violations, or are guilty of no crimes at all but are in the country illegally.
The new study – by Adam B. Cox, a professor at New York University’s School of Law, and Thomas J. Miles, a professor at the Law School at the University of Chicago – is the first comprehensive, independent analysis of the program’s efficacy in lowering crime rates, the authors said.
Secure Communities was rolled out over the course of several years, jurisdiction by jurisdiction, allowing the researchers to conduct essentially a randomized, controlled clinical trial to measure the effect of the policy.
Most efforts to study federal policy changes, Professor Cox said, become before-and-after experiments pivoting around one moment in time. But with Secure Communities, he said, “we were able to see the experiment repeated 3,000 times.”
The staggered rollout allowed them to better untangle the effect of the program from factors that also influence local crime rates, like rates of poverty and the size of the police force.
Through Freedom of Information Act requests, the researchers obtained all the “performance data” that Homeland Security collected about the program between 2008 and early 2013, Professor Cox said. In addition, they gathered Census Bureau data as well as crime reports and other municipal policing information from the F.B.I.
They concluded that there was no “empirical evidence” that Secure Communities caused “a meaningful reduction” in the rates of serious crimes.
“This is important because Secure Communities specifically, and criminal deportation policies more generally, have long been publicly justified primarily on grounds that they keep communities safer from violent crime,” the authors said in the study.
The study allowed that the program may have led to small reductions in the rates of motor vehicle theft and burglary. This result, the authors said, suggests that the people who are least likely to be deterred by the threat of deportation are the most serious criminals – the very people the Obama administration says the program is intended to catch.
Though the authors did not send the study to the immigration authorities for review, officials with Immigration and Customs and Enforcement, or ICE, an arm of Homeland Security, defended the program. They said it had helped lead to the deportation of more than 288,000 convicted criminals between October 2008 and May 2014, including more than 113,000 immigrants convicted of major violent offenses, including murder, rape and the sexual abuse of children.
“Secure Communities, by leaps and bounds, has allowed us to get the most egregious violators of our local statutes out of our communities and remove them from the country,” a senior agency official who was not authorized to speak on the record said in an interview. With criminal recidivism rates just below 50 percent, he added, the program likely prevented more than 100,000 people from going out and committing another crime.
“I don’t know how you can say that does not have an impact on community safety,” he added.
But Professor Cox and Professor Miles countered that while the program may have reduced “in absolute terms” the number of crimes being committed, it has not necessarily made communities safer. A more important measure of safety are crime rates – numbers of crimes per population – which measures the likelihood that someone might become a victim of crime, they said.
Professor Cox and Professor Miles acknowledged that while Secure Communities had led to the deportation of immigrants convicted of very serious crimes, the majority of those deported through the program were guilty of misdemeanors – including victimless immigration offenses not directly related to public safety – or of no crime at all.
During the period their data covered – late 2008 to mid-2013 – about 29 percent of the immigrants deported through the program fell into the most serious of Homeland Security’s three categories of criminals: those who had been convicted of at least one aggravated felony or of at least two felonies punishable by more than a year in prison.
“If crime rates fell as a result of Secure Communities, it would support an inference that ICE is removing immigrants who commit serious crimes at high rates,” the professors wrote in an email. “But we found that crime rates are in essence unchanged.”
The findings, they said, seem to underscore much of the existing social science evidence that immigrants on average offend at lower rates than the native-born. “If the folks deported through Secure Communities are as – or more – law-abiding than the average person living in their community, then deporting them won’t necessarily drive down crime rates,” they said. “It could even, in theory, cause crime rates to go up.”
They continued: “The program was labeled ‘Secure Communities’ and promoted as a tool for reducing crime – especially violent crime – and making communities safer. Our paper shows that there is no support for that.”