The New York Times recently highlighted a new analysis of immigration-enforcement data that is sure to be used in the coming months to undermine the initiatives of the Trump administration and Republicans in Congress. The Times article, by staff writer Eduardo Porter, argues that years of “tough” enforcement under Obama failed to improve conditions for working Americans. Been there, done that, as it were, in response to President Trump’s call for tougher enforcement.
This analysis is deeply flawed, for two reasons. First, it uses inappropriate, incomplete, and doctored enforcement statistics to present a misleading picture of recent enforcement trends. Second, Porter relies heavily on a working paper and a forthcoming analysis by pro-immigration economist Giovanni Peri of the University of California, Davis, to argue that enforcing immigration laws does not help natives and in fact harms the economy.
Let’s examine the arguments on enforcement first. Porter states that “President Barack Obama went on a deportation spree in his first term.” To illustrate this “deportation spree,” he includes a bar graph labeled “Immigration Shock.” It claims to show annual interior apprehensions by Immigration and Customs Enforcement (ICE), the leading interior enforcement agency, from the period 2000 to 2015. The graph depicts a distinct “surge” beginning in 2008 and accelerating in 2009 before dropping off in 2013. This is meant to show that enforcement was ramped up considerably in the Obama administration:
The first problem with this is that Porter chooses apprehensions, or arrests, as a metric to illustrate a deportation spree. But an arrest is not the same as a deportation, and is not by itself an ideal metric for measuring the effectiveness of enforcement. Not all those arrested for immigration violations are deported. Why not just show deportations, since those statistics also are available?
Further, especially under Obama’s “prosecutorial discretion” policy, a significant number of the aliens arrested by ICE officers were released after only a short time in custody. Some of these individuals had their charges dropped or thrown out of immigration court; others were let out of custody on bond or under “supervision,” but skipped out on their hearings and melted back into the illegal population. In fact, according to our calculations, the number of interior deportations over the period 2009–2015 (1.15 million) is only 63 percent of the number of interior apprehensions (1.84 million), illustrating that a very large number of those apprehended by ICE in the interior during this time were never deported from the country. If illegal immigrants who are arrested at some point nonetheless remain in the country, then so does their labor-market impact, undermining the enforcement-does-not-help-native-born-workers argument advanced by Porter’s piece.
It gets worse. These statistics aren’t just an unsatisfactory measure of enforcement; they are also inexplicably doctored, in a way that changes the timing and shape of the enforcement surge. The apprehension totals in the bar graph, which are sourced to the Department of Homeland Security and to Peri, do not match the official statistics published on the DHS website. We asked Peri about the discrepancy, and he told us that the Times had made adjustments to the numbers. Specifically, he said, the Times had subtracted from the annual totals any cases where the apprehension location was not specified, which are collectively labeled “Unknown” in the DHS statistical tables. Peri said that the Times wanted to count only interior arrests. But the DHS table makes clear that all arrests it attributes to ICE are interior arrests (as opposed to arrests made by Customs and Border Protection officers, including the Border Patrol). It is not clear what innocent explanation there could be for subtracting these “unknown” cases.
At the most basic level, the problems with Porter’s article stem from the fact that he talked to no one with a point of view different from his own.
When the actual DHS interior-apprehension figures are graphed, the arc of the enforcement “surge” looks noticeably different from the arc of Porter’s graph. The number of interior apprehensions actually spiked dramatically from 2007 to 2008, in the last years of the Bush administration, and then remained fairly flat until 2012, followed by a steep decline in Obama’s second term.
The actual trend line of the surge under Bush is impossible to determine, because the DHS statistical tables used by Porter and Peri are incomplete. A footnote in the official tables states that in certain years (including 2006 and 2007) the numbers include only certain ICE arrests — specifically those from operations targeting fugitives and those made by the Investigations division of ICE. The total number of ICE interior arrests in those years is not known, but is certainly understated in the official statistics published by DHS.
Data quality would not be an issue if Peri and Porter had simply used deportations as the appropriate metric. But, inconveniently for their argument, there was a steady decline in interior deportations under Obama, not a “spree” of interior deportations as Porter asserts. (Unfortunately, reliable and consistent official data on interior deportations prior to 2009 is not available; we obtained the post-2008 data illustrated in the graph directly from ICE, during discovery in a lawsuit.) In every year of the Obama administration, the actual number of people deported from the interior of the United States fell, from approximately 235,000 interior deportations in 2009 to 70,000 in 2015:
Finally, and most germane to the argument pushed by Porter’s piece, the interior enforcement that occurred under the Obama administration was focused almost entirely on criminal aliens targeted by ICE after an arrest on state or local charges — not on worksite-enforcement operations aimed at discouraging the hiring of illegal workers. Under the Obama administration, 87 percent of the aliens removed from the country were convicted criminals, and most were apprehended in jails, not in worksite operations.
Take a look at the graph above one more time. It shows that workplace enforcement came to a near standstill under Obama. The bottom lines show the number of workers and employers arrested. The number of worksite-enforcement actions, which was never very high relative to enforcement focused on criminals, fell significantly during the Obama administration, and represented less than one percent of all ICE arrests during this time. Under Obama’s policies, ICE agents conducted payroll audits and levied fines on employers caught hiring illegal aliens, but were told not to arrest illegal workers. The illegal workers who lost their jobs after ICE audits may have been dismissed, but remained free to take new jobs elsewhere. It’s no wonder that Porter, relying on Peri’s research, could find no impact of Obama-era enforcement on American workers.
If there really was a big increase in interior immigration enforcement, then we would expect to see a corresponding drop in the overall size of the illegal-immigrant population, unless there was a big increase in new arrivals to offset it. But no researcher thinks there was an increase in new arrivals during this time. Pew Research and others have shown that between 2009 and 2015 the overall size of the illegal-immigrant population seems to have held steady, after dropping in 2007 and 2008 during the time of the actual surge at the end of the Bush administration and the downturn in the economy. Thus, the labor-market impact of illegal workers almost certainly held steady as well. This fact by itself would seem to invalidate the entire argument of Porter’s piece.
At the most basic level, the problems with Porter’s article stem from the fact that he talked to no one with a point of view different from his own. To be sure, immigration data is hard to interpret and is often incomplete. But this makes it all the more necessary for reporters to talk to a wide variety of people, which Porter clearly did not.
The New York Times public editor wrote some years back that if you think it does not have a bias, “you’ve been reading the paper with your eyes closed.” Our colleague, Pulitzer Prize-winning reporter Jerry Kammer, has written about this bias as it relates to immigration coverage more than once. The newspaper of record has helped create the misleading narrative that there was a large increase in interior enforcement under Obama, which many people still believe. They should open their eyes.