"Over the last several years, 40 percent (of) the jobs created in Texas were created for illegal aliens."

Mitt Romney on Tuesday, October 18th, 2011 in a Republican presidential debate in Las Vegas

Mitt Romney hits Rick Perry, saying 40 percent of Texas job growth went to illegal immigrants

Former Massachusetts Gov. Mitt Romney and Texas Gov. Rick Perry gesticulate during the Oct. 18, 2011, Republican presidential debate in Las Vegas.

During the Oct. 18, 2011, Republican presidential debate in Las Vegas, Mitt Romney sought to undermine Texas Gov. Rick Perry’s argument that he’s presided over rapid job creation during his tenure.

"If you look over the last several years, 40 percent, almost half the jobs created in Texas, were created for illegal aliens, illegal immigrants," Romney said.

Romney was referring to a study titled, "Who Benefited from Job Growth In Texas? A Look at Employment Gains for Immigrants and the Native-Born, 2007 to 2011," published by the Center for Immigration Studies, a research group that supports reducing illegal immigration.

The report analyzed U.S. Census Bureau data and found that immigrants, both legal and illegal, "have been the primary beneficiaries" of job growth in Texas since 2007, even though native-born workers accounted for 69 percent of the growth in the state’s working-age population over the same period.

Specifically, the report said that, of the jobs created in Texas since 2007, 81 percent were taken by newly arrived immigrant workers, both legal and illegal. In the meantime, the group estimated that about half of the newly arrived immigrants who took jobs in Texas beginning in 2007 were illegal.

As a result, the report estimated that about 40 percent of job growth in Texas since 2007 went to newly arrived illegal immigrants.

So Romney’s comment described the results of the study pretty faithfully, though it would have been more accurate if he’d referred to jobs "given to illegal aliens," rather than "created for illegal aliens." The study doesn’t suggest that Texas employers were conspiring to create jobs specifically for illegal immigrants, only that illegal immigrants ended up filling them.

However, we see two additional issues that call into question the accuracy of the study itself:

The 40 percent figure uses gross numbers of immigrants rather than net numbers.

The report actually offered calculations using two sets of numbers -- one using the total number of immigrants coming into the state, and the other using that number minus the number of immigrant departures and deaths. Using the net immigrant figure -- the one Romney didn’t use -- results in a lower percentage.

The figure using net immigration is about 54 percent. And once you divide that by two to estimate the share for illegal immigrants, the relevant number is about 27 percent, not 40 percent.

So Romney has effectively cherry-picked the higher of two percentages offered in the report.

Both the 40 percent figure and the 27 percent figure are subject to a wide margin of error.

Jeffrey S. Passel, senior demographer for the Pew Hispanic Center, an independent research organization, said that "there are lots of methodological problems with the CIS study, mainly having to do with the limitations of small sample sizes and the fact that the estimates are determined by taking differences of differences based on small sample sizes." In addition, the assumption that 50 percent of immigrants are illegal is an educated guess, but still a guess.

Meanwhile, the Census data used in the report is a longitudinal survey, which means that Census employees check back periodically with the same participants. "How many undocumented folks do you honestly think would be participating in that?" asks Sarah Jane Glynn, a policy analyst with the liberal Center for American Progress.

Gary Burtless, an economist with the Brookings Institution, added that the report did not drill deeper into the data than "working age."

"Even within the working-age population, it might be the case that the immigrant population may have disproportionately grown in age groups -- say, ages 25-49 -- where employment rates are highest," he said. "A really sophisticated analysis might show that, given the age distribution of net immigrant population growth and net native population growth, immigrants got exactly the proportion of new jobs a good forecaster would predict." Given the data, though, we "just don’t know," he said.

The report does acknowledge that "no estimate of illegal immigration is exact." But the methodological shortcomings also weaken the certainty of Romney’s statistic.

On balance, we think that both the report’s authors and its critics have reasonable points.

In the big picture, we agree with Chuck DeVore -- a conservative critic of the study -- that "trying to draw conclusions about immigration and employment in Texas in isolation from other factors is problematic at best."

But we also agree with Mark Krikorian, the Center for Immigration Studies’ executive director, that "even if DeVore prefers a net-to-net comparison, immigrants still got a disproportionate share of new jobs."

Our ruling

In the debate, Romney cited the report’s headline figure pretty accurately. But he failed to note, first, that it was an estimate subject to significant uncertainty, and second, that the report itself included an alternative -- and arguably more accurate -- measurement that produced a significantly lower figure.

That said, the lower figure, even when adjusted for uncertainty, still backs up Romney’s underlying point that illegal immigrants have gained a disproportionate share of new jobs in Texas. It’s just that the number probably isn’t as high as Romney said in the debate. So we rate the statement Half True.