Report Credits Drop in Illegal Immigrants to Enforcement
Study Was Based on Census Data That Indicate Number of Less-Educated Hispanics Has Declined

By N.C. Aizenman
Washington Post Staff Writer
Thursday, July 31, 2008

A report released yesterday by a Washington think tank that advocates stricter limits on immigration says the number of illegal immigrants in the country appears to have declined significantly over the past year, at least partly because of the chilling effect of stepped-up enforcement.

The study by the Center for Immigration Studies based its findings on census data that indicate that the number of less-educated, working-age Hispanic immigrants, defined as 18-to-40-year-olds with a high school diploma or less, has dropped by more than 10 percent, or about 830,000 people, since last August.

Previous research suggests that a large share of less-educated foreigners is in the country illegally and that it makes up the bulk of the illegal immigrant population. Furthermore, although earlier declines in the number of these Hispanic immigrants have been linked to a rise in their unemployment rate, the current drop-off began last year almost immediately after Congress abandoned legislation to legalize undocumented immigrants and six months before any significant rise in their unemployment rate had occurred.

During the same period, the number of foreigners who were more educated or non-Hispanic, and therefore far less likely to be illegal immigrants, continued to rise or hold steady.

"The evidence is consistent with the idea that at least initially, more robust enforcement caused the number of illegal immigrants to decline significantly," said Steven A. Camarota, one of the study's authors. "Some people seem to think illegals are so permanently anchored in the United States that there is no possibility of them leaving. . . . This suggests they're not correct. Some significant share might respond to changing incentives and leave."

Several demographers who specialize in estimating the illegal immigrant population expressed concern about the limits of the study's methodology but said they found the possibility that the illegal immigrant population is decreasing plausible. Determining the actual amount of that decline, however, is far more controversial.

The census does not ask about immigration status. Instead, government and independent researchers use a variety of techniques to estimate the number of immigrants in the country illegally. One way is to subtract the number of visas, permanent residency permits and naturalizations granted each year from the total number of foreigners counted by the census. The difference between the number of foreigners who can be accounted for through such records and the total number tallied by the census is considered to be the size of the illegal immigrant population.

Camarota and co-author Karen Jensenius took a different approach, calculating the previous ratio between the number of less-educated Hispanic immigrants counted by the census and the total illegal immigrant population estimated by government researchers, and then applying that ratio to the new, lower number of less-educated, working-age Hispanic immigrants to come up with a new estimate for the total illegal immigrant population. According to their calculations, from August of last year to May, the illegal immigrant population declined by about 11 percent, to about 11.17 million from a high of 12.49 million.

One drawback of Camarota's and Jensenius's method, noted the Pew Hispanic Center's Jeffrey S. Passel, a widely regarded expert on estimating the illegal immigrant population, is that "it tracks something that correlates with the number of illegal immigrants rather than the actual number of illegal immigrants, and it assumes the correlation remains the same."

"If the ratio [between the number of less-educated Hispanic adults and the total number of illegal immigrants] has changed, then the trend could be very different," Passel said.

Even more contentious is the question of what, if anything, the study's findings indicate about the impact that recent national and local immigration policies might have had on the size of the illegal immigrant population. Since December, the unemployment rate of less-educated, working-age Hispanics has risen to 7.06 percent from 4.93 percent, making it that much more difficult to determine whether the continued decline in their population during this period was the result of anything beyond basic economics.

But Camarota and Jensenius suggest that the six-month decline that occurred after the failure of the legalization legislation and before the rise of these workers' unemployment rate is one of several examples of a link between immigration policy and immigrant choices. They note, for instance, that starting in May of last year, when Congress's consideration of the legalization plan began receiving widespread media attention, the number of less-educated, working-age Hispanics began to rise.

"I call it the amnesty hump," Camarota said. He noted that the population increase during this period might not have been statistically significant, but "it seems that what was happening was that fewer illegal immigrants left than might otherwise have done so because they were hoping to qualify for legalization."

Also up for interpretation is the degree to which the drop in the number of less-educated Hispanic adults (and, by inference, illegal immigrants) was the result of fewer foreigners entering the country or more of them leaving. The U.S. Border Patrol reported a 20 percent decline in apprehensions along the southern border over fiscal 2007, a possible indication that fewer illegal immigrants attempted to enter the country.

Camarota and Jensenius note that census data do not answer the question. But the authors suggest that if less-educated Hispanic adults were not leaving in greater numbers than before, their total population would merely grow more slowly, not decline steeply.

Among those who are leaving, the vast majority are probably doing so on their own. Despite a surge in work site raids and other enforcement measures, as well as decisions by various state and local governments to train their police to identify illegal immigrants, only 285,000 immigrants were removed from the United States last year, and many of those were formerly legal immigrants who lost their status after committing a crime.

Camarota and Jensenius said they take this as possible evidence that tougher enforcement can have a multiplier effect, scaring many more illegal immigrants into leaving of their own accord than authorities can pick up. And the authors suggest that if the trends they identify are sustained, "it would cut the illegal population in half within just five years."

However, Randolph Capps, a researcher with the Urban Institute who has studied the number of U.S. children born to illegal immigrants, cautioned against such reasoning.

Even if all the findings in the study by Camarota and Jensenius prove correct, he said, it is probable that the first million illegal immigrants to leave were those who had arrived more recently and had the weakest ties to the United States.

The remainder, including the more than half of illegal immigrant adults who have children in the United States, Capps said, are less likely to leave unless they are removed by the government.

"Having a kid in school provides a really strong incentive to stay," he said. In addition, "People who are more settled in the United States have more options. They can move to another [state or county] where enforcement is not as strict. If they lose a job, they can find another. If one member of the family is arrested and deported, they can find other relatives to stay with."

View all comments that have been posted about this article.

© 2008 The Washington Post Company