Immigration And Welfare Revisited

On May 24, the Wall Street Journal published an editorial on "Immigration and Welfare" that sparked my curiosity. Having published several papers on this very question (here and here and here), I felt something was missing. Let me quote the passage that bothered me:

The "welfare" charge is also refuted by the experience of the federal welfare reform passed 11 years ago. That law reduced the welfare eligibility of new immigrants on the sensible grounds that the magnet for America should be work, not a government handout. Ron Haskins, an architect of that reform and the author of a 2006 book on its consequences, concludes that "the use of welfare by noncitizens has declined rapidly" in the wake of that law.

Ron is clearly right--the use of welfare by noncitizens did decline rapidly after 1996. But look at the graphic attached to this post, summarizing the data from the Current Population Surveys. It shows what happens to the fraction of households that receives some type of assistance (either cash, food stamps, or Medicaid).

As they say, a picture is worth a thousand words. The steep decline in welfare use by noncitizens after the 1996 welfare reform was more or less undone by an equally steep rise in welfare use after 2000. Roughly speaking, we are now back where we were before Bill Clinton signed that legislation. Welfare reform did not work as expected--partly because many states with large immigrant populations, like California, replaced the federal cutbacks with state funds. (The WSJ editorial also makes this point when it notes that there was some increase in immigrant use of Medicaid in recent years, but fails to point out what that implies for the overall trend).

So does this mean that many immigrants come to the United States seeking welfare? I don't think so.

We have an elaborate welfare state designed to protect the disadvantaged. People get sick; they fall into hard times. Immigrants are not exempt from these life events. In the past couple of decades, immigration has been disproportionately composed of low-skill workers, so immigrants will tend to qualify for the safety net more often than the native-born. There is nothing sinister about their behavior; it is precisely what the welfare state is supposed to do.

But it seems to me that all the facts--not just some of them--should be on the table as the future of immigration policy is being debated.

Comments

[jimbino]

You fail to mention that most of the welfare goes to the breeders! They may be disadvantaged, but they have chosen to hobble themselves at our great expense.

[PatrickR]

Looks like noncitizens sharply increased their welfare use in the wake of the recession, and now it's dropping more sharply citizens'.

[Daveg]

Hello,

Can you give a little more detail as to what is included in the non-citizens graph and the source? Thanks.

[gborjas]

Daveg.

I calculated it myself using the IPUMS version of the March Current Population Surveys
http://www.ipums.umn.edu/

I downloaded the data, defined a household as nativee or immigrant by the status of the household head, and defined noncitizen households to be all households headed by a noncitizen. The data tell you whether someone in the household received cash benefits, food stamps, or medicaid. And I just calculated the percent of households that received such assistance. Hope this helps. I have a similar graph for earlier periods in my book Heaven's Door.