Daniel Costa of the Economic Policy Institute (EPI) criticized a piece I wrote for The Hill in which I called for the U.S. to accept more refugees. Costa took issue for my argument to limit their access to welfare once they arrive, which I wrote in the eighteenth paragraph of my piece. Conservatives criticized me for not mentioning welfare reform sooner in my piece. I wrote about allowing more refugees in for the first seventeen paragraphs of my piece because that is more important than denying them welfare.

Costa, however, stooped pretty low when he wrote: “[H]opefully refugees in America will never be forced to suffer their libertarian version of humanitarian relief.” Emphasis added.

The humanitarian relief that refugees need isn’t food stamps once they arrive to the United States – it’s an escape from violence and oppression. Refugees aren’t fleeing Syria because their Syrian equivalent of TANF benefits expired, they are fleeing because they are being murdered. 

Costa assumes my opposition to welfare means that I oppose all support for refugees. That is untrue. As I mention in my original piece, civil society, private charities, churches, previous immigrants, and other groups that do aid refugees are performing a valuable service. That aid is important in helping some, but not all, people who flee war, oppression, and dictatorship to thrive in their new country. That voluntary aid and support should continue and the generous people who donate their own money to such causes are to be commended. But welfare is not charity and it does not alleviate the real scarcity that affects these refugees: a lack of visas for them to come here in the first place. 

Welfare or Refugees: Choose One

Welfare is one reason why America admits so few refugees. It’s an open secret on Capitol Hill that President Obama didn’t raise the refugee cap as a part of his executive actions specifically because of the welfare cost. Limiting welfare for refugees is a key element of convincing the American public and policymakers to accept more immigrants of every kind – including refugees. It’s no coincidence that support for immigration increased after the 1996 welfare reform. Costa’s insistence that refugees receive welfare benefits once they arrive is tantamount to him insisting that fewer of them come in the first place – an outcome with far worse humanitarian consequences than the current situation. 

The public choice reality is that more handouts to non-citizens will lead to less public support for liberalizing humanitarian immigration. Here I document many of the polls that reveal how worried Americans are about immigrant welfare use. Those welfare fears are vastly overblown but they do affect public opinion and policy making. 

Since the political limitation on allowing more refugees is welfare, the only humane thing to do is to deny welfare benefits and thus allow more of them to come. Can anybody seriously suggest that Syrian refugees would be worse off than they currently are if more of them arrived in the United States but didn’t have access to means-tested welfare? Of course not.

Let’s allow the refugees to choose for themselves. We should allow more refugees to come beyond what the current quota allows but deny them access to means-tested welfare until they naturalize. Hundreds of thousands or millions of refugees would take advantage of that deal in the near future and they would all be better off than they currently are. 

Welfare Doesn’t Aid Assimilation

Costa argues that welfare benefits help immigrants succeed in the United States. Refugees don’t need welfare to succeed once they are here. Refugees outperform many classes of economic immigrants who are ineligible for means-tested welfare. Costa believes this is because those refugees received some welfare (I’m inferring this from his piece). 

However, the peer-reviewed version of the paper he cites by Kalena Cortes doesn’t even mention welfare as a contributing factor to their labor market success. She attributes the growth of income to refugee willingness to work more hours and their higher rate of human capital accumulation due to their longer time horizons – they can’t return to their home countries. It’s a testament to the perseverance of refugees that their willingness to work was not constrained much by welfare benefits. Refugees are responsible for their own economic successes, not the welfare agencies. 

There is some evidence that access to welfare can slow down English acquisition – an important part of labor market integration. Soviet refugees in New York State were less likely to work than their counter-parts in Maryland. Those in New York also had more access to welfare than those in Maryland. Those in Maryland were also more comfortable with the English language. Welfare decreases the incentive to work, on the margin, due to high effective marginal tax rates. Working was an important part of increasing English skills. To the extent that welfare decreased the likeliness of refugees to work, it slowed their accumulation of language skills and success in the labor markets. 

Some Vietnamese refugees did have access to means-tested welfare and they assimilated fine. But as David Haines points out in his book Safe Haven: A History of Refugees in America, refugees entered the labor force more rapidly in cities like Richmond, Virginia where public assistance was more limited than in places where it was more abundant. 

More important than public assistance is that many Vietnamese refugees came when the U.S. economy was growing or was on the verge of a major job expansion. The papers I cite above all point to economic and job growth as a better predictor of refugee assimilation and success than anything else. For instance, Haines points to the strong manufacturing job base in Richmond as helping refugees to that city do well. 

If welfare matters so much for refugee integration, then why was there no economic effect on refugees after it was decreased? As Bollinger and Hagstrom write:

“Second, the 1996 welfare reforms appear to have no effect on the probability of poverty for any group, regardless of the measure used. Nor is there much evidence for any substantial difference in the post reform between immigrants, refugees, and native born. Indeed, even when differences are measured in how well the programs moved these groups out of poverty, there appears to be no substantial differential impact of welfare reform.” 

The booming economy post-1996 certainly helped refugees enter the job market. Perhaps welfare reform aided that period of job growth, perhaps it didn’t, but it doesn’t seem to have mattered much either way. From 1994 to 1999 welfare use by refugees collapsed so that by 1999 refugee-headed households had similar welfare use rates as U.S. citizens. Refugees are still clamoring to come here. Sweden gives high welfare benefits to refugees that combine with their rigid labor market regulations to lock many into segregated poverty for decades. Fortunately the United States government does not provide nearly so much welfare nor does it have such a rigid labor market. Attempting to be humanitarian by providing welfare backfired in Sweden. 

At best, Costa can claim that welfare does not delay integration into the labor market at the current low levels in the United States, but welfare doesn’t seem to speed up integration either. Refugee time-horizons, inability to return to their home countries, their resources upon arrival, and acquisition of human capital aided by those realities seem to explain their relative success – not government welfare offices dispersing taxpayer dollars. Assuming for argument’s sake that welfare doesn’t delay labor market integration (it does), removing it does no harm to the refugees and could influence public opinion to allow more of them to enter the United States while saving taxpayers some money. Sounds like a win for everybody.