Friday, September 22, 2017 1:00 pm
Fact check: Do 10 refugees overseas cost as much as one in US?
GLENN KESSLER | Washington Post
"For the cost of resettling one refugee in the United States, we can assist more than 10 in their home region." -- President Donald Trump, address to the U.N. General Assembly, Sept. 19, 2017
In arguing for keeping refugees close to their home countries, rather than resettling them in other countries, the president offered an arresting statistic. He gave a very specific figure -- that 10 refugees overseas could be helped for the cost of settling one in the United States.
How did he calculate this, and is the calculation valid?
The White House says the president's estimate came from a 2015 report by the Center for Immigration Studies, a research organization in favor of lower immigration levels.
The report, titled "The High Cost of Resettling Middle Eastern Refugees," concluded that in the first five years in the United States, each refugee costs taxpayers $64,370, or 12 times what the United Nations estimated it needed to care for refugees during a five-year period. On an annual basis, refugees cost $12,874 a year, the report estimated.
The report based its estimate on a variety of data sources, calculating how refugees used such services as Medicaid, food stamps, Temporary Assistance for Needy Families and the like. About one-quarter of the figure reflected the cost of public education for the children in refugee families.
The report acknowledges that refugees pay "something in taxes" but said the data was relatively sparse for the first five years in the country. So it did not look at the revenue side of the ledger.
Though the research is extensive, we often warn readers to be wary of reports by groups with a particular agenda. It's striking that even at this stage of his presidency, Trump would rely on a report from an outside group rather than an internal analysis produced by government experts.
As it turned out, the Department of Health and Human Services in July had drafted a 52-page report that attempted to assess the cost of refugees, in response to a presidential order in March for the State Department and other agencies to tally the cost of helping refugees in the United States.
The report is comprehensive, with a review of the literature on the subject that included an examination of the CIS report. The HHS report is open about the shortcomings in the data and uncertainties in the estimates.
But as the New York Times first reported, the White House killed the draft HHS report, which attempted to tally costs and benefits.
The Times later published the squelched report on its website. HHS looked at a 10-year cost for all refugees, not just from the Mideast, and came up with a figure of nearly $7,134 a year. That's just more than half the estimate by CIS.
But then HHS took it one step further and estimated the taxes paid by refugees, including payroll taxes, income taxes, property taxes, sales taxes and the like. The net fiscal effect was $63 billion during a 10-year period, the agency said.
In other words, over time, the taxes paid by refugees easily exceeded the cost to taxpayers.
That was too much for the White House, which rejected the analysis. Instead, the agency was ordered to send just a three-page report to the State Department -- which is still working on the refugee-cost memo for Trump -- that looks only at costs.
It concluded the annual per capita expenditures for refugees was $3,346, compared with $2,501 for the general U.S. population. (One could argue that a more relevant comparison would between refugees and low-income Americans on assistance, not all Americans.)
Why are the numbers so different for the CIS study and the HHS study? They looked at different things.
CIS examined statistics on Mideastern refugees, who tend to be poorer and less educated, whereas the HHS report looked at all refugees. CIS focused on the first five years, when resettlement costs are higher.
A working paper published by the National Bureau of Economic Research in June indicated that the break-even point between costs and benefits came about eight years after a refugee arrived in the United States. So, after that point, the average refugee would be a net benefit to U.S. taxpayers.
Steven Camarota, CIS's director of research, faulted the HHS report for crediting taxes contributed by refugees but not the cost of public goods, such as police protection, paid with those taxes.
Beyond the varying estimates, there's a fundamental dispute about whether Trump's comparison is even valid. Refugees in the country of first asylum do not have many prospects; they have no jobs and their children probably have access to little schooling.
In essence, it is intended as temporary situation, not a permanent solution. Refugee experts say the point of resettling them in a country such as the United States is to allow them to move on from the conflict that drove them from their homes.
Julia Gelatt, a senior policy analyst at the Migration Policy Institute, also noted there are diplomatic reasons for taking in refugees, such as fostering the U.S. relationship with countries providing temporary shelter.
"It is important for showing we are taking some share in the responsibility for caring for refugees," she said.
Camorota said it was logical to ask whether it made sense to help some people at the expense of helping even more still overseas.
"If someone comes here, it is like winning the lottery," he said. "It's a reasonable question: Do you help a tiny group of people in a huge way or a lot of people in a small way?"
To a large extent, Trump's statement reflects a philosophical issue beyond the realm of fact-checking.
A case can certainly be made that the president is comparing apples and oranges -- and that he is relying on an estimate for costs higher than what was produced by his own government (and then squelched because it did not align with administration policy). One can also argue it makes more sense to balance the costs of caring for refugees with the contributions, monetary or otherwise, they make to society over time.
But like we said, that's a philosophical question.