The National Academies of Sciences, Engineering, and Medicine have just released what can fairly be described as the most comprehensive look at the economic and fiscal impact of immigration on the United States. It represents an update of sorts of a similar NAS study released in 1997, in the middle of an earlier immigration debate. Overall the report is quite balanced, with a lot of interesting findings.
Unfortunately, it is long and dense, and many of the most important points are buried in its nearly 500 pages. (Disclosure: I was an outside reviewer of the study, and my research is cited in it.)
The most straightforward part of the study is its assemblage of estimates of the current fiscal impact of immigrants. The study shows that immigrants (legal and illegal) do not come close to paying enough in taxes to cover their consumption of public services at the present time. The NAS present eight different scenarios based on different assumptions about the current fiscal impact of immigrants and their dependent children — and every scenario is negative. No matter what assumption the NAS makes, immigrants use more in public services than they pay in taxes. The largest net drain they report is $299 billion a year.
It should be pointed out that native-born American are also shown to be a net fiscal drain, mainly because of the federal budget deficit — Washington gives out a lot more than it takes in. But the fiscal drain created by immigrants is disproportionately large relative to the size of their population. Equally important, a fiscal drain caused by natives may be unavoidable. Adding more immigrants who create a fiscal drain, on the other hand, can be avoided with a different immigration policy.
Although the NAS do not say so specifically, prior research is pretty clear: The fiscal drain is caused by two groups: Legal immigrants who have no education beyond high school and, to a lesser extent, illegal immigrants, most of whom have no education beyond high school. These less-educated immigrants have modest incomes and make modest tax payments, while they (or more often their U.S.-born children) make heavy use of welfare programs. Skilled immigrants have high incomes and are not a fiscal drag.
In addition to estimating the current fiscal impact of immigrants, the NAS also projects the fiscal impact of new immigrants in the future under eight different scenarios, using different assumptions. Some of the scenarios show a negative long-term fiscal impact, some a positive one. But the projections are highly dependent on assumptions about the future education level of immigrants, government-spending levels, and tax rates. Plus, the projections go out 75 years, so they are little more than speculation.
With regard to economics — jobs and wages — the results in the NAS study, based on the standard economic model, show that immigration does make the U.S economy larger by adding workers and population. But a larger economy is not necessarily a benefit to natives. The report estimates that the actual benefit to the native-born could be $54.2 billion a year — referred to as the “immigrant surplus.” This is the benefit that accrues to American businesses because immigration increases the supply of workers and reduces American wages. Several points need to be made about this estimate.
First, to generate this surplus, immigration has to create a very large redistribution of income from workers to owners of capital. The model works this way: Immigration reduces the wages of natives in competition with immigrant workers by $493.9 billion annually, but it increases the income of businesses by $548.1 billion, for a net gain of $54.2 billion. Unfortunately, the NAS does not report this large income redistribution, though it provides all the information necessary to calculate it.
A second key point about this economic gain is that, relative to the income of natives, the benefit is very small, representing a “0.31 percent overall increase in income” for native-born Americans.
If new immigrants hurt the wages and English acquisition of immigrants already here, then a reduction in immigration may help the millions of immigrants currently in the country.
Third, the report also summarizes empirical studies that have tried to measure directly the impact of immigration on the wages of natives (the analysis above being based on economic theory rather than direct measurement). The size of the wage impact in those empirical studies is similar to that shown above. The NAS report cites over a dozen studies indicating that immigration does reduce wages primarily for the least-educated and poorest Americans. It must be pointed out, however, that there remains some debate among economists about immigration’s wage impact.
The fourth and perhaps most important point about the “immigrant surplus” is that it is eaten up by the drain on the public fisc. For example, the average of all eight fiscal scenarios is a net drain (taxes minus services) of $83 billion a year at the present time, a good deal larger than the $54.2 billion immigrant surplus.
The report shows somewhat higher self-employment for immigrants (about 10 percent) than for natives (about 8 percent). But if you slightly change the age range when calculating self-employment, you get different results. The NAS seems to be looking at adults 18 and older. But relatively few entrepreneurs are under age 25 — and relatively few immigrants, compared with natives. Most immigrants arrive in the U.S. after age 25. Including 18- to 24-year-olds reduces the self-employment rates of natives relative to immigrants.
RELATED: The Only Immigration Solution
Using the same data as in the NAS study, we find that the self-employment rate of persons 25 and older is basically the same for immigrants and natives — 11.4 percent versus 11.1 percent, respectively Adding in those who are self-employed part-time creates a slightly higher rate for natives. There is some evidence that immigrants are somewhat more likely to start businesses, but their businesses seem to fail at higher rates. In short, there is little meaningful difference in the share of immigrants and natives who operate their own businesses.
On the issue of innovation, it is possible that immigrants are more likely to file patents, the most common measure used to study inventiveness. The reason for this seems to be that they are somewhat more likely to have science, technology, engineering, or math (STEM) degrees. Comparisons of native and immigrant holders of STEM degrees do not show much difference. The patent data itself are hard to interpret; all patents are not the same. Many are useless, and others are filed by patent trolls. Also, some research cited by the NAS shows that immigration reduces the research output of American natives, making the issue even more complex.
Typically, immigrants who file patents are skilled, those with at least a college degree. At present only one-third of immigrants have even a bachelor’s degree, and maybe half of those are in a STEM. If we think that immigrants really are more likely to file patents, and that this is actually a measure of innovation, then we would have to move to a very different immigration system, selecting immigrants primarily for their skills and not, as we do now, on the basis of their family relationships.
Finally, it is worth noting that the NAS study concludes that immigrants do not close the economic gap between themselves and natives as quickly as they used to. The lower education level of newly arrived immigrants partly explains this difference, but even after controlling for education, the study finds a slowdown. The report also finds a slowdown in acquisition of English. It cites as a possible explanation the huge number of immigrants already here, who give new immigrants less incentive to learn English. If new immigrants hurt the wages and English acquisition of immigrants already here, then a reduction in immigration may help the millions of immigrants currently in the country.The slowdown in assimilation has one other important implication. The report shows that second-generation Americans (U.S.-born adults with immigrant parents) typically have closed the economic gap between themselves and natives. But today’s second-generation adults are not the children of today’s immigrants. Their parents arrived decades ago, in most cases before the assimilation slowdown. It is not at all clear that the children of today’s immigrants will close the economic gap the way the children of immigrants did in the past.
The new NAS study has an enormous amount of useful information. In my view, the best way to think about immigration is that it is primarily a redistributive policy, transferring income from some workers to owners of capital and from taxpayers to low-income immigrant families.
— Steven Camarota is director of research at the Center for Immigration Studies.