A Critique of Douglas Holtz-Eakin’s Analysis
Download a pdf of this Backgrounder
Steven A. Camarota is the Director of Research at the Center for Immigration Studies
In an opinion piece1 for the American Action Forum, which he heads, former John McCain (R-Ariz.) economic advisor and CBO director Douglas Holtz-Eakin tries to make the case that "immigration reform" would be a good deal for public coffers. Holtz does not outline the reform he has in mind nor does he ever address the fiscal impact of legalizing illegal immigrants. The article is so lacking in specifics that it seems as if it was written so that advocates for legalization and increased immigration could simply say the words "dynamic analysis" or "dynamic scoring" and this would by itself somehow overcome the traits of immigrants that are likely to make immigrants generally, and less-educated immigrants in particular, a significant fiscal drain.
The central point of Holtz-Eakin's "dynamic analysis" is to argue that immigration-induced population growth by itself will have a positive indirect impact on the economy and public coffers. To reach his conclusion, Holtz-Eakin ignores the economic literature showing that immigration only slightly increases the income of natives. He also ignores the development literature that indicates that population growth does not increase per capita GDP growth. He also assumes away any congestion and resulting inefficiencies that seem likely with population growth. But perhaps the biggest weakness of his analysis is that he ignores the actual characteristics of immigrants, including illegal immigrants, such as their relatively high poverty, welfare use, lack of health insurance, and their more modest tax payments.
The core of Holtz-Eakin's argument is that immigration makes the economy larger in the future by increasing the population. This he believes will change behavior, thereby increasing productivity, income, and resulting tax payments. To be clear, the author does not cite any research showing immigration itself increases per capita GDP. Nor does he ever discuss the fiscal impact of legalizing 11 to 12 million illegal immigrants. The article is really just speculation about the benefits of increasing population growth. Yet a number of advocates of high immigration have latched on Holtz-Eakin's claim that dynamic analysis means that amnesty would be a boon for the treasury.
One of the most glaring omissions of the article is that it never mentions the fiscal impact of legalizing 11 to 12 million illegal immigrants. He never discusses any research on the actual taxes paid and public benefits used by immigrants generally or illegal immigrants in particular. Moreover, he never clearly explains the reform he has in mind, though he seems to want to increase the number of people allowed into the country and move to a more skills-based system. He buttresses his argument by making some generalizations about immigrants that are at best overstatements and in some cases demonstrably incorrect.
Holtz-Eakin starts off by arguing that immigration can impact "the overall pace of population growth". This is certainly correct. But then he goes on to argue that "in the absence of immigration, low birth rates mean that the U.S. population will actually decline." But this is misleading because even if net immigration is zero, something that is extremely unlikely, the U.S. population is still projected to grow until about the year 2040 — adding 30 million to the population. After 2050 it will start to decline slowly. So yes the population would decline without any immigration, but not for nearly half a century.2
Ignoring the Immigration Literature
But the larger issue is immigration's impact on per-capita income. Economists have studied this question, but this research is entirely left out of Holtz-Eakin's analysis. As the nation's leading immigration economist, George Borjas of Harvard, points out in a recent paper, "immigration makes the U.S. economy (GDP) significantly larger, with almost all of this increase in GDP accruing to the immigrants themselves as a payment for their labor services."
Borjas concludes that "Although immigration makes the aggregate economy larger, the actual net benefit accruing to natives is small, equal to an estimated two-tenths of 1 percent of GDP."3 A larger economy from immigration is not a richer economy, though it is not a poorer one either. It may also be worth noting that to generate these tiny gains immigration has to redistribute income. In the United States, the workers who lose from immigration tend to be the least-educated and poorest workers, who very likely have to use more government services as their income declines.
In their magisterial 1997 study "The New Americans", The National Academy of Sciences came to the same conclusion as Borjas — immigration's main impact is to redistribute income. The Academy estimated that the economic benefit from the redistribution created by immigration was one- to two-tenths of 1 percent GDP for natives, which is very similar to Borjas' estimate. This is very far from the kinds of per capita gains Holtz-Eakin asserts in his article.
Holtz-Eakin never refutes or even addresses the research showing that immigration makes America bigger but not richer, even though it is directly germane to his analysis. His real focus is on arguing that we should allow in more immigrants — again, without addressing the current population of illegal immigrants. Holtz-Eakin estimates that the immigration reform he has in mind, which he does not explain in any detail, can raise per capita GDP by $1,700 a year.
To make his case he argues that because immigration makes the population larger, GDP growth will be higher and "there will be greater replacement of existing capital goods and investment in new capital goods." And "more rapid overall population growth would generate more rapid GDP growth, which would in turn raise productivity." The research he cites to support his argument does not deal with immigration; it is a theoretic work suggesting a relationship between a larger population and positive economic outcomes.
It is not at all clear the few studies he cites are even relevant to immigration-induced population growth. One of the studies he links to deals with manufacturing from 1949 to 1983.4 Even if the study was correct, a lot has changed in the last 30 years. The idea that population growth all by itself increases per capita income and productivity is certainly not a consensus position among economists.
Ignoring the Development Literature
In addition to not discussing the immigration literature, Holtz-Eakin ignores the literature that looks at the impact of population growth on per capita income in developing countries, which would appear to be directly relative to his argument. That research generally does not support the idea that by itself population growth increases per capita GDP. In fact, there is good evidence that rapid population growth works against improving incomes in such countries. A 2009 review of 29 different studies on the impact of population growth on economic development concludes: "Particularly strong is the evidence in support of the increasingly adverse effects of population growth in the post-1980 period."5
Maybe he feels that this work is not relevant to developed countries like the United States. But he does not say so.
Holtz-Eakin also assumes away issues of congestion and the potential inefficiencies it creates. Yet traffic alone has been found to cost the economy $121 billion annual in wasted time and fuel.6 Perhaps population density plays no role in congestion, so we can make our country much more densely settled without increasing traffic or congestion. This would seem to be an important question if your whole argument hangs on the idea that a larger population is a richer population. But he never deals with it.
The Actual Characteristics of Immigrants
One of the oddest features of the entire article is that it ignores the actual characteristics of immigrants that relate most directly to their fiscal impact. For example, of households headed by immigrants, 36 percent access one or more welfare programs, compared to 23 percent of native-headed households. And the federal income tax liability of immigrant households is only 71 percent that of native households.7 What are the fiscal implications of this? None of this seems to matter to Holtz-Eakin.
To the extent that he discusses the traits of immigrants, he often invokes positive stereotypes that are at best overstatements. He asserts that because "immigrants have higher rates of labor force participation" increased immigration would cause the labor force to grow more rapidly. Actually there is no meaningful difference in the labor force participation rate of immigrants and natives.
Holtz-Eakin cites a Census Bureau table from 2010 that shows somewhat higher employment rates for immigrants. But the table includes people over 65 and younger teenagers. These two groups have relatively low rates of employment, and there are relatively fewer immigrants in these groups. (There are relatively few immigrant teenagers because immigrants generally arrive after age 20, and the children they have once here are, by definition, not immigrants.) The same table shows that if you just focus on 25 to 64-year-olds, the core of the labor force, the labor force participation rate for immigrants is 77.9 percent and 78.4 percent for natives.8 Public-use data from the fourth quarter of 2012 show the labor force participation rate for 18 to 65-year-old natives was 74.9 percent vs. 74.7 percent for immigrants.9 Immigrant men have somewhat higher rates of labor force participation, as Holtz-Eakin argues, but immigrant women have lower rates, so there is no significant difference overall.
Immigration No Fix for an Aging Society
Holtz-Eakin also argues that immigration will help fix the problem of an aging society. In fact, immigration has a tiny impact on slowing the aging of the U.S. population. To make his case Holtz-Eakin cites an article in the Wall Street Journal that states, "if you strip these immigrants — and their relatively high fertility rates — from our population profile, America suddenly looks an awful lot like continental Europe, which has a fertility rate of 1.5." This is simply wrong.
The 2011 American Community Survey collected by the Census Bureau shows that the Total Fertility Rate or TFR in the United States overall (immigrant and native) is 1.98 children per women, and for natives alone it is 1.88. This is a difference of .1 children or about 5 percent. Immigrants do have higher fertility than natives. But their impact on the average TFR in the United States and thus the nation's age structure is quite modest. Immigration is not the reason America's fertility is higher than most Western countries.
Demographers have long known that immigration can have only a small impact on slowing the aging of lower-fertility societies. In an important 1992 article in Demography, the leading academic journal in the field, Carl P. Schmertmann showed that mathematically, "constant inflows of immigrants, even at relatively young ages, do not necessarily rejuvenate low-fertility populations. In fact, immigration may even contribute to population aging."10
The Census Bureau concluded in a 2000 report that immigration is a "highly inefficient" means for increasing the percentage of the population that is of working-age in the long run.11 In an article for the National Academy on an Aging Society last year I also showed that immigration has only a small impact on improving the share of the population that is of working age through 2050.12
Immigrants Are Not Especially Entrepreneurial
The author also makes the argument that immigrants are especially entrepreneurial. Like labor force participation, the actual data show there is no meaningful difference between immigrants and natives in entrepreneurship. Holtz-Eakin links to a report showing that in 2007 "10.5 percent of the immigrant work force owns a business compared with 9.3 percent" of natives. This is hardly a big difference. Equally important, these rates include teenage workers, who have very low rates of entrepreneurship and very few of whom are immigrants.
In 2011 the share of employed immigrants ages 25-plus who reported they were self-employed was 11.5 percent, and 11.7 percent for natives, which is not a statistically significant difference. Of the self-employed, the average income for natives was $21,116 and $22,372 for immigrants. Natives enjoyed a slight advantage in the number of workers they employ, with about 17 percent indicting they had more than 10 employees compared to 13 percent for immigrant entrepreneurs. An additional 2 percent of natives and 1 percent of immigrants were self-employed part time.13 The report Holtz-Eakin cites actually says the same basic thing. The report states that "Of total business income in the United States, 12 percent is generated by immigrant business owners." To place this in perspective, immigrants are roughly 12 percent of the total population, making them quite average.
It is very hard to argue that a high rate of entrepreneurship is a distinctive characteristic of immigrants overall. Nor is it lacking among immigrants. Immigrants from some countries do have high rates of entrepreneurship, while those from other countries have very low rates. But their overall rate is very similar to that of natives. Perhaps Holtz-Eakin would like to select immigrants from regions of the world whose immigrants to the United States have high rates of self-employment. But he does not make that argument.
Immigration's Fiscal Impact
One of the biggest weaknesses of Holtz-Eakin's article is that he is totally silent on the fiscal impact of legalizing 11 million illegal immigrants, which is at the very center of the current debate. On average, illegal immigrants have about 10 years of schooling. The research is clear that less-educated immigrants are a large net fiscal drain. The 1997 National Academy study cited above found that an individual immigrant without a high school education used $89,000 more in services than he paid in taxes during his lifetime, excluding any costs associated with his children. The figure was negative $31,000 for an immigrant with only a high school education. More educated immigrants were found to be a fiscal benefit. Who is allowed into the country and who is allowed to stay in the country has enormous fiscal implications, but not in Holtz-Eakin's analysis.
Another key feature of the Senate immigration plan is a new worker program that will admit "lesser-skilled" immigrants, many of whom will be able to apply for permanent residence. Is this a good idea given what we know about the correlation between education and fiscal effects? Again, Holtz-Eakin never mentions it. Yet he still gives a specific estimate of the economic and fiscal benefit immigration creates. Even if he thinks a bigger, more densely settled country is a richer country, he still must address the question of whether the gains that he supposes come from population growth offset the fiscal impact of the new immigrants.
In making his case that immigration-induced population growth by itself will have a positive indirect impact on the economy and public coffers, Holtz-Eakin ignores the academic research showing that immigration does not significantly increase the income of natives. He also ignores the research that has examined the actual impact of immigration on public coffers. He can do this because he is not really interested in the actual characteristics of immigrants, even though all available evidence suggests that the fiscal impact of immigrants depends heavily on their education at arrival. His view is that by making the population larger and the country more densely settled, immigration will make the country richer, all evidence to the contrary.
There might be a fiscal benefit from immigration-induced population growth. But Holtz-Eakin has certainly not made a convincing case for it.
1 Douglas Holtz-Eakin, "Immigration Reform, Economic Growth, and the Fiscal Challenge", April 2013.
2 See "Projecting Immigration's Impact on the Size and Age Structure of the 21st Century American Population", Center for Immigration Studies, 2012. The Census Bureau was kind enough to provide their fertility, mortality, and migration assumptions so we could remove migration to discern its impact on population size and structure.
3 See "Immigration and the American Worker A Review of the Academic Literature", Center for Immigration Studies, 2013.
4 See Charles R. Hulton, "Growth Accounting when Technical Change Is Embodied in Capital", National Bureau of Economic Research, 1992.
5 Derek D. Headey and Andrew Hodge, "The Effect of Population Growth on Economic Growth: A Meta-Regression Analysis of the Macroeconomic Literature", Population and Development Review, 2009.
6 David Schrank, Bill Eisele, and Tim Lomax,"Urban Mobility Report, 2012", Texas A&M Transportation Institute, 2012.
7 See "Immigrants in the United States: A Profile of America's Foreign-Born Population 2012", Center for Immigration Studies, 2012. See Table 12 for welfare use and Table 33 for average federal income tax liability.
8 See Table 589 in "The 2012 Statistical Abstract", the United States Census.
9 Figures are from the public-use files of the October, November, and December 2012 Current Population Survey. To be in the labor force one has to be working or looking for work.
10 Carl P. Schmertmann, "Immigrants' Ages and the Structure of Stationary Populations with Below-Replacement Fertility", Demography, 1992 (Vol. 29, No. 4).
11 See F.W. Hollmann, T. J. Mulder, and J.E. Kallan, "Methodology and Assumptions for the Population Projections of the United States: 1999 to 2100", U.S. Census Bureau Population Projections Branch, Population Division, 2000.
12 See Steven Camarota, "Immigration and an Aging Society", in Public Policy and Aging Report, Spring 2012 (Vol. 22, No. 2), National Academy on an Aging Society.
13 See Table 13 of "Immigrants in the United States: A Profile of America's Foreign-Born Population, 2012" Center for Immigration Studies, 2012.