Give Me Your Tired Arguments, Your Poor Reasoning

By Steven A. Camarota on March 28, 2015

National Review Online, March 28, 2015

In a very one-sided recent article for the New York Times Magazine, Adam Davidson tries to make the case for massive increases in immigration. He starts out by dicussing the racial prejudice of an older relative who always thought that Hispanics "were stealing jobs." Davidson believes this man's story is illustrative of the current debate over immigration.

Nothing delegitimizes someone in modern America more than the suggestion, or even the hint, that they hold unenlightened attitudes about race. It is not surprising that an advocate of high immigration would suggest that those with whom he disagrees are racists. But Davidson is supposed to be a professional journalist writing for a national news magazine.

Davidson argues that only a "handful" of Americans disagree with him about immigration, and this tiny minority may "make a lot of noise," but, like his racist relative, they are not voicing legitimate concerns. As Alexis de Tocqueville pointed out nearly two centuries ago, public sentiment is the ultimate source of legitimacy in American politics. So Davidson tells us that public-opinion polls show that a strong majority are in "favor of a path to citizenship for undocumented immigrants," citing a Pew poll to that effect. The whole point of this discussion is to make those who would like to enforce immigration laws and reduce immigration seem to be out of step with the American people.

In fact, the level of public support for amnesty depends heavily on how the question is asked. For example, this Reuters poll asked the question differently than Pew and found that a majority of Americans want most illegal immigrants to go home. And Kellyanne Conway of the Polling Company asked if people supported "encouraging illegal immigrants to go back to their home countries by keeping them from getting jobs and welfare" and found that two-thirds of the public agreed. Further, even if one thinks Americans overwhelmingly support legalizing illegal immigrants already here, it does not follow that they also support the current high level of legal immigration, let alone the vast increases that Davidson calls for.

The main point of the article is to assure us that the economic benefits of immigration may be "the most settled fact in economics." After this hyperbolic statement, Davidson makes the reasonable point that immigrants don't just increase the supply of workers and push down wages; they also increase demand for labor through their consumption of good and services. He then cites an influential 25-year-old study by David Card, a well-known economist, showing that the Mariel boatlift did not reduce wages in Miami some three decades ago.

Davidson seems unaware of the criticisms of cross-city comparisons of this kind, including that the movement of people, capital, and goods between cities makes this kind of analysis questionable. Also, Card estimated the impact just on Miami, based on a very small sample. And it is not even clear that the effect of immigration on one city 35 years ago tells us very much about the impact of immigration today on the nation as a whole. But in Davidson's mind, Card's study pretty much settled the question.

He does write more than once that Harvard economist George Borjas is a "lone" dissenter on these matters because he emphasizes the negative impact on some workers. But if Borjas is really so isolated, why are his immigration papers so frequently co-authored with other prominent economists, such as Jeffrey Grogger, Gordon Hanson, Richard Freeman, Lawrence Katz, et al.? Furthermore, Davidson seems unfamiliar with other studies showing that immigration reduces the employment of the native-born, such as this one by a Federal Reserve economist that indicates immigration has significantly reduced the employment of American teenagers, or this study and this study, both of which show that immigration reduces the employment of African Americans.

Davidson then turns to the favorite economist of high-immigration advocates, University of California economist Giovanni Peri. He uses Peri to make the argument that immigrants do not compete with natives, but instead complement them. The argument goes like this: The arrival of, say, immigrant maids lowers the cost of getting your house cleaned, and this allows more educated Americans, who can afford a maid, to spend more time working, producing valuable goods and services, and less time cleaning. But herein lines a dilemma. Immigration enthusiasts are loath to concede that there is any negative wage impact from immigration. But if that is so, how can immigration make hiring a maid more affordable?

One way around this dilemma is to assume that immigrants lower the wages only of other immigrants. Making this assumption allows one to say that the wages of natives are unaffected. The negative impact on immigrants already here does not seem to matter.

The idea that immigrants with little education do not compete with well-educated natives certainly makes good sense. But the idea that less-educated immigrants do not compete with less-educated natives is much harder to argue. As David Frum points out in the January issue of The Atlantic, the kind of analysis Peri does is based on some big assumptions, such as that native-born maids may get jobs supervising immigrant maids or move into another occupation and avoid competition with immigrants. Surely this does happen, but it often does not.

Analysis of government data by myself and Karen Zeigler shows that 51 percent of maids and housekeepers are native-born. The nearly 1 million American-born maids don't seem to have gotten the memo that this is now a high-immigrant job category and they need to get out. Looking at all 472 occupations as defined by the Department of Commerce shows that only six are majority immigrant, and they account for just 1 percent of the labor force. Relocating or changing jobs often involves significant costs, so staying put may make sense, even if one is getting hammered by immigrant competition.

As Frum observes, "a model based on unrealistic assumptions can still achieve perfect internal consistency. It just won't describe the real world very accurately." Peri himself does concede that immigration may lower wages significantly for less-educated natives. But this possibility never occurs to Davidson.

This points to something else in the labor market that is relevant to any discussion of immigration. If immigration reduces wages, those wages do not vanish into thin air. They may be retained by businesses in the form of higher profits, and there is certainly evidence that this is what's happening. There is general agreement that gains in productivity in recent decades have almost entirely accrued to owners of capital, not labor (see this study and this study). This is certainly consistent with the possibility that by increasing the supply of workers, immigration has shifted bargaining power to employers, allowing them to capture these gains.

It's also worth adding that, in addition to a long-term decline or stagnation in wages for most American workers, the share of working-age natives who are actually working shows a long-term decline that began well before the Great Recession. As recently as January of 2000, 52 percent of native-born Americans ages 18 to 65 without a high-school education had a job. Today it's 40 percent. For natives with only high-school education, it was 74 percent in 2000; today it's 65 percent.

The fact that workers in general and less-educated natives in particular have not fared well in the last 35 years is not what we would expect if immigration improved the employment prospects and wages of natives across the board. After all, the immigrant share of the work force roughly tripled after 1980, so if Davidson is right, this should have been a boom time for wages and employment. In fact, the opposite happened.

Davidson also misses half of the debate over how immigration impacts the economy — the impact on taxpayers. If you are going to argue that immigration makes Americans financially better off, as Davidson does, you have to at least mention the taxes immigrants pay minus the services they use. The National Research Council (NRC) estimated in 1996 that immigrant households (legal and illegal) create a net fiscal burden (taxes paid minus services used) on all levels of government of between $11.4 billion and $20.2 billion annually. This negative fiscal impact is larger than the economic benefit the NRC estimated from having immigrants in the labor market. Research by the Heritage Foundation also shows that the average household headed by a legal immigrant created $4,300 more in governmental costs than it paid in taxes in 2010.

As I reported in congressional testimony presented on March 17 of this year, Census Bureau data collected in 2014 show that 40 percent of immigrant households reported using one or more major welfare programs, compared with 25 percent of native households. Even if one thinks that immigration benefits all natives in the labor market, it is very possible that the likely fiscal drain they create entirely eats up that benefit.

Davidson ends his article by arguing that we could absorb as many as 11 million immigrants annually. That would mean 110 million immigrants in a single decade! That's far more than all the immigrants who have arrived in all of American history. The fiscal impact and absorption capacity of our public schools, health-care system, and infrastructure just doesn't seem to matter at all.

He states that the "State Department issues fewer than half a million immigrant visas each year" to people living abroad, apparently unaware that this represents less than half the annual number of people awarded lawful permanent residence, or green cards. That's because the rest of the year's green cards are issued to people who were admitted earlier with some temporary, "non-immigrant" status. Total permanent legal immigration is already about a million annually, twice what Davidson implies.

Davidson's article is similar to one he authored for the New York Times Magazine in 2013. One wonders why the editors of that publication feel no need to take a more nuanced view of both the costs and the benefits of immigration. Instead, they publish articles that read like press releases from an open-borders group.