Google+
Close

The Agenda

NRO’s domestic-policy blog, by Reihan Salam.

Michael Clemens Latest on Less-Skilled Immigration



Text  



I’ve often discussed Michael Clemens of the Center for Global Development, in part because I admire his highly original contributions to the study of labor migration and because I strongly disagree with his normative framework and the policy recommendations that flow from it. In Foreign Policy, he makes the case for a dramatic increase in less-skilled immigration, criticizing the Senate immigration bill on the grounds that it doesn’t go far enough to meet demand for less-skilled labor.

According to the U.S. Bureau of Labor Statistics (BLS), the United States will need 3 million additional workers over the next decade to fill the least-skilled jobs — jobs that do not require a high school degree — in order to achieve projected economic growth. These include jobs in home health, food preparation, freight, child care, cleaning, landscaping, and construction. Over the same period, the total number of U.S. workers entering the labor force at all skill levels, between the ages of 25 and 54, will be 1.7 million. (At younger ages, 24 years and below, the labor force will actually shrink.)

Think on that a moment. Even if every single young American dropped out of college and high school now, so that at the end of the coming decade they would be performing essential less-skilled jobs, they could only fill about half of these new openings. And of course all those kids won’t and shouldn’t stop getting educated. Around 10 percent of those new labor-force entrants will have less than a high school education; around 30 percent will have high school only.

Bottom line: At least three out of four of these new, essential jobs will be filled by workers coming to the United States from abroad, or they won’t be filled at all. This has nothing to do with laziness. It’s about numbers. It’s not only that there aren’t enough less-skilled Americans to do these jobs. There aren’t enough Americans period.

First, Clemens characterizes the BLS projections in a very unusual way. The BLS has not devised its ideal portrait of what the U.S. economy ought to look like a decade from now — targets that we must meet by, for example, dramatically increasing less-skilled immigration. If the less-skilled population grows by less than the BLS anticipates, firms and households will adjust. The following is drawn from the BLS:

Occupations that require some postsecondary education are expected to experience slightly higher rates of growth than those which require high school diploma or less. Occupations in the master’s degree category are projected to grow the fastest, about 22 percent; occupations in the bachelor’s and associate’s degree categories are anticipated to grow by about 17 percent and 18 percent, respectively, and occupations in the doctoral or professional degree category are expected to grow by about 20 percent. In contrast, occupations in the high school category are expected to grow by just 12 percent, while occupations in the less than high school diploma or equivalent category are projected to grow by 14 percent (Chart 7).

Nevertheless, because many of the occupations require a high school diploma or less, they will account for the majority—63 percent—of new jobs between 2010 and 2020 (Chart 8). [Emphasis added]

Note the use of the word “require” — jobs that require a high school diploma or less are not necessarily limited to workers with a high school diploma or less. The slack labor market of the past half-decade has seen many jobs that do not require a high school diploma filled by individuals who have a high school degree or more. This is not an ideal outcome. Yet skilled workers will have a more difficult time climbing the economic ladder as jobs that can take greater advantage of their skills emerge, while less-skilled workers will have a more difficult time doing the same. It is not at all obvious that every job should be held by the worker with the lowest possible skill level. Moreover, it is not obvious that jobs in home health, food preparation, freight, cleaning, landscaping, and construction can’t become more capital-intensive and less labor-intensive over time. Child care might be an exception, yet if the various other sectors evolve in such a way that labor demand increases, one assumes that there will be more workers available to work in child care. A number of affluent market democracies face shrinking populations in the near future, including countries widely regarded as fairly successful, like Taiwan, Germany, South Korea, and Japan, the last of which is facing a population collapse in the coming decades. The U.S., in contrast, will have a steadily growing population. Labor-saving innovations devised in these and other societies in the coming decades will travel across borders, allowing U.S. firms to deploy them as necessary as the supply of less-skilled workers shrinks. 

And of course the BLS is making assumptions about the medium-term economic future that may well be proven false, as has been the case with BLS projections in the past that assumed, for example, that manufacturing employment wouldn’t decrease as sharply as it has, etc. Economic growth reflects both productivity increases and increases in the size of the workforce. To observe that a smaller increase in the size of the workforce will yield a smaller increase in economic growth isn’t in itself terribly compelling. One can easily imagine a scenario in which the size of the workforce increases by less than the BLS projects, yet a higher proportion of the workforce consists of workers with a high school diploma or more. A more important consideration is GDP per capita or GDP per worker hour. And while better labor matching might yield somewhat higher GDP per capita — the CBO projects a 0.2 percent increase in GNP per capita after two decades and a slight decrease after one — technological change, discussed below, might yield a different result, and immigration policy is essentially irreversible. That is, if it turns out that labor-saving innovation does not completely grind to a halt in 2013 and the labor market position of less-skilled workers continues to deteriorate, we can’t decide 20 years hence that allowing for a substantial influx of less-skilled immigrants was a terrible mistake and that we ought to expel less-skilled immigrants who arrived after that date and their dependents, as a large number will be citizens. 

Clemens never mentions the unemployment rate for U.S. workers with less than a high school diploma, which is 10.7 percent as of June. Note that the unemployment rate reflects the share of workers with less than a high school diploma who are actively looking for work. I agree with Clemens — laziness is not the issue. Rather, less-skilled natives face a number of challenges, one of which is geographical mismatch. That is, less-skilled natives are often priced out of high productivity regions, due to restrictive land-use policies. Addressing restrictive land-use policies might greatly facilitate labor matching, while yielding other economic and environmental benefits as well. Another serious challenge is that a disproportionately large share less-skilled U.S. workers suffer from drug and alcohol dependence, a challenge that U.S. firms are disinclined to address. Sharply increasing the supply of less-skilled immigrants greatly reduces the costs of drug and alcohol dependence in the native-born population for firms. This in turn presumably makes it somewhat less likely that steps will be taken to mitigate drug and alcohol dependence

Clemens’ article continues:

The Senate’s recently passed bill would create relatively small numbers of work visas for low-skilled jobs. (The House has no bill yet.) The Senate bill would create only a few hundred thousand temporary work visas at a time — the “W” visa — with a floating cap that could rise to a hypothetical maximum of 600,000 people in the country at any one moment. And it creates a negligible number of low-skill permanent work visas.

Who will staff the millions of new low-skilled jobs the U.S. economy will create in the next decade, jobs American workers will not fill? There are the current undocumented workers in the country who may be regularized under immigration reform — but they’re already in the United States, so the BLS’s estimates of new jobs already account for the jobs they fill. Perhaps some of these jobs will be filled by people who come on family-reunification visas, but no one knows how many of them there will be, and it will almost certainly be too few. Family-based immigrants have not been filling many of the low-skilled jobs currently filled by unauthorized workers, and the Senate bill will reduce the number of family-based visas.

Who will staff the millions of new low-skilled jobs the U.S. economy will create in the next decade, jobs American workers will not fill? There are the current undocumented workers in the country who may be regularized under immigration reform — but they’re already in the United States, so the BLS’s estimates of new jobs already account for the jobs they fill. Perhaps some of these jobs will be filled by people who come on family-reunification visas, but no one knows how many of them there will be, and it will almost certainly be too few. Family-based immigrants have not been filling many of the low-skilled jobs currently filled by unauthorized workers, and the Senate bill will reduce the number of family-based visas.

Note that Clemens is characterizing “a few hundred thousand temporary work visas” as a small number, and that the W visa program allows workers to bring spouses and children with them to the U.S. In contrast, the Canadian guest worker program forbids workers from bringing spouses and children, to deter workers from overstaying their visas. Many observers, including Matt Yglesias, like Clemens a supporter of increasing less-skilled immigration, have suggested that the structure of the W visa program will lead to unauthorized immigration. And this is actually Clemens’ strongest argument — restricting the size of the less-skilled influx will tend to increase unauthorized immigration:

All this implies that even if the United States regularizes millions of immigrants now, it is likely to have another unauthorized immigration crisis several years ahead. The last mass regularization, under President Ronald Reagan in 1986, barely altered the stock of unauthorized immigrants in the medium term. Before the regularization there were around 3 million unauthorized immigrants; just five years later there were once again around 3 million. The main reason was that the reform was a political showpiece built around the politics of “amnesty” and “security,” rather than the needs of the U.S. economy. It was never designed to fill America’s economic needs for low-skilled labor, but instead was a rickety political compromise among farmers, labor and Hispanic groups, and other interests. In the aftermath, employers had an awful choice: either turn to the black market for labor or face the consequences of a low-skilled labor shortage — stunted businesses, closed farms, infants and grandparents without proper care. The latest efforts at reform might be the best that can be hoped for from Capitol Hill. But they will similarly herald a new wave of unauthorized immigration unless their low-skill work visa caps are made much more flexible, starting from the essential labor needs of U.S. employers. And that’s unlikely.

But the consequences of a less-skilled labor shortage can be viewed through another lens. Stunted businesses might be the product of environmental regulation, closed farms might be the result of the elimination of farm subsidies or the imposition of more broad-based inheritance taxes, and infants and grandparents might go without proper care due to higher minimum wages. Yet we might nevertheless endorse environmental regulations, the elimination of farm subsidies or the imposition of more broad-based inheritance taxes, and minimum wage laws on the grounds that these policies have other benefits. The rejoinder might be that less-skilled immigrants from the developing world are better off even if they labor under conditions that natives, including less-skilled natives, would find intolerable, and that is entirely true. But if our primary interest is in global poverty alleviation, the set of potential policy options grows considerably — we’re no longer having a debate about the policies that best serve the interests of U.S. workers. 

Finally, Clemens addressing matching:

How can that MBA launch a new product? Only by depending critically on a small army of essential low-skilled workers. She needs someone to clear the table after a client lunch, empty the garbage at her office, and run the lot where she parks. She needs someone to pick the vegetables she eats and resurface the road she drives home on. She might need someone to care for her child, or grandfather, so she can work late. All of that is just the beginning.

This other form of teamwork is less obvious because it’s often invisible. Apart from the care workers, this MBA might never meet or only briefly glimpse the rest of these less-skilled workers. Yet every step of her daily life critically depends on them as much as it depends on her skilled co-workers. They make her more productive, and she them.

This is simultaneously the reason that there are so many less-skilled jobs in America’s future and the reason that fewer Americans do those jobs. People who acquire higher skills create less-skilled jobs that they themselves aren’t suited for. Less-skilled immigrants fill that gap. Machines may be able to take over a few of these jobs; that’s why you’re seeing more self-checkout registers in retail stores. But no machine we’ll have anytime soon can help the elderly bathe safely, clear tables at restaurants, or profitably pick cucumbers. People who do less-skilled essential jobs make skilled work in America possible, and together they make American competitiveness possible. You might never see the people who clean Google’s offices, but the company’s massive contribution to U.S. competitiveness would not exist without them.

Assuming it really is true that not increasing the size of the less-skilled immigrant influx will make it impracticable for knowledge-intensive service workers to hire workers to clear tables, empty garbage cans, operate parking lots, resurfacing roads etc., one assumes that new technologies will emerge — disposable food containers, automated trash removal and automated parking lots, autonomous vehicles that can resurface roads – and that other functions can be offshored, e.g., the U.S. could import more vegetables from abroad. Child care and eldercare can be provided by skilled natives in some cases, or in nurseries or senior centers to increase efficiencies. Clemens acknowlesgt that automation might play a role — and so his argument rests on the notion that we need tens of millions of less-skilled immigrants to bathe the elderly, to clear tables, and to pick cucumbers. Again, cucumbers can be imported; the elderly might benefit from promising biomedical innovations that delay the onset of dementia, and also from the scale afforded by eldercare facilities; and restaurants in countries like Denmark, Switzerland, and Australia continue to exist despite relatively low levels of less-skilled immigration — they happen to be somewhat more expensive. 

I prefer New York city to Copenhagen, and I’m sure many other Americans would agree with me. But Copenhagen is not an economically unviable or unpleasant city, despite the fact that professionals outsource food preparation somewhat less. There are many things I’d like to be cheaper. The question is whether driving down the cost of labor-intensive services is worth some of the downside risks associated with sharply increasing the number of Americans who are not likely to earn enough to lead what citizens of affluent countries define as dignified lives in the absence of mean-tested transfers.



Text  


Subscribe to National Review