Skill Composition and Immigration

The purpose of this post is not to advance a policy idea, but to articulate what is at stake. Immigration is one of the most difficulty policy issues to talk about, because it overlaps so much with touchy issues of identity politics and race. But, particularly in light of President Obama’s recent order to halt the deportation of some categories of undocumented immigrants, I think it’s important to step back and take a look at immigration through more abstract economic and ethical lenses, in the hopes that this will help us have smarter and fairer conversations in the future. The way I want to come at this is first by reviewing the conventional wisdom, and then looking at how that conventional wisdom is being challenged by my favorite policy writers of the heterodox center-left and the heterodox center-right (Matthew Yglesias and Reihan Salam, respectively).

So, first, the conventional wisdom: Almost everybody in the policy world seems to agree that high-skill immigration is an unqualified good — a no-brainer. High-skilled immigrants — let’s define them as those with advanced professional degrees, or with STEM degrees, or who own their own companies — help drive technological and business innovation that makes the country as a whole more prosperous. We want to keep them here, in the U.S., so that American workers will have job opportunities at their firms, so that American companies can own their patents, so American governments can get more revenues generated therein, etc. A rising tide lifts all boats, and since high-skilled immigrants help make the American economy bustle, they make working Americans better off. The only group of people who could conceivably be hurt by high-skilled immigration is high-skilled Americans, who would, in the short run, face a more competitive market for their skills. But we high-skilled Americans are literally the most privileged and pampered class of people in human history, so it seems fair to expect us to fend for ourselves. So everyone agrees that high-skilled immigration = good.

The other half of the conventional wisdom is that the other half of immigration — that is, low-skill immigration — is, at the very least, a little bit more complicated. The thinking here is that low-skilled immigrants aren’t likely to invent great new things or make discoveries or start companies. They’re more likely to just accept very low wages for service jobs, which should push down the wages for, and contract the employment opportunities of, native-born low-skilled American workers. Now, plenty of intellectuals still advocate for low-skilled immigration — sometimes on the cosmopolitan ethical grounds that there’s no moral reason to privilege the economic needs of human beings who happen to be native citizens over the more dire economic needs of human beings who could potentially be citizens in the future. But this position isn’t considered politically palatable (American voters are, after all, theoretically already American citizens). This is why both parties try to stick to advocating high-skilled immigration, staying away from the trickier issue. Focusing on high-skilled immigration, as Matt Yglesias notes with some derision, has thus become the fashionable thing to do.

So now let’s complicate this conventional wisdom. From the center-left, Matt Yglesias provides some evidence that, actually, the idea that low-skilled immigration lowers the wages of low-skilled native-born Americans is false. In fact, the Economic Policy Institute (which is labor-funded, hence not exactly a neo-liberal shill group), has actually found that it tends to raise wages. How could this be? What’s the theory here? Well, in truth, the “low-skilled labor market” is not just a single, fixed, homogenous thing. This is because (#1) the low-skilled labor market itself divides into several relatively distinct tiers, and (#2) markets are dynamic over the long run. (#1) means that low-skilled immigrants in practice mostly occupy the most low-level service jobs, e.g, dishwashing, which employ relatively few native-born Americans — so low-skilled immigrants in this tier are mostly competing against other immigrants, and limiting immigration on those grounds, Yglesias notes, seems a bit perverse. And (#2) means that having more immigrants in these jobs, i.e., having more bodies to meet our economy’s demand in these domains, is good for the economy as a whole, over the long run, which lifts all other boats. In other words, over the long run, bringing more people into this very-low-skill labor markets benefits all other workers, because these labor markets are “complementary.” If a local restaurant can suddenly survive because it can get lots of cheap dishwashers, this could mean it would employ relatively high-skilled business managers and lawyers, etc. So increasing the labor supply in any particular employment tier benefits all the others. Got it? Okay. So, let’s, for now, accept the EPI’s empirical analysis and Yglesias’s theoretical explanation. (I note in passing the irony that, according to Yglesias’s analysis, the only people who have economic grounds to oppose low-skilled immigration are low-skilled immigrants. This is, of course, not how the debate plays out in politics, which shows how identity politics matter more than economic considerations in our policy debates…)

Now, from the heterodox center-right, Reihan Salam objects. He doesn’t contest Yglesias’ short-run analysis, but argues that we need to do a “multigenerational analysis,” i.e., thinking about “how the future will unfold in a realistic, unsentimental way.” He points to research suggesting that the children and grandchildren of immigrants from south of the border are not closing the skills-gap with native-born workers. Now, I think this research could be contested and critiqued from a variety of angles, but let’s, again, accept it for now, so we can think analytically about the issues at stake. If the children of immigrants from SOTB aren’t gaining the skills of the children of the native-born population, why would this be the case? Well, one possibility is that there are vicious cycles here — because they work low-wage jobs, low-skilled immigrants end up residing in downscale neighborhoods and school districts, where their children are less likely to acquire the cultural capital, education, ambitions, bourgeois-workplace mores, future-orientation, etc., needed to succeed in higher-skill professions. So low-skilled immigrants can get their families into intergenerationl ruts, so to speak. Another possibility is that there are deeply ingrained cultural factors at play here — the fact that the children of Chinese immigrants are more well represented in elite universities and professions than the children of Mexican immigrants, though the latter compose a larger portion of the American population, suggests that we may need to seriously consider these factors. But whatever the explanation, the implication is the same: large-scale immigration from south of the border could lead to a long-term decrease in the “skill composition” of the American population as a whole. And that would actually be bad for everyone. The reason why both high-skilled software engineers and low-skilled barbers in Israel make more money than their equivalents in Guatemala is that there are large, very good, very positive “network effects” from having a population that is, in high proportion, high-skilled. Having a higher concentration of smart, skillful people makes your economy as a whole more prosperous — heck, it even makes your block parties better.

The implication, then, is that if it is true that certain immigration patterns could lead to a long-term reduction in the skills of the American workforce as a whole, then those particular immigration patterns could be bad for America as a whole, bad for America’s place and influence in the world in the future, and bad even for future generations of aspiring immigrants in America. This could be particularly problematic, as Reihan notes, in light of ongoing technological changes that are increasing the wages for high-skilled workers, while displacing the employment opportunities for lower-skilled workers. This picture becomes seriously problematic when you couple it with America’s growing debt and burdened welfare state. Notably, immigrants are not, today, net drains on the the American welfare state (that is, current evidence suggests they pay out more in taxes than they receive in benefits). But, if low-skilled immigrants largely get sucked into communities with “vicious cycles” of underperformance, and technological changes further erode their employment opportunities, they could, in the future, become drags on the fisc.

Now, let me step back for a bit. I wanted to articulate Reihan’s argument because I think it deserves very serious consideration, particularly from readers who are inclined to suspect that objections to immigration can only be based on an underlying prejudice. But I’m not sure I accept that this line of thinking provides a strong argument for imposing greater limits of immigration. Here are a couple things that come to my mind: (1) As a matter of basic moral theory, it does not seem to me that America’s fiscal position, or even America’s national prosperity or greatness as a whole, should be our single, overarching ethical metric. Suppose we accept that more low-skilled immigration will hurt America’s long-term prosperity. Well, my landlady’s plans to live and collect Social Security checks until she is 105, also erode America’s fiscal position. Her desire for a long life is pit against my interests as a young person who hates paying taxes. But, despite this, I don’t hope she dies soon, because I love her. Loving our fellow human beings as ends in themselves — which is what taking morality seriously requires — means that we do what we can to meet their urgent needs (such as by allowing them to migrate to our prosperous, stable country), even if it could, over the long run, hurt the American per-capita GDP figure. (2) To a large extent, Reihan’s argument, as he himself notes, implicates other policy issues, most notably education and residential segregation, as they relate to social mobility. If it were true that the children and grandchildren of low-skilled immigrants became as highly skilled as those of native-born Americans, then there would be no strong argument against large-scale immigration. So improving education for the poor, such as through more extensive vouchers and experimentation with charter schools, ought to be the highest priority. More broadly, Reihan and I both believe that broad market-based reforms, and a reduction in the American welfare bureaucracy, could help produce broad-based prosperity and employment opportunities that could reduce welfare dependency, nulling concerns about the intergenerational fiscal impact of immigration. This, I think, shows how many of our policy issues really can’t be considered in isolation — they’re all bound together.

So let me close this with a few relatively modest claims that we can take forward into future immigration debates: (1) Figuring out what is the most rational immigration policy hinges on a lot of open empirical questions: i.e., how are current immigration patterns influencing the skill composition of the population as a whole, and what power do we have, through both education and immigration policy, to change that? (2) There are also open ethical questions. If, indeed, some immigration patterns are reducing the skill composition of the U.S. in ways that could hurt the country’s long-term prosperity, we need to talk seriously about ethical questions that aren’t always taken seriously in the technocratic/economics-oriented policy world — i.e., do we have a humanitarian obligation to weigh the interests and needs of all human beings equally? Or do we have a nationalist obligation to privilege the interests of American nationals? Is there an ethical reason to privilege the interests of undocumented immigrants who are already within American borders over prospective immigrants who are currently in, say, Nigeria or Bangladesh? And (3) we should be wary of allowing the issue of immigration in the U.S. to be transformed into a matter of ethnic identity politics. There are both humanitarian/cosmopolitan ethical grounds and practical/economic reasons (as Reihan and Yglesias both agree in their posts) to think that a larger portion of immigration slots should be allotted to more diverse groups of migrants from the poor parts of all of the continents of the world.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s