Machines are getting better and better at taking on tasks that until recently could only be performed by humans. Whether you welcome the prospect of labor-saving automation, as I do, or you dread it, few would dispute that it is one of the central facts of economic life in this century. Claire Cain Miller, writing for The Upshot, rounds up the usual suspects, led by Erik Brynjolfsson, co-author of The Second Machine Age, to make the case that while automation has long been seen as an economic boon for flesh-and-blood workers, who can increase their productivity by letting machines serve as beasts of burden, there is at least some reason to believe that things are changing — that instead of enhancing the productivity of all workers, automation might render at least some workers obsolete, or squeeze their wages even more than they’ve already been squeezed.
But which workers will be at greatest risk of obsolescence as technological progress marches on? While Miller observes that some white-collar jobs might in theory be displaced by automation in the years to come, and some already find themselves under pressure, she makes it pretty clear that it is less-skilled workers who are most vulnerable. Elsewhere, in Dancing with Robots, Frank Levy and Richard J. Murnane, two of the leading experts on the impact of automation on the labor force, find that “the spread of computerized work is increasing the importance of education.” This is almost a truism of the ongoing conversation about automation, and for good reason. So what I find odd is that Miller’s otherwise very thorough article never mentions that a disproportionately large share of less-skilled U.S. workers are immigrants, and that we’ve spent the last several years debating whether or not it is wise to welcome more less-skilled immigrants into the country.
For example, Miller notes that the average skill level of the U.S. workforce appears to have stagnated. Yet she gives no indication that this stagnation has occurred despite the fact that native-born adults have higher levels of educational attainment than their counterparts in earlier generations. The reason is that U.S. immigration policy, from our overemphasis on family unification over skill-based immigration to inadequate interior enforcement, has been lowering the average skill level of the American workforce. According to 2012 data from the Program for the International Assessment of Adult Competencies (PIAAC), 40 percent of foreign-born adults fall in the lowest two levels of literacy as compared to 14 percent of native-born adults. For numeracy, the picture isn’t much brighter: 48 percent of foreign-born adults fall in the lowest two levels, while the same is true of just 27 percent of native-born adults. So it should come as no surprise that “the American work force has gained skills at a slower rate than in the past.” Given the scale of the less-skilled immigrant influx, it is a minor miracle that the average skill level of the U.S. workforce hasn’t deteriorated.
For Miller, the most worrisome sign about how automation might impact the labor market is that the labor market is already functioning poorly for many workers:
More than 16 percent of men between the ages of 25 and 54 are not working, up from 5 percent in the late 1960s; 30 percent of women in this age group are not working, up from 25 percent in the late 1990s. For those who are working, wage growth has been weak, while corporate profits have surged.
Like Miller, I find these numbers alarming. Yet it is important to note that labor force participation is much higher for those with higher levels of education than for those with lower levels of education. The share of adults with less than a high school education who are not in the labor force was 59.9 percent in 2013; for those with a bachelor’s degree, it was 24.6 percent. Many of the immigrants that President Obama celebrated in his recent address will find themselves displaced as (to draw on a few examples from Miller’s article) machines administer sedatives to elderly patients and as robot bellhops help meet the needs of patrons. Or Miller might have cited the so-called “lettuce bot,” which is quickly mastering a task that once required many human hands, or the more prosaic ordering machines that many European quick-service restaurants are using to economize on labor.
There will, of course, be non-routine manual tasks that only humans can perform for some years to come. But as workers are displaced from other sectors and job roles, the competition for these positions will likely intensify. Is it urgently necessary that we increase the number of workers who will be competing for these positions even so? To be sure, this strategy might encourage employers to rely on less-skilled labor rather than to invest in labor-saving technology. I’m not clear on how this, on balance, will improve the negotiating position of less-skilled workers, and thus how it will foster wage growth.
Assuming that increased automation has a more deleterious impact on labor market outcomes for less-skilled workers than for high-skilled workers, increasing less-skilled immigration seems profoundly unwise. For one thing, it will increase the number of U.S. workers threatened by economic marginalization. For another, it will have an impact on the children of less-skilled immigrants, as less-skilled workers facing severe wage pressure will struggle to provide their offspring with the resources and the foundational skills they will need to successfully pursue post-secondary education. If I’m wrong, I’d like to see someone carefully explain exactly why I’m wrong.