The March 6 issue of the New York Times contains an interesting article by the economist Paul Krugman entitled “Degrees and Dollars,” available online at http://www.nytimes.com/2011/03/07/opinion/07krugman.html. In it Krugman challenges the conventional view that we need to invest more in education because “everyone knows that the jobs of the future will require ever higher levels of skill.” Krugman argues that since about 1990 “the U.S. job market has been characterized not by a general rise in the demand for skill, but by ‘hollowing out’: both high-wage and low-wage employment have grown rapidly, but medium-wage jobs—the kinds of jobs we count on to support a strong middle class—have lagged behind.” He expects the trend to continue, noting a recent newspaper article about the growing use of software to do legal research, potentially replacing hordes of lawyers and paralegals engaged in document review in big cases. He argues that computers are good at doing both cognitive and manual work that can be performed effectively by following rules, whereas work involving a high degree of discretion or imagination, ranging from writing poetry to inventing a new social network to running a corporation (not his examples—he doesn’t give any examples of the type of nonmanual work that he thinks will resist automation successfully), along with some manual labor (he instances truck drivers and janitors, but could add waiters and retail sales personnel), cannot be. He argues that demand will grow for jobs in the categories in which work cannot be automated, and will decline for jobs in the categories that can be.
Hence, Krugman concludes, it would be “wishful thinking” to believe that “putting more kids through college can restore the middle-class society we used to have.” Instead, the focus should be on “restor[ing] the bargaining power that labor has lost over the last 30 years” and “guarantee[ing] the essentials, above all health care, to every citizen.” These are pathetic prescriptions. Unionization spurs automation by making labor more costly relative to capital, and providing health care to every citizen will increase the tax burden on persons with middle-class incomes, and not just on wealthy persons, because the costs of universal health care are too staggering to be borne entirely by the wealthy.
But the more interesting question is whether Krugman is right to be pessimistic about the future returns to a college education. The market disagrees. If the market agreed with him, college enrollments would be plummeting because college is expensive, not only in tuition but also in opportunity costs—the income forgone by being in school. College enrollments continue to increase relative to population, and, more important from a market perspective, more and more high-school students express a desire to go to college, even though if Krugman is right their college education will not produce lifetime earnings increments sufficient to offset the cost of tuition and the cost of their forgone earnings during their college years. (Of course a high unemployment rate, by reducing the opportunity costs of college, reduces overall college costs and so helps to sustain high enrollments, but this is not a major factor driving demand for college education.)
The market could be wrong, or (though this is unlikely) the nonpecuniary benefits of college could be increasing faster than the pecuniary benefits are falling. A high school student, and his parents, are hardly in a good position to predict the structure of the labor market in ten or twenty years. Most people base most of their expectations for the future on simple extrapolation from the recent past. Often that’s the best one can do in the presence of profound uncertainty. But the economic crash that began in 2008 has made us more alert for discontinuities in economic trends. The future doesn't always repeat the past. Krugman may be right that computers will replace many midlevel jobs—the current example is employment by bookstores, which is being killed by Amazon.com. Income inequaltiy has been growing in America (illustrating that there are negative as well as positive trends) and may continue to grow. One can even imagine the emergence of a large servant class, as in nineteenth-century England. Technological advance continues at a dizzying rate; no one can be sure where it is leading us.
One reason for skepticism, though, is that Krugman may have too limited a conception of the benefits of a college education. He may think that what a college education does is impart knowledge that the graduate needs either for his career or for the next stage in his education, such as law school or medical school. But it may be that the greater value of college lies elsewhere: in providing association with people of above-average intelligence, in imbuing young people with general work skills (discipline, working under supervision, following directions, being evaluated, basic writing and speaking skills, some foreign language proficiency), in fostering ambition, and in providing information to future employers about job applicants that enables better matching of workers with jobs. It is plausible that these noninformational benefits of a college education are valuable across a vast range of jobs and will enable the holders of those jobs to obtain good middle-class incomes. Police officers, prison guards, firemen, noncommissioned officers in the military, sales personnel, nannies, tour guides, secretaries, hotel receptionists, IT staff, medical technicians, store managers, auto mechanics, and many other workers who do not require a college education for their jobs may nevertheless be of significantly greater value to their employer if they have such an education and may be compensated accordingly. Finally, a college education may not only increase incomes in ways different from building up a stock of knowledge but also may reduce people’s living costs by making them more adept at household management.
But even if this is right (and Krugman wrong), it doesn’t follow that there should be massive increases in public expenditures on education with a view toward increasing the number of people who obtain a college education. The number of people who have the IQ and character traits that would enable them to benefit, in the ways described above, from a college education is inherently limited, and maybe all or most of them will obtain a college education under existing conditions for financing such an education.