I think this likely a substantial underestimate unless something is done for a number of reasons. First everything we hear and see regarding technology suggests the rate of job destruction will pick up. Think of the elimination of drivers, and of those who work behind cash registers. Second, the gains in average education and health of the workforce over the last 50 years are unlikely to be repeated. Third, to the extent that non-work is contagious, it is likely to grow exponentially rather than at a linear rate. Fourth, declining marriage rates are likely to raise rates of labor force withdrawal given that non-work is much more common for unmarried than married men.
On the basis of these factors, I expect that more than one-third of all men between 25 and 54 will be out work at mid-century. Very likely more than half of men will experience a year of non-work at least one year out of every five. This would be in the range of the rate of non-work for high school drop-outs and exceeds the rate of non-work for African Americans today.
Will we be able to support these people and a growing retired share of the population? What will this mean for the American family? For prevailing ethics of self-reliance? For alienation and support for toxic populism? These are vital questions. Even more vital is the question of what is to be done.
These questions should preoccupy social science researchers. They are vital to our future.
Lawrence H. Summers, the Charles W. Eliot university professor at Harvard, is a former Treasury secretary and director of the National Economic Council in the White House. He writes occasional posts on Wonkblog about issues of national and international economics and policymaking.