By Lawrence H. Summers
The Washington Post
The impulse behind the latest “big” progressive idea of creating a federal job guarantee is entirely valid. Studies show that those without jobs are much more likely to be dissatisfied with their lives, to become addicted to alcohol or drugs and to be abusive within their families than even those working at low wages they find inadequate.
On this point, the U.S. economy is falling short of its potential. The fraction of the adult population between ages 25 and 54 that is working or seeking work has declined over the past 20 years. Despite America’s vaunted labor-market flexibility, the chance that a 25- to 54-year-old man will be out of work is much greater than it is in France and not very different than what it is in Spain . And in sharp contrast to the rest of the world, the fraction of adult women working in the United States has been declining since 1999 .
These trends are important causes of the increasingly bitter nature of U.S. politics and of resistance to technological change and overseas trade. President Trump received disproportionate support in parts of the country where joblessness increased most.
If the United States could guarantee jobs in even a modestly efficient manner and in a way that significantly increased employment, it would be a very good thing. I want to be enthusiastic about job-guarantee proposals. But at a time when cynicism about government runs strong, it is important for progressives to avoid making promises that they cannot keep. We must rigorously examine the practicality of a job guarantee.
A first question is how much to pay. A program of last-resort employment could likely provide the minimum wage and low benefits. But that will not help those laid off from highly paid manufacturing jobs or those who expect to earn wages well above the poverty line. While such a proposal could help many young people, it is far from clear that it would connect with the principal concerns of Rust Belt adults.
On the other hand, if the guaranteed jobs paid premium wages, say double the $7.25 per hour national minimum wage, they would be an attractive alternative for a quarter or more of the workforce, raising questions of cost and economic disruption.
Suppose a $15 per hour guaranteed job drew 4 million additional people into the workforce and also attracted 10 million existing employees, just one quarter of those for whom it would represent a wage increase. The cost, once benefits, materials and supervisory needs are included, would, conservatively, be $60,000 per worker. That would increase government spending by $840 billion — one-fifth of the current federal total. If wages for the 30 million lower-wage workers who remained in the private sector went up by just $4 per hour, private-sector costs would rise by $240 billion. The burden would largely fall on small businesses and disproportionately hurt restaurants and other major employers of low-wage labor.
A second question job guarantees raise is what all these new workers would do. The current federal government civilian workforce comprises 2 million people . Meaningfully increasing employment or offsetting adverse trends even if all hired came off the sidelines of the workforce through a jobs program would require boosting the federal workforce by at least 50 percent.
The United States has large needs, for example in infrastructure and care of the aged. At the federal level, these are met through contracting, not direct hiring. Using an employment guarantee to address these national problems would require significant restructuring of the way services are provided, likely with an efficiency cost.
A final question concerns the macroeconomics of a jobs guarantee. If the Federal Reserve saw the budget deficit expanding substantially, a tightening of the labor market or upward pressure on wages, it would probably respond by raising interest rates significantly. This would discourage spending and offset the employment gains from a guarantee scheme. If, on the other hand, the program was financed with new taxes, demand from those who paid the taxes would go down. That would reduce private-sector employment and offset the gains from an employment guarantee.
It would be terrific if these questions had persuasive answers. But right now, I am inclined to think that the idea of a jobs guarantee should be taken seriously but not literally. A combination of wage subsidies, targeted government spending, support for workers with dependents, and increased training and job-matching programs represent a more viable strategy for meeting demand for guaranteed employment.