My 1991 Critique of Extended Unemployment Benefits

Some things never change. Another President Bush was ambushed with extended unemployment benefits shortly before another presidential campaign. Some data in this oldie are dated (though not wildly different from today), but the arguments seem worth another look:

The Cure for Unemployment
Alan Reynolds
The Wall Street Journal , October 3, 1991

Democratic Congressmen hope to make George Bush look like a hard-hearted villain because of his reluctance to spend an extra $6 billion to extend unemployment benefits beyond the usual six months. Yet the current job situation is scarcely an emergency. Unemployment was higher than it is today in all but two of the dozen years from 1975 through 1986. Today, the average spell of unemployment — 14 weeks — is still lower than it was even as recently as 1987. Half of the unemployed find new jobs in fewer than seven weeks.

The congressional push to extend unemployment benefits aims to help a relatively elite minority of the unemployed. Last year, only 39% of the unemployed collected any benefits at all. This was largely because about half of those unemployed did not lose their jobs. They either quit their jobs, were reentering the labor force after a prolonged absence or were young people who had not held jobs before. Another reason many unemployed do not qualify for benefits is that they already have another job lined up, and are just taking some extra time off between employers. Or, they find a new job within three weeks — the waiting period to qualify. And, of course, unemployed illegal immigrants are less than eager to register with government agencies.

Those who are not eligible for unemployment benefits rarely take six months or more to find a job. Conversely, those unemployed for long periods are usually among those who do receive benefits, and often receive supplemental union benefits that can approximate their usual after-tax wages (particularly with some casual labor “off the books”). Moreover, cyclical layoffs account for most of the long-term unemployed, who, because of their seniority, have good reason to wait to be recalled.

Robert Topel of the University of Chicago figures that unemployment benefits could be extended to an entire year without spending another dime. How is that possible? Simply make people wait four weeks rather than three before they qualify for their first check. There are so many more people who collect benefits for a few weeks than for a few months that the savings from that one week would cover the costs.

Regardless of how extended benefits are financed, though, the unemployment rate would surely be significantly higher than otherwise, simply because more people would be subsidized to remain unemployed for longer periods. Studies in the Monthly Labor Review have shown that those who have supposedly been “unable to find a job” in 26 weeks miraculously find one within a few weeks after their benefits run out. They either quit waiting to be rehired in cyclical industries, or accepted second-best jobs that required, for example, moving to a new city.

Giving people almost a year to search for the “right” job could nonetheless be justified, in theory, because it is not in society’s interest to have many people working below their ability. But too long a period of subsidized job search is likely to reduce the intensity with which people look for work, and to delay economically desirable relocation of workers away from areas of high unemployment to areas where they are needed.

Another negative effect of prolonged benefits is that it would further subsidize employers that frequently lay off workers at the expense of those that do not. The tax employers pay for unemployment benefits is already too high, in an actuarial sense, on firms that provide stable employment, and too low on firms that do not. If layoff-prone employers had to bear more of the cost of the dole, they would adopt less volatile strategies of hiring and firing.

Over the longer haul, the more serious problem is not a shortage of jobs, but a shortage of people willing to work at the after-tax wages offered. From 1980 to 1989, the percentage of working-age people who were either working or looking for work rose from 63.7% to 66.5%, as marginal tax rates fell. But labor force participation rates began to dip at the start of 1990, and have now fallen back to 66.2%.

Many wives with working husbands, young people living with parents, and people of early retirement age have simply dropped out of the job market since the 1990–91 increases in Social Security taxes and in marginal federal and state income tax rates. They are indeed “discouraged workers,” but they are discouraged because their added work brings little added after-tax income, not because they couldn’t find jobs if they tried. Leaving the labor force means not trying.

Considering that the economy emerged from recession only a few months ago, the percentage of the unemployed who have quit their jobs is quite high — over 12%, compared with fewer than 8% in 1982–83. And new jobseekers (graduates) account for an unusually small share of the unemployed — 8.4% at mid-year, compared with 13% in 1984. Like the decline in labor force participation, this suggests the problem is not simply a shortage of jobs, but insufficient incentive, after taxes, to accept job offers and stay on the job. Congress should be less concerned about subsidizing lengthy periods between jobs, and far more concerned about tax policies that are shrinking the labor force and the tax base.