David R. Francis
The Concise Encyclopedia of Economics

Unemployment Insurance

by David R. Francis
About the Author
[Editor's note: some of the data have changed since this article was written in 1992. The overall structure of the unemployment insurance, however, has remained intact.]

The United States unemployment insurance program is intended to offset income lost by workers who lose their jobs as a result of employer cutbacks. The program, launched by the Social Security Act of 1935, is the government's single most important source of assistance to the jobless.

A second goal of the program is to counter the negative impacts on the national economy, and especially on local economies, of major layoffs, seasonal cutbacks, or a recession. Unemployment benefits help sustain the level of income and hence the demand for goods and services in areas hard hit by unemployment. In short, unemployment insurance supports consumer buying power.

Not all unemployed workers are eligible for unemployment insurance. In fact, from 1984 to 1989 the proportion of the unemployed receiving benefits was at or below 34 percent every year. Benefits are not paid to employees who quit their jobs voluntarily or are fired for cause. Nor are they paid to those who are just entering the labor force but cannot find a job, nor to reentrants to the labor force who are looking for work. In February 1991, 76 percent of the target population of "job losers"—those involuntarily laid off—received benefits.

The proportion of unemployed workers who receive benefits is always higher during recessions than during expansions. This is because during recessions a higher fraction of the unemployed are people who were laid off. By January 1991, 46 percent of total unemployed workers claimed unemployment benefits, the highest percentage for that month since 1983.

Under the joint federal-state program, most states pay a maximum of twenty-six weeks in benefits, starting after a one-week waiting period. A few extend the duration somewhat longer. These benefits replace about one-third of gross wages for people with average or below-average incomes. The average weekly benefit in 1991 was about $161. When a state's unemployment is substantially above the national average, the program provides for up to an additional thirteen weeks of benefits. Five states were paying "extended benefits" in the winter of 1991, but this number approximately doubled by the end of April as the recession and unemployment worsened. The state and federal government share, approximately equally, the cost of extended benefits. During the eighties many states raised their "triggers"—the unemployment rate that must be reached—for extended benefits. As a result relatively few workers were eligible for extended benefits.

The federal government makes grants to the states for the administration of the unemployment insurance program. These grants exceeded $2 billion in fiscal 1991, ending September 30, 1991. The money helped pay the wages of about thirty-seven thousand state workers who administer the program and who dispense benefits from state unemployment insurance funds. In that fiscal year states collected about $16 billion in unemployment taxes from employers to cover the cost of the program; the federal government collected approximately $4.4 billion. Outlays on benefits were expected to run about $18.7 billion in fiscal 1991.

Federal law requires all state governments to impose a tax on employers of at least 0.8 percent on each employee's first $7,000 of pay. The tax base exceeds $7,000 in thirty-six states, with a national average of about $8,500. The highest base is $21,300 in Alaska. Most states levy a higher tax rate on businesses that have higher layoffs. However, the tax rate cannot go below the minimum even for businesses that have no layoffs. Nor do states set the maximum high enough so that employers with high layoff rates generate enough tax revenues to pay all the benefits to the workers they lay off. The result is that workers and businesses in industries with low layoff rates subsidize workers and businesses in industries, such as construction, with high layoff rates. Harvard's Martin Feldstein suggested in 1973 that this subsidization of layoffs would cause more layoffs. The evidence indicates that he was correct. Economist Robert Topel of the University of Chicago estimates that if employers could expect to repay (in taxes) the full value of unemployment benefits drawn by their laid-off workers, then the unemployment rate would fall by as much as 1 full point (e.g., from 6 percent of the labor force to 5 percent).

A basic tenet of economics is that when an activity is subsidized, people do more of it. Does unemployment insurance—a subsidy for being unemployed—increase unemployment by prompting the unemployed to delay their search for a new job or to search longer for a better position? Economists have found that it does. A 1990 study by Bruce D. Meyer, an economist at Northwestern University, found that a 10 percent boost in the "replacement ratio"—the proportion of after-tax work earnings replaced by unemployment benefits—causes unemployed people to extend their time without work by an average of 1.5 weeks. (During fiscal 1990 the average duration of benefits for the jobless was 13.6 weeks.)

Most people who receive unemployment insurance find a job or are recalled to work in the first several weeks. Meyer also found that among those who remain jobless for a longer period, the chance of a person on unemployment insurance going back to work increases rapidly as the time of benefit exhaustion approaches. Indeed, the chances of an unemployed person getting a job triples as the length of remaining benefits drops from six weeks to one week. Meyer suspects some of the jobless may have arranged to be recalled to previous work or to begin new work about the time their benefits expire. "If workers are bound to firms by implicit contracts, moving costs, specific human capital [education, experience, skills, etc.], or other reasons, firms have an incentive to base recall decisions on the length of UI [unemployment insurance] benefits," noted Meyer in a study done for the National Bureau of Economic Research. Unionized firms tend to take greater advantage of this "layoff subsidy" than do non-union establishments. And not surprisingly, given the incentives, layoffs are more common for those eligible for unemployment benefits than for those not eligible. If benefits are extended beyond twenty-six weeks, the unemployed tend to stay out of work nearly a day longer, on average, for each week of the extension.

Lawrence H. Summers, chief economist at the World Bank, and chief economic adviser to Democratic presidential candidate Michael Dukakis in 1988, reaches similar conclusions. Summers, along with Harvard economist Kim B. Clark, found that unemployment insurance almost doubles the number of unemployment spells lasting more than three months, thereby encouraging long-term joblessness. Summers and Clark suggest that unemployment insurance benefits cause many of the long-term unemployed to have high "reservation wages." Translation: to accept a job, these unemployed workers insist on getting a high wage, and if they aren't offered that wage, they stay on unemployment insurance as long as possible.

Economists have proposed various reforms to reduce the adverse effects of unemployment while still assisting people who lose their jobs. One of the more modest reforms suggested has been to reduce the minimum tax rate on employers and raise the maximum tax rate, so that the taxes they pay more closely reflect their layoff rates. A more extreme proposal, made by Robert Topel, is to experience-rate individual workers so that workers with a history of long unemployment spells pay higher tax rates. The federal government has already adopted one reform suggested by economists across the ideological spectrum. The 1986 Tax Reform Act eliminates the tax bias in favor of unemployment insurance by taxing unemployment benefits just like other income.

About the Author

David R. Francis is an economic journalist with the Christian Science Monitor.

Further Reading

Becker, Joseph M. Experience Rating in Unemployment Insurance: An Experiment in Competitive Socialism. 1972.

Feldstein, Martin. "The Economics of the New Unemployment." Public Interest 33 (Fall 1973): 3-42.

Summers, Lawrence H. Understanding Unemployment. 1990.

Topel, Robert. "Unemployment and Unemployment Insurance." Research in Labor Economics 7 (1986): 91-135.

Topel, Robert. "Financing Unemployment Insurance: History, Incentives, and Reform." In Unemployment Insurance: The Second Half Century, edited by W. Lee Hansen and J. Byers. 1990.

Return to top