Report | Wages, Incomes, and Wealth

The broad reach of long-term unemployment

Issue Brief #194

Download PDF

The broad reach of long-term unemployment

by Andrew Stettner and Jeffrey Wenger
National Employment Law Project & Economic Policy Institute

Long-term unemployment is the scourge of a declining economy. After the unemployment benefits run out, the bills are overdue, and the retirement fund is tapped, many Americans are still looking for jobs. Some people may believe that the long-term unemployed, those unemployed for more than six months, are not looking hard enough for work, do not have enough education, or are not good workers. The reality is that the long-term unemployed are better educated, older, and more likely to be professional workers.

In 2002, 18.3% of the unemployed spent more than six months looking for work; this number had increased to 21.8% by April 2003. In particular, long-term unemployment became more concentrated among three groups:

• mid-career workers (those over 45 years old)—34.8% of long-term unemployed workers were over 45 years old;

• college graduates—18.1% of the long-term unemployed had college degrees; and

• executive, professional, and managerial workers—20.1% of the long-term unemployed came from these occupations.

In April 2003, unemployment increased to 6.0%. Throughout this recession, analysts have anticipated higher unemployment rates resulting from job losses and weak economic growth in the United States. Despite the weak economic performance, the unemployment rate has remained low, largely because many people have given up looking for a job. Some people have returned to school to receive training, while others look for work only occasionally because they are discouraged by the lack of job prospects.

The weak demand for workers coupled with ongoing economic weakness means that losing a job may be relatively rare, but once lost, finding a new job is increasingly difficult. A good indicator of the difficulty many of the unemployed have finding a job is the average number of weeks spent looking for work. As of April 2003, the average spell of unemployment increased to 19.6 weeks. The average unemployment duration has not been that high since January 1984; at that time, the national unemployment rate stood at 8%. Although some analysts claim that the relatively low unemployment rate reflects an “adequate” labor market, in truth, the labor market is considerably worse than adequate for the unemployed.

Characteristics of today’s long-term unemployed

Analysis indicates that the current economic downturn has had a surprisingly severe impact on “better-off” groups.1 Workers normally thought to have the skills and backgrounds necessary to avoid prolonged unemployment during a recession experienced long periods of joblessness; many were unemployed for more than six months. Once mid-career, college-educated, or executive and professional workers lose their jobs, they are more likely than other workers to remain unemployed for a long period of time (See Table 1).

Today’s equal opportunity labor market recession has also substantially affected less-advantaged workers, in particular, black workers and those previously employed in manufacturing. Once unemployed, these workers are more likely than others to remain unemployed for six months or more.

Table 1

The data show that long-term unemployment is more common among better-educated workers: 43.6% of the long-term unemployed in 2002 were educated beyond high school, compared to only 39.6% of the total unemployed in 2002. In other words, better-educated workers are overrepresented among the long-term unemployed as compared to the total unemployed population. A similar pattern is found among occupational backgrounds. In 2002, one in five long-term unemployed workers were from white-collar executive and professional occupations, compared to only one out of six of the total unemployed population. That is, more of the long-term unemployed are from white-collar occupations, as compared to the overall unemployed population.

Lower-wage workers, generally thought of as particularly vulnerable in a recession, are also an important part of the long-term unemployment picture. In terms of race, black workers are more prevalent among the long-term unemployed (making up 25% of this category) than among the total unemployed (20%), reinforcing the notion that lower-skilled black workers are likely to be the “last hired, first fired.” Manufacturing workers, who have experienced great economic dislocation in the past several recessions, have also fallen victim to long-term unemployment. One in five long-term unemployed workers in 2002 were from the manufacturing sector, compared to just one in six in 2000.

Growth patterns and the likelihood of long-term unemployment

The steep rise in the percentage of the unemployed looking for work for more than six months is evidence of the continued overall weakness of the economy. The percentage of the unemployed searching for work for more than six months has increased from 11% in 2000, to 18.3% in 2002, to more than 20% so far in 2003.

Another component of the joblessness picture is the expansion of long-term unemployment during the sustained job decline of the last two years. The data present a compelling picture of the breadth and depth of today’s long-term unemployment compared with pre-recession unemployment levels. Table 2 examines the proportion in each subgroup that has been unemployed for more than six months. From this we calculate the likelihood of long-term unemployment for different groups of workers as the economy moved from full unemployment to a sustained decline across 2000 and 2002.

Table 2

Age: The data show that long-term unemployment has grown equally among workers in their prime earning years (25-44) and among mid-career workers (45+). However, it is these mid-career workers who have suffered the most during the current downturn, with more than a quarter (25.6%) becoming long-term unemployed. Mid-career and older workers often suffer difficulties with long-term joblessness during economic downturns because they must replace higher-than-average wages and overcome age discrimination in the labor market.

Education: Once college graduates lose their jobs, they are more likely than workers at other education levels to become long-term unemployed. Throughout 2002, 22.7% of unemployed college graduates had been out of work for more than six months, markedly higher than the 2002 national average of 18.3%. In contrast, only 17.0% of jobless workers with high school degrees or less were long-term unemployed, and 18.7% of those with some college education endured such long spells—rates similar or less than the national average. Along these lines, the number of long-term unemployed college graduates has tripled since 2000, and the number of long-term unemployed workers with some other education beyond high school has almost tripled. National Employment Law Project (NELP) interviews have revealed the shock that highly educated unemployed workers experience when they find that their skills and education do not ensure an easy transition between jobs. These data underscore the fact that the lack of available jobs—and not the skills of America’s unemployed workers—is responsible for long-term joblessness.

Race: Like the industry data that follow, data on race and long-term unemployment provide evidence for the broad character of the current decline. Black workers are significantly mo
re likely than unemployed workers from other races to be long-term unemployed. Unfortunately, this seems to confirm a familiar pattern for black workers. During economic good times, black workers are able to overcome racial discrimination and labor market disadvantages (education, etc.) but not when jobs are scarce. On the other hand, the ranks of white long-term unemployed have nearly tripled during this labor market downturn, growing faster than long-term unemployment among minority workers. In short, black workers are more likely than white workers to be long-term unemployed, but long-term joblessness among whites is growing at increasing rates.

Gender: Long-term unemployment is affecting male and female workers in a relatively equal fashion during this recession, with men experiencing slightly higher proportions and growth in long-term unemployment. This contrasts with data from the last recession. In 1992, 22% of unemployed male workers were out of work for more than six months, compared to just 17% of jobless women. The trend of men and women more equally bearing the brunt of joblessness has occurred throughout this current slump—including slower-than-average recessionary service sector growth (where women are more likely to work) and continued high levels of female labor force participation (meaning women are more likely to keep looking for work rather than drop out of the labor market when job opportunities shrink).

Occupation: A similar pattern emerges with regard to occupation. More than one in five jobless workers (22.8%) from executive, managerial, and professional occupations qualified as long-term unemployed in 2002. No other major occupational group experienced as great a probability of long-term unemployment; all other nonagricultural occupations have long-term unemployment rates near the national average. White-collar long-term unemployment has more than tripled during this recession, substantiating the popular perception that the current downturn is a more “high-end” recession than those in recent memory.

Industry: The industrial composition of the long-term unemployed demonstrates the wide-ranging nature of the downturn; joblessness clearly reaches more than white-collar and highly educated workers. Nearly one in four (24.1%) jobless manufacturing workers have experienced long spells of unemployment. The signature features of the current downturn—the decline of the technology and stock markets and the 9/11 attack—are also reflected in these data. The sectors of finance, transportation and public utilities, and wholesale trade (intermediate sellers of goods) all experienced greater than 20% long-term unemployment, with the ranks of the long-term unemployed more than tripling in the latter two industries.

Combined statistics from all of the subgroups illustrate who is most likely to experience long-term unemployment during this downturn. Figure A shows the characteristics of the unemployed with the highest probabilities of long-term unemployment. Figure A shows that the increase in the long-term unemployed population is being driven both by workers who have been historically affected by long-term unemployment (manufacturing and older workers) as well as by more-skilled workers who have been caught in this current downturn (college graduates and executive, managerial, and professional workers).

Figure A

State-level analysis

In addition to the national tables, we analyzed state data in seven large states (California, Florida, Illinois, New Jersey, New York, Pennsylvania, Texas). In 2002, these seven states had slightly less than half of all the long-term unemployed workers in the nation. Except for Florida and Texas, all have above-average state unemployment rates.

Some general observations can be made about these seven states. First, workers in these large states are more likely to be long-term unemployed than workers in the rest of the nation. Moreover, these states have all experienced substantial increases in long-term unemployment; New Jersey saw long-term unemployment increase by 27,000 workers between 2000 and 2002, while California saw long-term unemployment increase by 120,000 workers during the same period. To compare these states, we also present the proportion of jobless college graduates and black workers who have been out of work for more than six months. On a national level, these two groups happen to have the same rate of long-term unemployment, and have higher rates of long-term unemployment, than the national average. Thus, these workers provide an easy source of comparison to analyze how more- and less-advantaged workers experience long-term unemployment (despite the fact that many workers fit in both categories).

New York and New Jersey: The two states most severely affected by the terrorism of 9/11 had the highest incidence of long-term unemployment through 2002 of the seven states analyzed (New York had 24.9% long-term unemployment and New Jersey had 21.9%). In New York, out-of-work college graduates and black workers were more likely to be long-term unemployed than other workers in the state, with general long-term unemployment following a broad pattern similar to the rest of the nation. In New Jersey, college graduates appear more vulnerable than black workers to long unemployment spells.

California and Pennsylvania: Workers in both of these high-unemployment states are also more likely to be out of work for long periods after being laid off. In both states, long-term unemployment is affecting broad sectors of the labor market. College graduates and black workers both are more likely than the average worker to go long periods without finding employment.

Illinois: The rate of long-term unemployment among the jobless is above the national average in Illinois. However, college graduates have the same kind of unemployment spells as other workers in the state, with a 19.7% long-term unemployment rate, while more than 25% of black workers experience extended joblessness. In Illinois, long-term unemployment appears to be tilted toward more-disadvantaged workers.

Florida and Texas: Unlike the large northern and western states above, long-term unemployment is less prevalent than the national average in these two southern states. Of the two states, Florida has experienced a greater increase in joblessness since the start of the recession (on par with the national average). In Florida, the increase appears broad based, with both college graduates and black workers experiencing long-term unemployment at rates that are greater than the state average.

Policy recommendations

Long-term unemployment can only be fixed by creating enough jobs for all the workers who want one. Unfortunately, the fiscal crises in the states and current tax cut proposals will provide no short-term stimulus for the economy. While the Federal Reserve Board has acted repeatedly to lower interest rates, fueling a refinance boom and helping to maintain a robust housing market, the economy is running against the limits of monetary policy.

Fiscal policies are needed to stimulate demand for goods and services, stem the loss of jobs, and continue to provide social services for those in need. Any economic stimulus will take time to heal the economy, but in the short run, workers who are caught in this economic malaise should be assured that they can make ends meet. To that end, we recommend temporary policies that will tide people over until the economy recovers.

Extend unemployment insurance
During recessions, unemployment insurance (UI) benefits play a crucial role—they allow workers to avoid poverty, mitigate economic stress, and preserve some savings as well as provide support during long job searches. Workers from all backgrounds benefit from unemployment insurance, as most
workers only have sufficient assets to provide for themselves for 5.4 weeks of unemployment (Gruber 1999). Also, unemployment benefits provide a potent economic stimulus, with UI benefits credited with lessening the severity of unemployment. Recent analysis indicates that UI benefits reduced the severity of recessions an average of 15%; that is, previous recessions would have been considerably deeper and longer without the economic stimulus provided by unemployment insurance benefits (Chimerine et al. 1999).

However, workers normally only qualify for, at most, 26 weeks of regular state unemployment insurance benefits—far less than it takes many of the most-skilled workers to find a job in today’s economy. The extended duration of unemployment is causing a record proportion of jobless workers to exhaust the six months of regular state unemployment.

Following the example of prior recessions, Congress enacted the Temporary Emergency Unemployment Compensation (TEUC) program in March 2002. The program provided up to 13 weeks of extended benefits to workers in all states, and up to 26 weeks to workers living in a limited number of “high-unemployment” states. As of March 31, 2003, the TEUC program has pumped $13 billion dollars into the nation’s economy, and over 4.6 million workers have received TEUC payments. Given the breadth of long-term joblessness, we recommend that Congress and the Bush Administration take action in the following ways:

• Extend the TEUC program: Unless Congress acts, the TEUC program will end at midnight on May 31, 2003. An average of 350,000 workers across the nation have exhausted their regular state benefits each month for the past six months, and the Center on Budget and Policy Priorities estimates that an additional 2.1 million workers will exhaust their regular state benefits without finding work from May through November 2003 (Shapiro and Goldberg 2003). Congress and the Bush Administration should continue the TEUC program for at least six months beyond the May 31 deadline.

• Strengthen the TEUC program: The wide reach of long-term unemployment demonstrates that the job slump has profoundly affected those unlucky workers who have lost their jobs. This persistent unemployment is strong evidence that today’s tough labor market is at least as bad as past recessions. Despite this fact, the extended benefit program has been much weaker than in previous economic downturns. During most of the 1990s recession, almost all workers received a minimum of 20 to 26 weeks of extended benefits. But because benefits have been so limited during this downturn, 2.8 million workers have exhausted all federal and state benefits.

The TEUC program should offer at least 26 weeks of extended benefits to all unemployed workers. With the shortage of jobs across the economy, long-term joblessness is likely to stay at higher-than-average levels. Unemployed workers are experiencing great hardship, and deserve additional support to weather economic forces beyond their control.

The National Employment Law Project is a national advocacy organization for the unemployed and the working poor. NELP can be contacted at 55 John St., 7th Floor, New York, N..Y 10038, (212) 285-3025, or at

1. In this analysis we use data from the 2000 and 2002 Current Population Surveys. These are the same data used to generate the official unemployment figures each month. Our sample includes only unemployed workers as defined by the U.S. Department of Labor.

Chimerine, Lawrence, et al. 1999. “Unemployment insurance as an economic stabilizer: evidence of effectiveness over three decades.” U.S. Department of Labor, Unemployment Insurance Occasional Paper, 99-8.

Gruber, Jonathan. 1999. “The wealth of the unemployed: adequacy and implications for unemployment insurance.” National Bureau of Economic Research, Working Paper No. 7348.

Shapiro, Isaac, and Jessica Goldberg. 2003. “Nearly four million of the long-term unemployed will be directly affected by upcoming legislative decisions.” Center on Budget and Policy Priorities, May 6.

See related work on Income and wages | Wages, Incomes, and Wealth

See more work by Andrew Stettner and Jeffrey Wenger