International Monetary Fund

Please send us your feedback

Share This

Surviving the Third Wave

Finance & Development, December 2009, Volume 46, Number 4

Hyun-Sung Khang

PDF version

After the financial and economic crises, a “third wave” is engulfing the labor market, leaving millions without work and changing the course of their lives

Unemployment rates in OECD countriesPOWER plant operator Zeki Kilic was afraid to tell his wife when his firm announced it was introducing short-time working in response to the global slump. Plummeting demand for steel meant his employer, German industrial conglomerate ThyssenKrupp Steel, could not maintain its workforce at full capacity, but it didn’t want to lay off its highly qualified employees. So it introduced short-time working, or Kurzarbeit, as it is known in German.

“We were in a doomsday mood,” Kilic says. “The world economic crisis was the phrase on everyone’s lips. And you didn’t know where all this was heading.”

Kilic was reluctant to break the news to his wife because he feared short-time working was simply a precursor to an eventual layoff. His worries are understandable. The aftershocks of the global crisis are resonating through workforces around the world. Tens of millions of people have already lost their jobs and millions more will soon be sharing the same fate. While countries are slowly returning to positive growth, for many people, the worst is yet to come.

“Imagine the worker who will lose his job in the months ahead. For that worker the crisis is not behind him, but still ahead,” says the head of the IMF, Dominique Strauss-Kahn. He sees an emerging third wave of the crisis hitting labor after the initial financial markets crash, which swiftly engulfed the wider economy.

“The pace and magnitude of the increase in unemployment in the OECD area are unprecedented in the postwar period, and we need to go back to the economic downturns of the 1970s and early 1980s to find something similar,” says Stefano Scarpetta of the Organization for Economic Cooperation and Development (OECD). The OECD calculates that unemployment will continue to rise until the end of next year. If the forecast is correct, the number of jobless within its 30 member states will be over 20 million higher than at the beginning of the downturn—the worst hike since the end of World War II.

Output will eventually recover, but the risk is that the rise in unemployment will prove enduring. Even if jobs are created eventually, the slump is etching lasting scars on the lives of millions of new graduates unable to secure jobs, young people sentenced to a lifetime of lower wages, and temporary workers whose shaky grip on the world of work has become even more tenuous, alongside the psychological distress of the jobless and their families.

Impact varies geographically

The impact of the global slump on unemployment has varied across regions and countries. It is not restricted to the geographical source of the crisis. While the United States has breached the psychologically significant 10 percent unemployment mark, Japan, which initially viewed the U.S. financial crisis as “fire on the other side of the Pacific,” has also confronted record levels of unemployment during this crisis. Contrast this with the Netherlands, where—at 3.6 percent—unemployment was up less than 1 percentage point over the previous year, according to OECD harmonized rates.

In Africa, the slump in global demand, coupled with a sharp decline in the prices of some major commodities, pushed the continent’s largest economy, South Africa, into its first recession in almost two decades. Its legacy: about a quarter of the population lacks work, according to the country’s statistical agency, Stats SA. The picture for developing countries is more difficult to distinguish given the absence of reliable figures. Where those statistics do exist, they can distort a complicated social picture and hide considerable human distress. In developing countries, workers cannot afford to stay unemployed, and in the face of a crisis people are often forced into less productive and lower-paying jobs.

Duncan Campbell, of the Geneva-based International Labor Organization (ILO), recalls his experience working in Thailand during the 1997–98 Asian crisis when, contrary to expectations, unemployment fell. “Anecdotal evidence tells us that it was effects like the Thai factory worker who lost his job, opened up a street stall, and then drew in his wife to work on the stall, and then also his child, who previously used to be in school. So even where there has been no rise in unemployment, what we do see is a rise in working poverty, income-related underemployment, and an increase in vulnerability.”

Disaffected youth

The impact of unemployment has also been unevenly distributed across sectors and types of workers. History suggests that immigrants and low-skilled, temporary, and older workers are likely to be shown the door first as layoffs are rolled out. One trend causing particular anxiety is the number of young people exiting high school and college into a jobless market. The ILO has warned that youth unemployment worldwide will rise from 12 percent in 2008 to 15 percent in 2009. In Spain, the unemployment rate for teenagers and young adults has hit almost 40 percent.

Unemployment, especially for those just starting out, can scar individuals for years, and possibly all their working lives. “The effects of a period without work do not end with that spell,” David Ellwood—now dean of the Kennedy School of Government at Harvard University—has written (Ellwood, 1982). Ellwood and, later, other economists concluded that early unemployment influenced workers’ prospects and reduced their wages over the span of their working lives: the absence of a work record lowered their likelihood of being hired, and disillusionment and despair impaired their search for work.

Workers in precarious, temporary employment

Joining young people in the job queue are temporary workers, who, with limited access to safety nets, face dire straits without employment. During the boom, many companies in OECD countries took on temporary workers, largely to skirt hiring and firing regulations. The predicament of temporary workers is particularly acute in countries such as Japan, where housing is often part of the employment package. This was vividly illustrated by Yoshinori Sato in the September 2009 issue of Finance & Development. A vehicle assembly worker, Sato found himself out of a job and facing homelessness with four days to vacate his company-owned dormitory room.

According to OECD calculations, some 95 percent of the 158,000 layoffs in Japan since October 2008 involved nonregular workers (part-time, temporary, and contract workers; those on loan from other companies; and workers hired out by agencies), who usually do not qualify for company-paid severance or unemployment insurance. The precarious existence of the Japanese temporary worker is mirrored in the lives of his French and Finnish counterparts, with governments in all three countries trying to ease the vulnerability of their nonregular working population.

The crisis increasingly affects women

In the early days of the crisis, high numbers of layoffs in construction and manufacturing meant men were initially hit harder than women. But as developed economies continue to bleed jobs, women are increasingly joining their male counterparts in the unemployment line. In the United States, for example, service industries—a sector where women are heavily represented—now account for half the overall decline in employment.

While more women will soon find themselves forced out of work in the richer economies, paradoxically, in poor countries, more women may have to return to the world of work, distorting the historical pattern of female participation in the labor market. In the early stages of development, women’s participation in the labor force is high, but as a country moves along the development path, their participation drops as production moves from the household, family farm, and small business to the wider market. Women begin to reenter the work force in large numbers as they become more educated and the value of their human capital rises. For this reason, female labor force participation is often described as U-shaped. “In recessions, the shape of the ‘U’ gets distorted,” says Campbell. “It probably gets flatter.”

Jobless recovery?

Even though unemployment is a lagging indicator, the fear is that growth and employment are decoupling—that along with subdued growth and costlier credit, high structural unemployment (or at least higher than before the crisis) could now be part of a “new normal.”

Historically, as an economy emerged from recession, growth was accompanied by a rise in employment. However, as Andolfatto and MacDonald (2004) point out, after the two most recent recessions in the United States, employment growth lagged the recovery in gross domestic product by several quarters—a phenomenon known as “jobless recovery.” It is a description economists such as William Darity of Duke University—who has studied the psychological impact of unemployment—disdain as a contradiction in terms. He believes any meaningful recovery must involve job creation.

Explanations for this recovery-minus-job-creation phenomenon are varied. After a recession, the natural rate of unemployment may simply end up higher. Some economists argue that it results from the churning effect of workers reallocating their skills from a declining to a burgeoning sector, or perhaps sustained unemployment rates reflect efficiency gains from technology.

Robert Gordon—of the National Bureau of Economic Research, which is responsible for pronouncing the official end to a U.S. recession—argues that jobless recoveries result from the tendency of firms to overhire in the late stages of business expansion. This buildup of labor kicks in following the end of a slump, and this “end-of-expansion effect” can lead to an increase in output without a corresponding increase in labor (Gordon, 1993).

In later writings, Gordon added a complementary hypothesis—the “early recovery productivity bubble”—which he describes as the corollary of the jobless recovery. “In the first few quarters of the recovery profits are still squeezed, and business firms are aggressively attempting to cut costs by reducing labor input” (Gordon, 2003)—hence the rise in output with fewer workers. Given his belief that the U.S. recession ended in June of this year, he has no doubt the developed world will see—indeed may already be in the throes of—a jobless recovery.

Continued high unemployment


Zeki Kilic on the job at
ThyssenKrupp Steel.

Gordon believes positive employment growth will resume at the beginning of 2010. Others are more pessimistic, believing that many economies are confronting the specter of a structurally higher rate of unemployment. The idea of a permanent effect of a transitory shock, such as a recession, is encapsulated in the concept of hysteresis. Traditionally, economists believed that high unemployment was a cyclical phenomenon—unemployment would cause people to lower their wage demands, so new jobs would be created and unemployment would fall. But in a paper written more than 20 years ago, Olivier Blanchard and Lawrence Summers (Blanchard and Summers, 1986) asserted that unemployment rates have a ratcheting effect.

Whether hysteresis occurs in unemployment rates has been debated by economists, but in the current climate it is experiencing something of a return to fashion. Blanchard, now chief economist at the IMF, says he is still undecided about the precise channels for hysteresis. During the intervening two decades, he has variously explored the possibility of an insider-outsider theory of wage bargaining (where remaining employed workers increase their wage targets, preventing the unemployed from getting their jobs back); the changing behavior of the unemployed (who become unemployable); and, most recently, unemployment protection measures introduced by governments at the height of a crisis. These, Blanchard suggests, can discourage a return to work. “The reasons for unemployment go away, but the institutional structures remain the same. When you have done this, you have screwed up the labor market.”

The creator of the concept of hysteresis in employment rates is skeptical about its role in this crisis. (Blanchard finds a more plausible explanation in the modest scale of the recovery: “Output growth is going to be low, productivity growth is going to be normal, and employment growth is going to be very small,” he says.) But others—such as Laurence Ball of Johns Hopkins University—seize on hysteresis as a possibly critical explanation for continued higher unemployment rates into the future. “With interest rates near zero in the United States, there is no room to cut in this crisis, and that, in my view, is why we are going to get hysteresis effects,” says Ball.

Impact of long-term unemployment

Long-term unemployment results in serious problems for the jobless as well as for the overall economy (though a few economists have investigated the positive externalities to downturns). The problems of long-term unemployment are intuitive. Workers, unemployed for an extended period, lose their skills and are less able to keep up with current work practices; their ties to the world of work weaken; and even if they do return to work, the formerly jobless can end up less efficient and proficient than they used to be.

The endpoint for a discouraged job seeker is complete withdrawal from the labor market. Estimates from the Current Population Survey suggest that almost half the unemployment episodes completed in 2003 ended with the individual leaving the labor force (Ilg, 2005). The OECD assumes that two out of every three workers in mainland Europe who remain jobless for more than a year will not resume work thereafter.

Governments respond

The social and financial cost of long-term unemployment—not only to the individual, but also to the state—has led many governments to channel a sizable proportion of stimulus funds into programs to get the unemployed back to work. Depending on the fiscal situation, political inclination, and cultural preferences, these measures range from the creation of public sector jobs and retraining programs to tax incentives to encourage hiring.

In India, for example, the government has used double its intended budget on the “National Rural Employment Guarantee Act.” This offers 100 days of work a year to any adult member of a rural household—man or woman—willing to do manual labor on a public works project for the minimum wage. In Mexico, President Felipe Calderón announced reforms to ease red tape and lower costs for investors in public works projects to foster job growth. And in the United States, the Obama administration is considering a tax credit for new hires next year.

In recent years, many governments have favored supply-side measures—including reducing the power of unions, cutting red tape, and the introduction of expensive training programs—to boost employment. This followed the bitter experience of the 1970s, when many countries tried to overcome the effects of the oil shock by promoting demand-side policies, but in the process ended up stoking inflation. Now, record-low interest rates and fiscal deficits will limit some governments’ room to maneuver. But even measures to promote employment have a mixed record, and governments are often motivated more by political pressure to cushion their workers from the worst of the economic crisis than by efficient outcomes.

A shortcut to shorter unemployment lines?

Intervening while at-risk workers are still employed can ultimately prove less expensive than waiting for them to lose their jobs and funding unemployment benefits or costly (and arguably ineffective) job training programs. Short-time working, for example, one of the most popular job-protection measures in the euro area, is praised by its supporters as an appropriate response to this crisis, in which many firms face the combined pressure of a severe short-term contraction in demand and an inability to access credit. According to the OECD, short-time working programs have been adopted in various forms by 22 of its 30 member countries. The German version, Kurzarbeit, is designed to distribute the economic pain between employee, employer, and the government.

This was the system adopted by Zeki Kilic’s company, ThyssenKrupp Steel. Workers work fewer hours, and their salaries are paid by the government and the employer. A peculiarity of the system is that the hourly wages of workers can jump, often by a substantial amount (a source of finger-pointing by critics). ThyssenKrupp attempted to keep wage losses for all employees to a maximum of 10 percent of their salary. In Kilic’s case, when Kurzarbeit was in full force, he was working three-quarter time, but saw only about a 10 percent reduction in his wages each month.

Designed to bridge the gap between full-time employment and unemployment, short-time working recognizes that it often makes economic sense to keep workers in anticipation of a rebound rather than pay the costs of hiring and firing. “Without Kurzarbeit, there would certainly have been massive layoffs,” said a spokesman for ThyssenKrupp Steel. “If you lay off people during a crisis, you lack a specialized workforce when orders rise again. We wanted to avoid this.”

But critics say Kurzarbeit is an expensive option that often fails to deliver the most efficient outcome. Moreover, the experience with short-time working subsidies has not been unconditionally encouraging. Compensation often goes toward retraining workers whose employers would have retrained them anyway or ends up supporting firms that prove nonviable, even when business conditions improve. “It can be a very costly measure by the government, and there is at least a danger of delaying necessary structural change,” says Martin Schindler of the IMF.

Contrasting patterns of unemployment are informing the debate about how best to create and protect jobs. European joblessness has risen less than in the United States, though it was higher to begin with. At 9.7 percent (latest available figures), unemployment in the euro area was just 2.5 percentage points higher than at its lowest point in the cycle, 21 months ago. Over the same period, unemployment in the United States surged by almost 5 percentage points. The European model, previously criticized for its supposedly sclerotic worker retention policies and excessive job protection—as opposed to apparently more nimble and mobile, hire-and-fire U.S. working practices—is now viewed with renewed confidence.

The argument about the respective virtues of each economic model has yet to be played out. German unemployment, for example, may yet see a sharp upswing as short-term working programs run their course, and Kurzarbeit may be vindicated—or not—by the shape of the cycle. “Eventually, fiscal support for such schemes will run out, and the speed of the recovery will be among the factors that determine whether they have really helped to avoid unemployment or merely delayed it,” says Schindler.

Kilic, however, has no doubt that short-term working saved his job. Although the ThyssenKrupp steel plant is not yet running at full capacity, he and his colleagues returned to full-time work in August after six months on short time.

Kurzarbeit did help, clearly,” he said. “We are out of the woods now. It could have been much worse.”


Andolfatto, David, and Glenn MacDonald, 2004, “Jobless Recoveries,” EconWPA Macroeconomics Paper 0412014.

Blanchard, Olivier J., and Lawrence H. Summers, 1986, Hysteresis and the European Unemployment Problem, NBER Macroeconomics Annual (Cambridge, Massachusetts: MIT Press), pp. 15–77.

Ellwood, David, 1982, “Teenage Unemployment: Permanent Scars or Temporary Blemishes?” in The Youth Labor Market: Its Nature, Causes, and Consequences, ed. by Richard B. Freeman and David A. Wise (Chicago: University of Chicago Press), pp. 349–85.

Gordon, Robert J., 1993, “The Jobless Recovery: Does It Signal a New Era of Productivity-Led Growth?” Brookings Papers on Economic Activity, Vol. 24, No. 1, pp. 271–316.

———, 2003, “Exploding Productivity Growth: Context, Causes, and Implications,” Brookings Papers on Economic Activity, Vol. 34, No. 2, pp. 207–79.

Ilg, Randy, 2005, “Analyzing CPS Data Using Gross Flows,” Monthly Labor Review, Vol. 128, No. 9, pp. 10–18.

Hyun-Sung Khang is a Senior Editor on the staff of Finance & Development.


Finance & Development Logo

Letters to the Editor

Francisco Vergara: Will American Unemployment Ever Start Falling?

Write to us

F&D welcomes comments and brief letters, a selection of which are posted under Letters to the Editor. Letters may be edited. Please send your letters to

Free Email Notification

Receive emails when we post new items of interest to you.
Subscribe or Modify your profile