November 13, 2011
Structural Unemployment:
Structural Unemployment: Posner
The term "structural employment" means high unemployment that persists through the business cycle, rather than being high only during economic downturns, and so is likely to reflect features of the structure of the economy. Structural economy was a characteristic of the post-World War I British economy. Keynes explained it as the consequence of an overvalued pound, which by making British exports very expensive reduced exports and hence employment in production for export, and by making imports cheap reduced the consumption of domestic products and hence the producers' demand for workers. Demand for labor was so weak that the going wage for many workers was lower than their unemployment benefits. There were also impediments to relocating from geographical areas of high to low unemployment within England; Becker points out that the depression in our housing market complicates relocation to low-unemployment areas in this country.
Unemployment in the United States rose rapidly beginning in the fall of 2008, and it remains abnormally high three years later. Since the economy is still in the doldrums, it is hard to tell whether the abnormal unemployment rate is structural—in which event it may be the "new normal"—or whether it is cyclical. Another complicating factor, emphasized by Becker, is the extension of unemployment benefits to almost two years. Unemployed persons often wait until their benefits are about to expire before they undertake a serious search for a new job, though my guess is that the longer the benefits period, the less the delay in job search; the worker worries about the erosion of his job skills and becomes financially pressed because unemployment benefits are lower than wages.
Although it is premature to say that we have a problem of structural unemployment, it may also be premature to say that we do not. The financial crash and ensuing global economic crisis was a business-cycle phenomenon, but a crisis of such magnitude can bring about or be correlated with or highlight a structural change in the economy. The most inclusive measure of the U.S. unemployment rate—what is called U-6 and includes discouraged and involuntarily underemployed workers as well as unemployed ones looking for jobs (unemployment in the narrow sense)—has risen from a shade above 7 percent in 2000 to a shade above 16 percent at present. True, it fell sharply between 2004 and 2007, but those were years in which the economy was artificially pumped up by massive cheap borrowing interacting with deep tax cuts. Without the tax cuts and the cheap borrowing, and without a recession (more realistically, a depression), there might have been steadily increasing rather than irregularly increasing unemployment over the last eleven years.
The economic crisis is related to, and may have accelerated, trends that could create structural unemployment. One trend is increased competitiveness of foreign producers, resulting in Americans' substitution of cheap imports for domestic products (and hence domestic production and therefore domestic employment), and in a reduction in exports. This is the same combination that caused or at least contributed to the structural unemployment in the U.K. that I mentioned. The severe ongoing economic crisis in Europe has also reduced demand for U.S. exports. And it is difficult for the U.S. to devalue its way to increased exports and reduced imports because of the role of the U.S. dollar as the principal international reserve currency.
A related trend, also adverse to U.S. employment, is the decline in educational performance of American students relative to students in other countries. Furthermore, quite apart from the minimum wage and lengthy and generous unemployment benefits, American workers are expensive to employers because of workers' legal rights and the extensive regulation of workplace safety. The costliness of these rights and regulations leads American companies to relocate as much of their production abroad as they can and accelerates the substitution of capital for labor inputs into production—and much of that capital equipment is produced with few workers.
Persistent unemployment can feed on itself, because the unemployed have lower incomes and so spend less on consumption (and consumption drives production and therefore employment), and because the long-term unemployed lose skills. And if the pattern of employment shifts, many workers discover that they have not been trained for the types of work in which there are jobs.
If all public benefits for the poor were abolished, along with unions, unemployment benefits, the minimum wage, and regulations of workplace conditions, the going wage would plummet and unemployment would fall. But as such draconian measures are not in the cards, we have to worry about the possibility of structural employment and think of civilized ways of heading it off.
Has Structural Unemployment Become Important in the United States? Becker
The persistently high unemployment rate in the United States during the Great Recession has led to claims that much of American unemployment is "structural". According to this view, the demand for workers by companies is insufficient to employ all unemployed workers because there is a mismatch between the skills possessed by many American workers and the skills required by companies. The structural advocates believe the skills demanded by companies tend to exceed or otherwise be different from the skills possessed by many unemployed workers. As a result, so goes the argument, these unemployed workers cannot find jobs and remain unemployed for a long time.
Although I will argue that not much of American unemployment is "structural" or due to such a mismatch, the structural theory is on the surface supported by the large number of long-term unemployed, the most disturbing feature of American unemployment during the Great Recession. Structural advocates claim that unemployed individuals with skills that are only weakly demanded face prospects of remaining unemployed for a long time. Since the unemployment rate rose above 9% in 2009, the fraction of the unemployed who have been out of work for over 6 months has grown to over 40%. Prior to the start of the recession in 2008, long-term unemployed were a little under 20% of total unemployment. Although long-term unemployment usually rises during prolonged recessions, the magnitude of the rise during the current recession is unusual for the United States.
While long-term unemployment in the American labor market jumped up during this recession to unusual heights, there is no evidence of any large mismatch in US labor markets prior to the recession. In 2007, for example, the total unemployment rate was still under 5%, and less than 20% of the unemployed were out of work for six months or more. It is not credible to believe that the underlying structure of labor demand in the US has shifted so much in the few years since the recession began that almost 4% of workers (0.4x9%) will not have employable skills once the American economy gets out of its doldrums, and begins to grow at its "normal" long-term rate of about 2% per capita per year.
The JOLTS monthly data prepared by the federal government on new hires and job vacancies in the United States show over 3 million job vacancies in recent months, and over 2 million openings even during months at the height of the recession. Given this large number of job openings, I conclude that the millions of individuals who have been out of work for 6 months or more could find jobs if they are willing to be flexible on wages and other conditions of employment. Put a little differently, millions of job openings each month in a labor market with considerable flexibility in wages, which describes the American labor market, makes it really hard to argue that many of the long-term unemployed cannot find jobs.
How then can we explain the high and persistent levels of long-term unemployment during this recession? One clearly important factor is the extension in July of 2010 of unemployment insurance to cover workers who have been out of work for 99 weeks instead of having their payments end after 6 months of unemployment. Studies have shown that many unemployed find jobs just about when their unemployment payments expire, after 6 months of collecting benefits under the old system. It does not require any complicated economic analysis to conclude that the extension of unemployment payments to 99 weeks will encourage many individuals to remain unemployed after 6 months of unemployment, although they would have found jobs under the old rules. They will turn down jobs that may not be as good as those they had before becoming unemployed in order to continue to collect unemployment benefits. For that reason I opposed the extension to 99 weeks in my post on 7/25/2010 ("Should Unemployment Compensation be Extended?").
Long-term unemployment may also partly be the result of uneven employment opportunities and uneven prospects for housing prices in different states and regions of the US. States like California, Nevada, and Florida that had major construction, jobs, and housing booms in the years before the onset of the recession suffered the largest hits to employment and to housing prices. Unemployed workers in such states may be reluctant to move to states where more jobs are available at good wages since they may be involved in foreclosure litigation and other issues related to the depressed housing market. They may also turn down jobs that pay much less then the unusually good wages they had during the boom times to continue to collect unemployment benefits for up to 2 years.
Other factors are also involved in the sharp rise in the number of workers unemployed for over 6 months. Yet overall they do not justify the conclusion that many unemployed individuals are unable to find jobs because the American economy has too little demand for their set of skills.