May 6, 2012
Why Unemployment Remains So High
Why Unemployment Remains So High—Posner
In 2007 and 2008, the U.S. unemployment rate was only 4.6 percent, though it started to rise after the financial crisis hit in September 2008, reaching 10.1 percent in October 2009. Since then it has fallen to 8.1 percent. The unemployment figure is not terribly illuminating, as it defines the unemployed as persons who are looking for a job, thus excluding those who have stopped looking because they don't think they'll find one; in the latest quarter, the unemployment rate fell (from 8.2 percent in the previous quarter) only because of the large number of persons who dropped out of the labor force, thus reducing the denominator in the fraction of unemployed. The 8.1 percent figure also excludes the underemployed. A further concern is that more than 40 percent of the unemployed have not been working for more than six months, and it hard to land a job after having been out of the workforce so long; work skills and attitudes tend to atrophy, making employers reluctant to hire the long-term unemployed.
The current depression (calling it a mere "recession" that ended in June 2009 when GDP stopped falling strikes me as ridiculous) began almost four years ago. That employment has not recovered is not surprising. When the financial crisis struck in September 2008 the personal savings rate of Americans was only 1 percent, the reason being that the housing bubble had inflated the apparent savings of homeowners by increasing their home equity relative to their debt. The rapid plunge in the market value of people's personal savings caused them to reduce their consumption spending in an effort to rebuild their savings. At the same time, the uncertainty of the economic environment and the weakness of bank balance sheets caused banks to restrict lending (other than to the federal government and a handful of other reliable borrowers), so credit for consumer purchases became harder to get, and this further reduced consumption spending. The Federal Reserve flooded the banks with money, but the banks hoarded the money. With credit scarce and consumer spending down, companies reduced production and so laid off workers, which further reduced consumer spending, both directly by reducing incomes and indirectly by increasing uncertainty about the economic future.
When the downward spiral stopped, and consumer spending revived, companies were reluctant to hire back the laid-off workers because of continued uncertainty about economic conditions arising in part from the deepending economic crisis in Europe and slowing economic growth in countries like China, India, and Brazil.
I do think the government could have broken the downward spiral earlier, and had it done so unemployment would be somewhat lower today, though I doubt it would be dramatically lower. The $800 billion stimulus (Keynesian deficit spending as a depression remedy) should have been larger, should have been introduced earlier (in the fall of 2008 rather than in February 2009), should have been designed differently, with a focus on financing projects requiring labor rather than on transfer payments (since much of the transferred money could be expected to be saved rather than spent—and in a depression one wants spending not saving), and should have been executed with a greater sense of urgency, as in 1933, when literally millions of the unemployed were put to work within months and the economy began a rapid recovery, though the recovery stalled in 1937 when the government increased taxes and reduced the money supply.
In addition to restoring jobs during a depression, government can preserve jobs, as it did with the financial aid to the Detroit automakers, which saved hundreds of thousands of jobs at a particularly critical time (the first half of 2009). It now appears that the government should also have given financial aid to the states (which cannot create money to pay their obligations and thus can engage in deficit spending only with the consent of their creditors) to enable them to avoid having to lay off public workers, such as teachers, police, and firefighters—the states have laid off almost 600,000 of their employees during this depression.
I don't think there's much else the government could have done to arrest the depression. Attempts at fundamental economic reform should not be made during a depression, because they create uncertainty, which increases the incentive of businesses and consumers alike to hoard rather than spend; hence the Administration's health-care reform was mistimed. The extension of unemployment benefits, along with more generous means-tested benefits programs like stamps, may also have been a mistake, by reducing economic pressures on the unemployed to find work. On the other hand, these transfer payments, although an inefficient method of promoting employment (because as I said transfer payments may to a significant extent be saved rather than spent), do increase incomes; and some—probably most—of the increased income is spent, in turn increasing demand for goods and services and so production and employment. Obviously, however, indefinite extension of unemployment benefits cannot be the solution to the unemployment problem.
A depressing possibility is that the depression has accelerated what may be a long-term trend to reduced employment in many industries. Until quite recently, the depression had fostered significant productivity gains, as employers tried to get more work out of fewer workers, as by speeding automation and adopting more efficient personnel management policies. By "accelerating" I mean adopting these measures earlier than they would have done without the spur of falling demand. So the depression may have brought forward inevitable changes in production that will result in reduced employment and earnings. Eventually the displaced workers will find jobs in industries not experiencing the same productivity advances, but the transitional period of high unemployment may be protracted.
The fact that the unemployment rate was less than 5 percent before the depression may refute the suggestion of a long-term decline in employment. But remember that the unemployment rate can be misleading. If we look instead at the percentage of the working-age population that is employed, it has declined from its peak of 67.1 percent in 2000 to 63.6 percent this year—and the decline began before the onset of the depression.
Why has the Recovery in Employment in the US been so Slow? Becker
The jobs report from The Bureau of Labor Statistics this past Friday is not good reading. The economy added about 115,000 workers, the slowest increase in 6 months. To make matters worse, over 40% of the unemployed have remained out of work for at least six months. The unemployment rate did drop a notch, but this was because many discouraged workers left the labor force. In fact, the recovery is the slowest in the post World War II period. No single factor explains this slowness, but a combination of several explains most of the slow recovery.
Recoveries after major financial crises are notoriously slow. This is well documented in the book This Time is Different: Eight Centuries of Financial Folly, by Carmen Reinhart and Kenneth Rogoff. The authors study many financial crises, and the recoveries from these crises. Recoveries are slow partly because the dire nature of a financial situation is not recognized quickly, and policies that try to end a crisis are usually implemented slowly.
While slow recoveries from major financial crises are common, employment would have increased considerably more rapidly, and unemployment would have fallen much faster, were it not for several factors special to this recovery. Scott Baker, Nicholas Bloom and Steven Davis have studied changes in economic policy uncertainty since 1985, and have constructed an index of the degree of economic policy uncertainty during the past 26 years (see their "Measuring Economic Policy Uncertainty", October 2011). The index spikes sharply after major events, such as Black Monday's stock market fall in 1987, the 9/11 attack, and the beginning of the 2nd Gulf War. Economic policy uncertainty according to their index also rose to extremely high levels during the past several years, especially during the 2010 midterm election, the debate over the stimulus package in 2009, the Lehman bankruptcy and TARP legislation in 2008-9, and the Eurozone crisis and the US debt ceiling disputes in 2011.
These shocks to the degree of uncertainty about the economy and economic policy significantly affected hiring and investments by American businesses. The authors estimate sizable negative effects on employment for up to two years after an increase in the degree of economic policy uncertainty equal to the actual change between 2006 and 2011. Greater uncertainty encourages firms to delay investments in business capital and in new hires. Households faced with a more uncertain economic environment postpone purchasing consumer durables, not only housing but also cars and appliances. Companies and households delay these investments partly because they want to wait to see how the uncertainty is resolved before undertaking potential risky investments. They also wait because greater economic uncertainty raises the cost of capital.
Heightened uncertainty is inevitable during recessions, especially serious ones. However, some of the uncertainty during this financial crisis was avoidable if Congress and the president had not passed an ineffective stimulus package over a divided Congress, if they had resolved the budget deficit and debt ceiling issues (especially by trying to get entitlements under control), if agreement on tax policy toward broader and flatter taxes had been achieved, and if clearer policies were adopted about which companies would be allowed to go bankrupt and which would be bailed out.
Another force slowing down the recovery in employment was the result of more generous means-tested policies introduced during the recession. My colleague Casey Mulligan discusses these policies and their employment effects in "Do Welfare Policies Matter for Labor Market Aggregates?" (January 2012). He shows that they raised unemployment and reduced employment because they gave perverse incentives on the supply side of the labor market. Many others have also argued that the extension of unemployment compensation to 99 weeks clearly encouraged some of the unemployed to remain unemployed much longer than they would have if unemployment compensation ceased after a year.
Mulligan measures in addition the impact on employment and unemployment of several other policies adopted during the recession. For example, the food stamp program SNAP) was made more generous and more easily available, which enabled additional families to qualify for food stamps by reducing their earnings through remaining unemployed, withdrawing from the labor force, or working part time rather than full time. Policies were introduced to reduce mortgage debt of lower income families. Mulligan's calculations suggest that the aggregate effect of all these policy changes on both US employment and unemployment was large because they reduced the supply of men and women who wanted to work, or to work full time.
The literature on financial crises shows that the employment recovery from this recession would likely have been rather slow even with clear and productive government responses. Unfortunately, such responses were not forthcoming. Instead, many of the policies discussed or implemented discouraged both hiring by companies and job seeking by members of the potential labor force through raising the uncertainty about where the American economy was headed. They also reduced employment by creating greater incentives to remain unemployed, or to stay out of the labor force.