Unemployment: How Bad Can It Get?
The downturn in the labor market is fast becoming the worst since the 1930s. Of course, it's a long way from March's 8.5% unemployment rate to the 25% peak during that era, but for the post-World War II period, the speed of deterioration has been shocking. Of the 5.1 million jobs lost since December 2007, 72% have disappeared in only the past six months, with more losses to come. The jobless rate, already up 4.1 percentage points from its low point two years ago, is well on its way to surpassing the trough-to-peak rise in the 1948-49 recession of 4.5 points, a postwar record. The question on everyone's mind is: How high will the unemployment rate go?
Current trends are not encouraging. Two weeks after the Labor Dept.'s mid-month survey of the job markets for its March employment report, weekly jobless claims climbed higher, suggesting another dismal report for April. The economy has to grow about 2.5% to generate the 100,000 or so jobs per month that are necessary just to stop the jobless rate from rising. Much faster growth would be needed to bring it back down to the 5% level, generally accepted as full employment.
It's normal to see the largest increases in the jobless rate during the deepest declines in real gross domestic product. If economists are right and GDP losses diminish this spring, then monthly increases in the unemployment rate should become smaller. So should payroll declines. Economists at JPMorgan Chase (JPM) show that job losses tend to be largest in the quarter following the biggest quarterly contraction in GDP. By that pattern, if the 6.3% drop in fourth-quarter GDP is the worst, as now expected, then the first quarter's payroll losses, averaging 685,000 per month, would be the recession's biggest, and the second-quarter declines would be smaller.
Still, joblessness seems certain to breach 9% by midyear, and 10% is increasingly possible later on. The question about peak unemployment is important not just because of its implications for consumer confidence, incomes, and spending. A 10% jobless rate is the worst-case scenario for Washington's current round of stress tests for banks. As more people lose their jobs, loan defaults rise, forcing more writedowns and placing even more pressure on bank balance sheets.
More important, as the rate nears 10%, growing slack in the economy raises the risk of deflation. Amid persistently weak demand, prices and wages could begin to spiral downward. The climb in unemployment has already caused hourly pay to slow to an annualized growth rate of 2.2% in the past three months, about half the 4.2% pace in the final three months of 2008.
Unemployment's peak will depend mainly on the depth of the recession. A classic relationship between GDP growth and changes in the jobless rate implies that a peak-to-trough drop in GDP of 3.5%, which is the consensus forecast, would yield an unemployment rate of about 8.75% by the end of 2009. But because the rate has already risen more than this model predicts, economists at UBS (UBS) think the GDP data may greatly understate the economy's weakness. They expect economic growth in recent quarters to be revised down when the Commerce Dept. issues its revisions this summer.
Even if economic growth turns positive in the second half, as generally expected, the upturn may not be strong enough to keep the jobless rate from rising into 2010. In the past, deep recessions have been followed by strong recoveries, allowing unemployment to retreat quickly. That seems unlikely this time, because of the long healing time the credit system and household finances are likely to require.
Much will depend on consumers and the help they get from Washington's stimulus programs. Some signs are encouraging. Consumer spending and housing both appeared to be stabilizing last quarter. That would not be unusual. In past recessions, household outlays have often picked up before employment has. That said, the jobless rate's steep climb is one factor that could derail this hopeful trend.