English > Finance > Finance-Featurestory>'Jobless' Recovery Invalidates Okun's Law

'Jobless' Recovery Invalidates Okun's Law

08-04 11:35 Caijing

Recession-related U.S. GDP and unemployment rates were somewhat predictable before the latest downturn changed the game.

By J. Bradford DeLong

(Caijing Magazine) The United States has or will soon see the bottom of the business cycle, at least as far as the semi-official National Bureau of Economic Research (NBER) is concerned. But the recovery that has or is about to begin won't feel like one. From the perspective of average American workers -- and from the perspective of its impact on American politics -- the economic downturn is more likely than not to drag on for quite a while.

The financial crisis gathered force from summer 2007 through summer 2008. It then exploded with a macroeconomic catastrophe after the collapse of Lehman Brothers last fall. It did more damage to the American economy than most forecasters had dared to imagine.

Conventional wisdom and consensus forecasts by economists last December concluded that the official, headline U.S. unemployment rate in 2009, as reported by the government's Bureau of Labor Statistics, would average something like 7.8 percent. Yet, as of late July, average 2009 unemployment seemed more likely to settle above 9.3 percent. Real GDP in America for 2009 also looked to be lower than predicted. It was likely to come in at US$ 11.4 trillion rather than the US$ 11.53 trillion forecast by the incoming Obama administration just last December.
It's worth pausing to note these magnitudes. The unemployment rate will be 1.5 percentage points or more above, and real GDP will be 1.2 percent below, what was forecast in December 2008.

Back in the 1960s, one of then-president Lyndon Johnson's economic advisers, economist Arthur Okun from the Brookings Institution, set a rule of thumb that quickly became known as "Okun's Law." Okun found that if the workings of the industrial business cycle were causing real GDP in the United States to rise or fall by a bit over 2 percent, one could expect the American unemployment rate to fall or rise by 1 percent. The magnitude of employment swings was always roughly a little less than half the proportional magnitude of swings in real GDP. This was thought to be the case for four reasons.

First, American businesses had a tendency to "hoard" labor in business cycle recessions. They sought to keep useful workers on the payroll, even when there was temporarily nothing for them to do on the factory floor.

Second, American businesses cut back work hours whenever unemployment rose, reducing output more than proportionate to numbers of workers because a decline in total hours would exceed employee totals.

Third, plants and equipment at American factories ran less efficiently when work hours were artificially shortened. And fourth, some workers who lost jobs did not show up in unemployment statistics because, instead of reporting themselves unemployed, they chose to retire or leave the labor force altogether.

For all these reasons, rising unemployment during America's business cycle recessions used to be a fraction of the declines in real GDP relative to trend.

According to Okun's Law, the unexpected extra 1.2 percent decline in real GDP in 2009 should have been accompanied by a 0.5 or 0.6 percentage-point rise in the unemployment rate. Instead, the American economy looks likely to experience a 1.5 percentage point increase in unemployment.

I confess this comes as a surprise. But it should not. Evidence that Okun's Law has been broken -- especially with regard to worker retention during a downturn -- has been mounting for decades.

In 1993 -- two full years after NBER said the 1990-'91 recession had ended -- the unemployment rate was still higher and the employment-to-population ratio lower than it had been at the recession's trough.

In 2005 – long after the 2001 recession had ended – the American economy was still seeing this kind of "jobless recovery." It was not until 55 months after the 2001 recession trough tied to America's post-dot-com-bubble downturn that a greater share of Americans were working than had been working when real GDP was at its lowest.
Now, the American economy is starting once again to show signs that its economic recovery will be "jobless." Real GDP will grow, but the unemployment rate will increase, the employment-to-population ratio will fall, and real wages for average, poor and above-average -- but not rich -- workers will stagnate. So get ready for another jobless recovery.

Why the shift? What's been different about the American economy over the past two decades?

Economist Paul Krugman has a theory. He says past recessions "were very different.... Each of the slumps -- 1969-'70, 1973-'75, and the double-dip slump from 1979-'82 – was caused, basically, by high interest rates imposed by the Fed to control inflation. In each case housing tanked, then bounced back when interest rates were allowed to fall again. Since the mid-1980s, however... recessions haven't been deliberately engineered by the Fed; they just happen when credit bubbles or other things get out of hand."

These recessions have "proved hard to end," Krugman says, "precisely because housing -- which is the main thing that responds to monetary policy -- has to rise above normal levels rather than recover from an interest-imposed slump."

I would guess -- but it is only a guess -- there is a second factor at work as well in America. Manufacturing firms used to think their most important asset was a skilled, experienced workforce. Especially when an economic recovery began, firms did not want to lose hold of this prime, productive asset. Skilled workers were the franchise.

Now, by contrast, it looks as though American firms think their workers are much more disposable, and that their brands or machines or procedures and organizations are the key assets they must safeguard. They do not need to keep current workers to maintain productivity.

American firms generally, however, still want to keep their workers happy. The hypothesis is that firms think workers left on the payroll will not be upset if they fire other employees and speed the pace of operations near the bottom of a downturn. Firms will plead that labor force reductions and work rule changes are forced upon them by economic necessity, and workers will accept that plea at the bottom of a recession.

But they won't accept it at other times. Hence, the start of the recovery is an American business' last chance to trim its labor force and become more efficient and, thus, more profitable in the coming business-cycle expansion. At least that's the theory and my guess.

America is indeed likely to see an economic recovery. The prevailing forecast right now is for real GDP to contract at a rate of 1 percent per year or less between the first and second quarters 2009, followed by growth between the second and third quarters at an annual rate of 2 percent or so. Eventually, the NBER Business Cycle Dating Committee is most likely to say the recession ended in June or, as a second most likely alternative, April. A turnaround date after June 2009 is a less likely possibility.

Yes, that would mean the recession is already over. One reason is the federal government's much-maligned stimulus package, which probably boosted the real GDP annual growth rate by about 1 percentage point in the second quarter 2009, and will boost it another 2 percentage points between July and summer 2010.

This matters because it means American demand for imports is likely to start growing again – from now. But it also matters politically. The question, "Did the Democratic Party's stimulus package work?" could be answered "yes" if observers focus on real GDP and NBER business cycle turning-point dates. Democratic members of Congress seeking re-election in 2010 will be able to point to real GDP growth and an official end to the recession in the second quarter 2009.

However, that is probably not the most relevant political question, nor is this how the issue is likely to be framed. Between this year's first and second quarters, work hours in America declined at a 6 percent annual rate. Work hours from the second to third quarter are likely to decline at an annual rate of about 3 percent. These numbers imply that American productivity is now growing at 5 percent per year not because of investments in new technology but rather because businesses are continuing to fire workers, believing that remaining workers will accept their pleas of economic necessity. If the question in the 2010 and '12 elections becomes "Did the Democratic Party make it easier to find jobs or get raises?" the answer will probably be "no."

And this matters. A return to power by the Republican Party, in spite of its very disappointing policy performance between 2000 and '08, is not out of the question. Neither is it out of the question to expect attempts to escape political dilemmas by blaming foreigners -- especially Asians and their manufactured goods -- for high American unemployment.

J. Bradford DeLong is a professor of economics at the University of California-Berkeley and a research associate at the National Bureau of Economic Research.


Please contact Caijing Magazine for any inquiries. Reproduction in whole or in part without Caijing's permission is prohibited.
[ICP License: 090027] IDC License:[B2-20040250] Advertising Business License:[京海工商广字第0407号] 京公网安备11010502005607号
Copyright by Caijing. All Rights Reserved