Read

The Unemployment-GDP Disconnect: Can We Solve the Puzzle?

A puzzle is emerging as the U.S. economy slowly recovers: Why is the unemployment rate falling so fast when Gross Domestic Product growth is just...

A puzzle is emerging as the U.S. economy slowly recovers: Why is the unemployment rate falling so fast when Gross Domestic Product growth is just creeping along? At the Bush Institute, we have made sustainable GDP growth of 4 percent annually our target. When a nation’s output of goods and services grows powerfully and consistently, unemployment falls and so, all else being equal, does the federal debt. The relationship between growth and unemployment was first established by economist Arthur Okun 50 years ago. His rule of thumb was that unemployment falls at one-half the rate that GDP rises above the long-term trend. So, if the trend is 3 percent and the economy grows at 5 percent, then unemployment will drop by a full percentage point. In fact, during 2011, GDP growth was only 1.7 percent — well below historical averages. Unemployment should have gone up. Instead, it dropped from 9.6 percent on Dec. 31, 2010, to 8.8 percent on Dec. 31, 2011 – nearly a full point. That decline is consistent with a GDP rate well above 4 percent, not the reality of below 2 percent. The unemployment has continued falling. It’s now 8.3 percent. That is 1.7 percentage points lower than at the peak in October 2009. Since then, GDP has grown at only 2.5 percent, which is considerably below the post-World War II average of 3.3 percent. So what’s going on? Two articles addressed the question yesterday, March 12. In the Wall Street Journal, Jon Hilsenrath paraphrased Christina Romer, President Obama’s first chief economic advisor, with this explanation: “…company managers were so shocked by the financial crisis in 2008 and 2009 that they fired workers more aggressively than they would in a conventional downturn.” In other words, they overshot to the downside. Now, even though the economy is recovering in an unimpressive way, the company managers are overshooting to the upside. Romer worries that Okun’s Law will reassert itself. “The only way unemployment will keep coming down is if GDP growth picks up substantially,” she said. Unfortunately, as Hilsenrath writes, “The economy by many estimates is on track to grow at an annual rate of less than 2 percent in the first three months of 2012.” The Congressional Budget Office, as I noted earlier, is projecting growth of just 2 percent in 2012 and 1.1 percent in 2013. As a result, the CBO sees unemployment rising to 8.9 percent in December 2012 and 9.2 percent in December 2013. A second piece, by James Pethokoukis in The American, offers three explanations:

  1. The unemployment rate does not present an accurate picture of the labor situation since so many Americans have dropped out of the workforce. The ratio of employed persons to the civilian population age 16 and older has declined from about 63 percent to 58 percent since the recession. You can see that dramatic drop on one of the “Growth Snapshots” on the Bush Institute’s 4 Percent website here. There is also an excellent graph with Pethokoukis’s piece showing where the unemployment rate would be if discouraged workers are counted as unemployed: at over 11 percent.
  2. Trend growth is actually lower than most economists think it is. I find this explanation quite weak. Even if trend is 2.5 percent — or 2 percent! – Okun’s Law would not have applied during 2011.
  3. Or maybe growth is a lot higher than we think it is, and the data does not reflect reality. I find this argument weak too. GDP would have to be two or three points higher for Okun’s Law to make sense.

Pethokoukis asks readers for their responses, and one writer, BSR, makes an interesting point, writing, …more productive employment (finance, construction, manufacturing) is being replaced by less productive employment (retail, leisure, hospitality, etc.), i.e., auto assemblers and homebuilders becoming Walmart associates and McD workers. Okun’s ‘law’ has no room for employment skill substitution that affects GDP but not employment numbers.” The only problem is that durable goods manufacturing is up by 444,000 jobs since the trough in January 2010 while food service employment is up by not much more, 531,000, over the same period. Still, BSR may be on to something. Hilsenrath quotes Robert Gordon, the Northwestern University economist, making a similar argument. Gordon sees “clear signs everywhere” that productivity is slowing. The annual rate has been just 0.9 percent over the past seven quarters. A quick-and-dirty calculation for GDP is productivity growth plus population growth, so productivity is the single most important figure to watch in examining growth. By hiring less productive workers – in whatever the sector — companies may be lowering the unemployment rate (probably just temporarily) while having little impact on GDP. I find this argument convincing, and I think there is truth as well to Romer’s explanation and, even more, to the notion that so many people have dropped out of the workforce that the unemployment rate is not as meaningful as it has been in the past. Finally, take note that the disconnect between unemployment and GDP is nothing new. Edward S. Knotek II of the Federal Reserve Bank of Kansas City noted in a thoughtful paper that a similar phenomenon occurred in 2006. Even if Okun’s Law does not seem to be operating in a sluggish economy, there is no need to dismiss its effect in a strong economy. If the U.S. makes the public policy changes that will ignite a sustainable growth rate of 4 percent, you can be sure unemployment will fall consistently and substantially. The recent experience leads to an inescapable conclusion: watch the GDP rate, not the unemployment rate. This post was written by James K. Glassman, Founding Executive Director of the George W. Bush Institute.