Nonfarm business sector labor productivity increased at a 9.5 percent annual rate during the third quarter of 2009, the U.S. Bureau of Labor Statistics reported today. This was the largest gain in productivity since the third quarter of 2003, when it rose 9.7 percent.That's a huge jump, but what does it mean in the broader context of unemployment?Before any analysis, let me explain what productivity specifically measures. According to BLS:
Labor productivity, or output per hour, is calculated by dividing an index of real output by an index of hours of all persons, including employees, proprietors, and unpaid family workers.So essentially it's output divided by hours of employment. Like any ratio, its value can increase in two ways: if the numerator increases or if the denominator decreases. In this case, BLS says both are happening:
Output increased 4.0 percent and hours worked decreased 5.0 percent in the third quarter of 2009.This isn't good news for unemployment. What you're seeing here is employers squeezing more output out of workers putting in fewer hours. This makes total sense, considering third-quarter GDP grew by 3.5%, while employment and hours worked continued to decline. In order for output to have increased, fewer workers must have been producing more. A demand for more output is what generally drives employment to increase. Yet, in this case, that demand is being satisfied with fewer hours worked. As a result, it's pretty clear that employers have decided to simply get more out of their current workers, rather than turn to the labor market to ramp up hiring. The fear, then, is that this trend will continue. Output may continue to increase, but employers may simply require current workers pick up the slack, rather than look to the giant pool of unemployed Americans. But how does this productivity trend look from a historical perspective? Here's a chart that has productivity (blue) versus unemployment rate (red), since 1970, per BLS data:
I wouldn't get too concerned about the lines crossing, or specific values, but I think the trends are telling. As you can see, a productivity spike around peak in unemployment isn't uncommon.
The last time unemployment was this high was in 1982. At that time, productivity spiked as well, but not until after unemployment began to decrease. That makes sense, because once growth started, employers started hiring, but not as quickly as the growth implied.
Here, I think we're seeing something else. Output increased in large part due to the government stimulus. As a result, employers didn't see their real long-term growth prospects change much, so it didn't make sense to hire a lot more workers. Instead, they're just having current workers pick up the slack, since they understand that the stimulus is temporary. That's why I think productivity growth is leading the labor market's improvement.
At other times with unemployment spikes, like 1975 and 1992, productivity growth appears to be more in line with unemployment peaks. In these cases, stimulus measures likely had a more immediate impact in GDP than in 1982 -- like in 2009.
So really, I wouldn't be too alarmed by the productivity number. I just wouldn't be too optimistic about it either. Just like we saw in previous recessions, once unemployment declines, so will productivity. But I think what it does likely show is that businesses understand the stimulus-induced growth isn't sustainable in the near-term, so they aren't matching their hiring to the increase in output that results.
This article available online at:
http://www.theatlantic.com/business/archive/2009/11/what-does-productivity-mean-for-unemployment/29672/
