Comments Guidelines

All comments are pre-moderated. No spam, slurs, personal attacks, or foul language will be allowed.

Wednesday, August 22, 2012

Is the Growth of Manufacturing Production a Mirage?

A lot of people lament the decline in manufacturing employment, which has fallen by about 1/3 since 2000. As Upjohn Institute economist Susan Houseman points out in the linked article, we're talking about 5.5 million lost manufacturing jobs in that time frame. Here's what it looks like in long perspective

Instead of recovering as it did in previous recessions, after the 2001 recession manufacturing employment continued to fall, as Houseman points out.

But a number of commentators, including Matthew Yglesias and some more conservative ones cited by Houseman, have argued that what we really ought to be looking at is manufacturing output, which has risen steadily except for small blips during recessions.

What's wrong with needing fewer people in manufacturing due to greatly increased productivity?

Houseman argues that the increased productivity is a mirage, due to a single industry, computers. She writes:
Real value added in the computer industry grew at a staggering rate of 22 percent per year from 1997 to 2007 and 16 percent per year from 2000 to 2010. In contrast, average growth of real value added in the rest of manufacturing was just 1.2 percent per year from 1997 to 2007; real value added in the rest of manufacturing was actually about 6 percent lower in 2010 than at the start of the decade.
With that kind of growth, many multiples of GDP growth, we must be an export powerhouse in computers and electronics, right? (Insert joke here.)*

Of course we aren't, so where does that gigantic growth rate come from? If you remember the debates over inflation that gave us the Boskin Commission, you will recall that one of its criticisms of Bureau of Labor Statistics Consumer Price Index (CPI) data was it did not adequately account for improvements in quality over time. Houseman argues that the huge increases in computer power and semiconductor processing speed are what are beneath the apparently massive growth in productivity in the industry. In other words, the price deflators used to calculate real growth are the real reason productivity is apparently growing so rapidly in computers.

If Houseman is right, it means that falling manufacturing employment really is a problem; we have not become so productive that we simply need fewer manufacturing workers. And the fact that productivity is growing by leaps and bounds, yet our trade deficit in electronics keeps getting worse, seems to me to be strong evidence that she is on to something.

* From the Austin Lounge Lizards song, "The Drugs I Need."

Cross-posted with Angry Bear.

1 comment:

  1. There was an article in the NYTimes about five years ago saying that the productivity of labor making white goods, like washers and driers, had increased four-fold in the 90s. According to our appliance repair guy, the number of manufacturers of white goods collapsed during that period with many brands becoming marketing shells, fewer workers, bigger factories and higher output. I'm guessing that there really have been real productivity gains in manufacturing outside of computers.

    If you look at manufacturing, you can't help but notice the increased use of large presses, alternate materials, optimal shaping, robotic assembly and computerized sensing, control and actuation which greatly simplify the design of a lot of devices. If we didn't have low wage countries like China subsidizing their manufacturing sectors, odds are we'd have even higher global productivity as robotics became more and more cost effective. The next generation of sighted, sensored, flexible robots which can be manufactured at more modest prices is going to change the sector yet again over the next decade.

    This has been going on for a long time. Put together a chart of manufacturing workers as a percentage of the workforce since the early 20th century. It falls in a straight line with a few blips for recessions.