Economic reports are not usually the kind of thing that gets the heart racing. But earlier this week, a non-profit think-tank called The Information Technology & Innovation Foundation put out a report that amounts to a bombshell.
We first read about the report in the Washington Post. The basic claim is that manufacturing in this country is not doing nearly as well as advertised. At Changing Gears, we’ve made a lot out of the productivity gains in manufacturing over the past couple of years. According to everything we’ve heard, manufacturing productivity has led the way out of the recession, and Midwest manufacturing has been a major driver of growth.
But the ITIF report provides a blunt challenge to that story line. Some of the claims in the report are controversial, and not widely accepted. But even the federal government now says there could be problems with how it measures manufacturing productivity.
And that could have big implications for the policies our leaders consider in the future.
In one sense, the report tells us what we already know – that manufacturing jobs in the U.S. dropped significantly during the last decade. But it says those job losses were more severe than most economists acknowledge – worse even than during the Great Depression. And the recovery hasn’t been nearly as strong as advertised:
“In short, the United States lost two million manufacturing jobs during the Great Recession, and after the recession just 166,000, or 8.2 percent, returned. That leaves 91.8 percent of jobs to be recovered. At the rate of growth in manufacturing jobs in 2011, it would take until at least 2020 for employment to return to where the economy was in terms of manufacturing jobs at the end of 2007.”
Most economists look at this job loss, and see a silver lining. Even though the U.S. has lost manufacturing jobs, output has been pretty steady, according to federal statistics. That means the manufacturing sector is more efficient, and more productive.
But the researchers at ITIF say that’s not true. They say the federal government makes mistakes in how it measures manufacturing output, and grossly overestimates productivity. From the report:
“Correcting for biases in the official data, ITIF finds that from 2000 to 2010, U.S. manufacturing labor productivity growth was overstated by a remarkable 122 percent. Moreover, manufacturing output, instead of increasing at the reported 16 percent rate, in fact fell by 11 percent over the period.”
This is where disagreements over the report begin. The debate gets a bit technical, but it the core disagreement is about the best way to measure output.
Most of the statistics on our economy come, in one way or another, from the federal government. Our stats on output come from an agency called the Bureau of Economic Analysis.
To understand the debate over how the BEA measures output, it helps to think of your computer. The BEA looks at that computer and says it’s twice as powerful as a computer you would have bought five years ago. So it counts as twice as valuable to the economy – in effect, it counts as two computers being produced.
But the ITIF researchers look at the data and say it’s still just one computer.
The example of the computer is an important one, because it turns out the computer industry has been by far the leading source of manufacturing growth in the U.S.
To understand these numbers better, I called Brent Moulton at the Bureau of Economic Analysis.
He takes issue with the way the ITIF report looks at productivity in the computer industry.
“There’s just a lot more computing technology, so I think counting the number of boxes doesn’t measure the output of computers in a sort of meaningful way,” Moulton told me.
He says these computers do add more value to the economy. And most economists would agree that this value has to be measured in the statistics somehow.
But Moulton agrees that the BEA numbers do have some problems.
For example, the BEA helped fund a recent report that found another problem with its measurements. In this case, the error is in how the BEA accounts for imports. Basically, this report claims the agency is underestimating how much of a role foreign trade plays in the economy.
The report says the government could be overestimating manufacturing growth in non-computer industries by as much as 49 percent.
This report was co-authored by Susan Houseman at the Upjohn Institute, based in Kalamazoo, Mich.
“Our statisital agencies were never set up to measure the kinds of economic activities that are happening now, in a rapidly globalizing economy,” Houseman told me over the phone.
She says the bias in federal output statistics are a real issue. But she also says you don’t even need to take the biases into account to realize that manufacturing is struggling pretty badly. You just need to look at the individual sectors within manufacturing.
“If you back computers out, the remaining 90 percent doesn’t look very good,” Houseman says. “And that’s in the BEA statistics. That’s just facts.”
So why does it matter that the productivity of manufacturing in this country has been oversold?
These numbers help determine policy. And when presidential advisors in the past have looked at rising productivity, and falling jobs numbers, they’ve claimed that our economy is becoming more efficient. They’ve claimed that manufacturing is going through the same transformation that agriculture went through: fewer workers, more value to the economy.
But Houseman doesn’t buy that argument. And neither do the researchers at ITIF.
What they see is an economic sector that is struggling to compete with increasing international trade. And they say it’s time for new policies to address the problem.