Earlier this week we discussed the myth of manufacturing productivity, and how the number may not account correctly for reduced labor resulting from subassemblies being outsourced offshore but finished in the U.S.. Apparently it struct a chord, and this morning it was mentioned on TheStreet.com Real Money. That, along with a couple of emails I sent to blogger colleagues, has generated a flood of emails in response, with a sampling of the comments posted below. Right or wrong, these guys forget more about economics than I will ever know to begin with.
UPDATE: A reader provided some studies that helped quantify the effect on productivity. Read about it here.
Columnist Alan Reynolds wrote,
The value of manufacturing output really has been going up, not down, and rather strongly. People think otherwise because they're thinking of older products -- T-shirts rather than electronic equipment. Even auto output is quite strong in the USA -- the "Big 3" may be moving more production overseas but Germany, Japan and Korea are moving here.
The data are at bea.gov, but it's not easy.
That's why I wrote a textbook.
And we've lost fewer manufacturing jobs by far than Japan, Germany and (yes) even China. The data and sources are in my book, Income and Wealth, but I think I must have touched lightly on the topic in past columns. They're all online at townhall.com, or at cato.org.
Donald Luskin of Trend Macrolytics and Conspiracy to Keep You Poor and Stupid wrote,
As a first observation, I would point out that the issue you raise is very real – but it only makes the published productivity numbers incorrect if those numbers didn’t already account for it, and I have no idea whether they have or haven’t. These things are always a combination of “economics” and the real-world process of census-taking.
Russell Roberts of Cafe Hayek wrote,
Interesting point, Kevin. I assume that manufacturing output is only "value added" but that could be wrong, too.
Alex Tabarrok of Marginal Revolution wrote,
Interesting post. There are many different measures of productivity. The one you seem to be relying on is physical output per worker in which case I think you are correct. But suppose we measure productivity as the value of output per worker. In that case it's not obvious to me that overseas labor should be counted. Suppose that instead of shipping the intermediate assembly overseas a bunch of robots were used. Then U.S. labor productivity would increase. Why is overseas labor different than robots? In other words, if the value of the output goes up relative to the US labor input that is an increase in the productivity of American labor regardless of how that increase in output relative to input was created. True, the difference is not an improvement in technological productivity but it is an improvement in economic productivity.
Steve Conover of Skeptical Optimist wrote,
Although we are turning into a service economy, Brian Wesbury has said recently that even manufacturing is picking up.
Here's a link to an article I wrote a while back that explains some of the things that aren't so apparent in the official numbers:
http://www.optimist123.com/optimist/2006/02/the_economic_st.html
Brad Setser of RGE Monitor wrote,
I am not an expert on productivity/ how productivity is calculated/ how intermediate subassemblies are accounted for. Theoretically, productivity is measure of the US component of output relative to US inputs -- it tries to measure how efficiently work in the US is done, not how efficiently work outside the us is done.
For the tiny bit it is worth, I have often wondered the same thing -- i.e. if some of the reported gains in productivity weren't mismeasured gains from offshoring/ pure mismeasurement.
So there you have a sampling of comments. I'm going to have to think a bit about the comparison of offshore labor to robots... I'm sure there's a knowledge and creativity component there that should be accounted for. Stay tuned for more analysis. Now back to the vacation...