Tuesday, August 29, 2006

How Reliable Are Productivity Estimates?

Reporting on the Federal Reserve retreat in Jackson Hole, the Financial Times said:
Gene Grossman, a professor at Princeton, offered the central bankers a new way of thinking about the problem [of globalization's downward pressure on most wages], which paints offshoring in a more positive light. He argued that we should think about trade not as an exchange of good but an "exchange of tasks". If some tasks that used to be performed onshore are offshored to lower-cost locations, the result will be an increase in the productivity and wages of workers who perform related tasks that cannot easily be offshored.

Analysis of the US from 1997 to 2004 suggests this positive productivity effect could outweigh the negative labor supply effect on wages in sectors where there had been a lot of offshoring. But it does not appear large enough to offset the broader negative effect on low-skilled wages arising from the ongoing fall in the relative price of labour-intensive goods.
Although even this analysis concludes that offshoring is a mixed blessing, to me it indicates the difficulty in separating offshoring from productivity estimates. I am inclined to believe that much of ex-Fed chair Greenspan's touting of productivity increases to account for the lack of inflationary pressure is really due to offshoring. For example, because of the lower wages, businesses can hire several foreigners for the salary of one American. They will produce more work (be more productive) for the same cost to the business. Thus, if you look at an income statement, it looks as if you get more output for the same input, but the only difference is that more work is being done overseas. Automation, computers, networking, etc., do make American workers more productive, but can you separate out those factors from offshoring? I don't think so, at least not well. All of which means that the only people benefiting from the increase in "productivity" are those at the top of the pyramid, CEOs and a handful of senior executives. The average workers suffer, not only the "low-skilled" mentioned by Grossman.

Interestingly, the article below this one in the FT is "Flows that slip through statisticians' hands."

No comments: