Friday, April 02, 2004

DeLong on Outsourcing 

Brad DeLong and Stephen Cohen have written a superb article on offshore outsourcing, with some excellent (although perhaps somewhat overblown) thoughts about the likely future scale of the phenomenon. In particular, they make these extremely important points:
The economists say that this wave of service trade-driven globalization will, like the last wave of goods trade-driven globalization, be a positive-sum game. There will be more winners than losers, and the winners will win more than the losers. As long as the Federal Reserve does its job to make Say's Law (the claim that the demand will always appear to soak up increases in supply, and so higher productivity means higher incomes and not higher unemployment) true in practice even though it is not true in theory, the most that the losers will suffer in the long run is a fall in their incomes. And other workers and consumers will see their incomes rise and their spending power increase to make the process taken as a whole a good thing for the country, and for the world.

The economists are right. In response to Forrester and to the wave of political upset, McKinsey (as well as others) have hurried out quick pro-"outsourcing" studies. McKinsey (using its truly proprietary--that is, opaque--methods) calculates that every $1 spent on offshoring business practices generates 50 cents in business cost reductions--which show up as higher profits and higher investment in America, generates higher exports to the offshore site, and calculates that displaced American workers do find new jobs, and suffer on average only small wage losses.

But the economists are also wrong. For we have not, we do not as a country make the investments in retraining and rebuilding needed to transfer some of the gains from the winners to the losers, and so make the process of economic change truly a win-win one. There were no regional adjustment funds provided to the cities of Lowell and Fall River Massachusetts in the 1940s and 1950s as their textile manufacturing base pulled up stakes and headed for the lower-cost Carolinas. There was little money spent on Flint and Detroit in particular and Michigan in general in the late 1970s and 1980s to cushion the economic impact of the coming of Toyotas and Hondas to America's shores. Consumers in Boston and San Francisco drove their Accords and Corollas and pocketed the gains rather than having them diverted to rebuilding the Midwest.
This is an important point, well explained and well argued-for. I've noted similar calls for repairing the social safety net here.

DeLong and Cohen predict that at some point in the near future, outsourcing of high-paying service jobs is really going to heat up, causing disruptions an order of magnitude greater than those of the 1970s and 1980s. They may well be right, but it's worth examining some possible flaws in their logic. First, it's not clear (to them or to me or to anyone I know of) just how many formerly non-tradeable services are going to become tradeable. The canonical example of a non-tradeable service is the haircut. It's really, really hard to import a haircut from China or India. I've jokingly suggested that with a webcam, a robotic Flowbee and broadband access, this could all change, but the fact that this seems a laughable idea should give some indication of the problem with assuming that just because a service can be imported cheaply, it will be.

There's a famous (well, famous to economists) paper ("National Borders Matter" by John McCallum, in the June 1995 American Economic Review) showing that despite basically a complete lack of trade protection between the US and Canada, economic activity between Canadian cities was enormously greater than between US and Canadian cities, even when controlling for things like distance and relative wealth. In other words, despite the lack of tariffs, quotas or any sort of border controls on traded goods, Canadians still ended up purchasing domestic products. To my knowledge, there's no consensus on why this effect exists and is so dramatic. But I do think that if it has anything to do with local tastes or personal relationships, the effect should only be magnified in the case of trade in services. To return to my prior example, you're probably a lot more willing to consider buying a pair of scissors made in China than buying that trans-Pacific haircut. To the extent I'm right, DeLong and Cohen are overstating their case. But their basic analysis is still sound, well-argued, and worth your time to read.

This page is powered by Blogger. Isn't yours?

Weblog Commenting and Trackback by HaloScan.com Referrers: