Speaking Freely

Krugman unwittingly illustrates the problem of relying on a single number for overall production rates when he provides an alternative to a competitiveness approach. To say that 91 percent of the slowdown in the growth of per capita GNP "was explained by a decline in domestic productivity growth" does not explain the decline but rephrases it. To say that gnp grew slowly because the growth in output per hour grew slowly is simply to push aside the real question: What caused the decline? Krugman's numerical exercise does not even adequately fulfill the smaller role he assigns to it, to show that foreign competition played a trivial part in lowering the rate of growth of national welfare. This failure occurs because Krugman counts only the prices and quantities of imports, not their impact on profits, investment, jobs and wages. The typical case outlining the advantages of trade to the U.S. economy always focuses on these elements because they are much bigger than the simple, first-round effects of the prices and quantities of imports, which, with a modicum of craftsmanship, can be manipulated to demonstrate whatever one wishes. Similar problems of logic and data flaw the calculations that yield Krugman's most sweeping single-number assertion, that the U.S. trade deficit in manufactured goods has only a very small impact on wages, a reduction at most of only 0.3 percent. The problem, again, is not just with the single number but with the static approach Krugman adopts. Only a dynamic understanding and methodology can appreciate those impacts because that is how they proceed, iteratively, with real and consequential feedback. Finally, national productivity data have several smaller, technical difficulties that radically reduce the reliability of the numbers. It is impossible to get reliable productivity numbers for the core sectors of the service economy, for example, well over a third of GNP. And operationally, market and institutional structures lead economists to assign low productivity growth rates to industries such as semiconductors although engineers know that productivity has grown at fabulous rates.

The clean simplicity and apparent analytic power of the simple, one-number approach, though it fits snugly with the models and methods of traditional American economics, has given rise to efforts to define a different organizing concept, competitiveness, in order to open a broader, more open-minded and modest approach. The competitiveness approach poses a sensible question: How are we doing as an economy? No single number sums it all up, especially given the follow-up: How are we doing compared to the other guys? And why? Competitiveness is reconsideration of a broad set of indicators, none of which tells the whole story but that together provide a highly legitimate focus.