Tuesday, 14 October 2008

Krugman's Nobel

Many people have many reasons for not liking the awarding of the Nobel Prize to Paul Krugman and to be honest I share many of them. But I also have to admit that there are also good reasons why he got the prize.

The Royal Swedish Academy of Sciences gave the award to Krugman for "for his analysis of trade patterns and location of economic activity", that is, for his work on strategic trade theory and the new economic geography.

What is important about Krugman's trade theory? The standard theory of international trade stated in the early 1800s, when the English economist David Ricardo launched the theory of "comparative advantage" to explain the range and composition of trade. This theory, which was extended during the 1920s and 1930s by the Swedish economists Eli Heckscher and Bertil Ohlin, implies that foreign trade is based on differences among countries. Ricardo studied countries which differ in terms of technology. Heckscher-Ohlin considered countries which differ in terms of access to factors of production; some countries have a relatively abundant supply of labour but a scarcity of capital, whereas the opposite prevails in other countries. The result is that some countries should specialize in and export, for example, industrial products and import agricultural products – and vice versa.

Since its inception, this theory seemed capable of explaining the much of international trade. But over the last 50 years or so researchers found increasingly large deviations from the trade patterns predicted by Ricardo and Heckscher-Ohlin. This is what is now called intra-industry trade. Over recent decades it has expanded, in particular between rich countries. Such trade implies that a country both exports and imports more or less the same goods. New Zealand and Australia both import and export wine, for example. This would not be compatible with the theory of comparative advantage unless the production of New Zealand wine required a wholly unique technology, or a completely different combination of labour and capital than, for instance, the production of Australian wine. But this seems somewhat unlikely.

Around 30 years ago, Krugman introduced an new theory of international trade. It was intended to deal the occurrence of intra-industry trade and was based on an assumption of economies of scale whereby mass production diminishes the cost per unit produced. The car industry would be one where such economies of scale would exist. In addition to economies of scale in production, Krugman's new theory was based on an assumption that consumers appreciate diversity in their consumption. We like lots of different wines rather than just one. After our basic needs for food and housing have been satisfied, it seems as if we opt for diversity and variation in our consumption. In 1977, Avinash Dixit and Joseph Stiglitz had published a model for analysing consumers’ preferences for product diversity. According to this model, each producer, working under increasing returns to scale, becomes more or less a "monopolist" in terms of his own brand, even though he is subject to sharp competition from other brands.

Such a model can be used to show that foreign trade will arise not only between countries which are different (as in the traditional theory noted above), but also between countries which are identical in terms of access to technology and factor endowments. Moreover, it can be demonstrated that extensive intra-industry trade will occur. In fact, it becomes advantageous for a country to specialize in manufacturing a specific car, and to produce it for the world market, while another country specializes in a different brand of car. This allows each country to take effective advantage of economies of scale, thereby implying that consumers worldwide will benefit from greater welfare due to lower prices and greater product diversity, as compared to a situation where each country produces solely for its own domestic market, without international trade.

Today, I would guess that the general view is that the basic mechanisms specified by Krugman complement the traditional Riccardo-Heckscher-Ohlin theory. The truth is that reality encompasses features of both theories.

Economic geography deals not just with what goods are produced where, but also with the distribution of labour and capital between countries and regions. Krugman's approach used in his trade theory – in particular the assumption of economies of scale in production and a preference for diversity in consumption – has also was also found to be useful for analysing geographical issues. Thus Krugman was able to integrate two disparate fields in a cohesive model.

The ideas which would form the building blocks of the "new economic geography" had already appeared in Krugman's trade work. In one paper he asks what would happen if foreign trade became impossible, for instance due to excessively high transport costs or other obstacles. His line of reasoning is as follows. If two countries are exactly alike, then welfare will be the same in both countries. But if the countries are alike in all respects except that one of them has a slightly larger population than the other, then the real wages of labour will be somewhat higher in the country with more inhabitants. The reason is that firms in the more highly populated country can make better use of economies of scale, which implies lower prices to consumers and/or greater diversity in the supply of goods. This, in turn, enhances the welfare of consumers. As a result, labour, i.e., consumers, will tend to move to the country with more inhabitants, thereby increasing its population. Real wages and the supply of goods will then continue to increase even more in that country, thereby giving rise to further migration, and so on.

Krugman's work evolved into the so-called core-periphery model, which shows that the relation between economies of scale and transport costs can result in either concentration or decentralization of communities. Under some circumstances, the forces which contribute to concentration tend to dominate. Regional imbalances arise and most of the population will be concentrated in a high-technology core, whereas a small minority will inhabit the periphery and live off agriculture. Such a mechanism could underlie the explosive urbanization witnessed around the world, with rapidly growing megacities surrounded by increasingly depopulated rural areas. Such an outcomes is not, however, the only possibility. Under a different set of conditions, the forces which give rise to decentralization will dominate. This promotes somewhat more balanced development. Krugman's model can be used to account for the mechanisms at work in both directions. For example, his model indicates that declining transport costs easily generate concentration and urbanization – which seems particularly noteworthy since transport costs have exhibited a declining trend throughout the twentieth century.

This work does justify the award of the Nobel. I do however share the surprise of Mark Koyama at Oxonomics over the award.
I always envisioned Krugman winning jointly with Dixit or perhaps with Bhagwati or another trade theorist.
A joint prize between two or all three of Krugman, Dixit and Bhagwati would have made a lot of sense. Koyama also says
This is also surprising because Krugman is relatively young (is he the youngest winner?) and because the new economic geography is a youthful subfield of economics. I would have imagined a nobel prize going to endogenous growth theory or to incomplete contracts and the theory of the firm before it went to the new economic geography.
I must agree, endogenous growth theory, incomplete contracts and the theory of the firm seem to me to be areas with a longer history which are richly deserving of the prize. And there are others.

Why are some other economists not that happy with the award? As Peter Boettke has written
Krugman is far more politically partisan than any of the recent award winners. Joe Stiglitz became more ideological and partisan after he won the prize, Ned Phelps used the platform of the prize to think "big" thoughts about the capitalist system, but Krugman became ideological and partisan more than a decade prior to the announcement of his prize. And he has not really written serious academic papers or books in economics during that time span. Krugman more or less abandoned scientific economics when he decided to start writing for a broader audience in the 1990s.
Note that the work that got Krugman the prize was done before this time. Boettke goes on to say
Unfortunately, and unlike both [Milton] Friedman and [John Kenneth] Galbraith, Krugman's work devolved from science to ideology and finally to political partisanship. Friedman and Galbraith had always kept (though from differing perspectives) on the scientific to ideological spectrum, but neither became overtly partisan in their writings. This cannot be said for Krugman and the way he has used his platform as an economist and as a columnist for the New York Times for his Democratic partisanship purposes.

This would be innocent enough if Krugman were just another political pundit, but now the prize has given him an enhanced platform from which to pronounce his partisan positions as if they are grounded in economic science. The casualty of this in the public imagination will be the subtle and fine points of the economic way of thinking that we inherited from a long line of political economists from David Hume and Adam Smith to James Mill, Ricardo, J. B. Say and John Stuart Mill to Frank Knight, Ludwig von Mises, F. A. Hayek, Milton Friedman and James Buchanan.
I share Boettke's concerns.

Bryan Caplan makes an interesting prediction,
When Obama wins, Krugman will quickly drop his partisan hackery. He's unfair to his enemies, but he does not suffer fools gladly. And it's safe to say that a year into Obama's presidency, there will be plenty of folly for Krugman to decry.
We just have to wait and see.

No comments: