Geoscience Reference
In-Depth Information
trade are substitutes, and if there are barriers to trade such as transport costs—which
there must be for space to matter—factor mobility will crowd out trade.
What about technology differences? Since we don't have any good way to model
these formally, there's nothing inside the model to say that technologies can't be
different between, say, Chicago and New York. It's pretty obvious, however, that in
practice high mobility of people and easy communication are likely to make such
technology differences transient. With apologies to my New York Times colleague
Tom Friedman, the world isn't really flat—but America is, with people, capital, and
ideas flowing very easily from one region to another.
This in turn means that you can't explain interregional trade flows, let alone
agglomeration and divergent regional growth, in terms of comparative advantage.
Instead, you need to appeal to some form of increasing returns, in which regions
specialize because there are inherent advantages to large-scale production. Mathe-
matically, increasing returns can be consistent with perfect competition if they take
the form of pure positive external economies; in practice, plausible stories about
local specialization, even if they are about explaining industry clusters rather than
the location of individual plants, just about always involve indivisibilities and some
kind of monopolistic competition. So Isard was right in saying that spatial econom-
ics—the economics of location and regional activity within a national economy—
required a different kind of model than what trade theorists were doing in the 1950s.
Now, you might say that this is all well and good, but we had a major rethinking
of international trade in the 1980s, with a new emphasis on increasing returns and
imperfect competition. Did this restore the unity between regional science and
international trade?
The answer is, partially but not completely—and the subjects have been diverg-
ing again with the rise of modern globalization, as I will now try to explain.
3.2
Trade: Everything Old Is New Again
The intellectual path to the “New Trade Theory” of the 1980s began with empirical
observations in the 1960s, inspired mainly by developments in Europe. Both
Balassa (
1966
) and Grubel and Lloyd (
1971
) observed that the formation of the
European Common Market in 1958 was followed by rapid growth in manufactures
trade among Western European countries. And this rapid growth posed a puzzle.
First, why were these countries trading so much? They were similar in resources
and technology; you might say that Europe in the 1960s was (and still is) flat in the
same way I've just described as being true for America, even though labor mobility
wasn't as high. (Actually, the mobility of guest workers from southern Europe and
beyond, who were effectively the marginal source of labor supply in a number of
countries, may have made labor mobility effectively quite high even though
Western European workers weren't themselves mobile.) So where were the sources
of comparative advantage?
One might try to infer European patterns of comparative advantage by looking at
the industrial composition of trade—but when Balassa and Grubel/Lloyd tried to do
