Friday, March 27, 2009

Paul Krugman's Nobel Prize, and Economic Geography

Though it seems like eons ago, it was just a couple of months ago that Krugman was awarded the Nobel. The day of the announcement (October 13, 2008), I wrote the following to my department colleagues:
hey, this is exciting.
am all the more excited because most of the courses i teach are either directly or indirectly about economic geography .... and krugman was one of the first neoclassical economists to systematically talk about a "new economic geography"
in his book Geography and Trade, krugman writes, "About a year ago I more or less suddenly realized that I have spent my whole professional life as an international economist thinking and writing about economic geography, without being aware of it"
it will be neat if neoclassical economics alters its course thanks to the 1-2-3 punch from Amartya Sen, Joseph Stiglitz, and Paul Krugman. but, maybe that is asking for too much, eh!
Thus, I was looking forward to the special session on Krugman and economic geography that I noticed in the program for the annual meeting of the premier professional organization for geographers--the Association of American Geographers.
But, the session did not live up to its billing. So, after thinking about it for a day or so, I emailed the following to the session organizer, with a copy to Professor Ron Martin:
Dear Professor Maskell:

I was delighted when I noticed in the program a session devoted to Paul Krugman's Nobel Prize for his work in economic geography.

As a member of the geography department in a teaching university, one of the reasons I look forward to the AAG is for the continuing education benefits from attending sessions. Particularly when the demands of teaching require me to spread my intellectual interests across a range of courses. However, for the most part, the session on Paul Krugman did not advance the discussion more than how the Economist summarized it in 1999--almost exactly to the day of the session, ten years ago.

After my comments, I have copied/pasted the article from the Economist, and the word "geographers" is highlighted throughout because of the keyword search I had employed. I remembered that piece because (a) it came at a time that Krugman was rapidly transitioning into a public role, and (b) I have used that many times over the ten years, including in a book review in Professional Geographer.

I found it equally puzzling that the geographer mentioned in that article, Professor Ron Martin, was not one of the panelists. Well, there could have been any number of logistical reasons behind it. But, puzzling nonetheless. Anyway, my thanks to Professor Martin for his observations towards the end of the session when it was opened up to comments from the floor.

Well, even if I did not find the session at a level I initially expected, my thanks to you for organizing this session. I hope that this session will be the beginning of a series of conversations on forging a truly new economic geography that marries the best ideas and methods of economists and geographers.

************
"Knowing your place"
Economist; 03/13/99, Vol. 350 Issue 8110, p92-92, 1p

Economists say they have rediscovered geography. Geographers are interested to hear it. They didn't know they had been away

THE leading light of the "new economic geography" is Paul Krugman, professor of economics at MIT. These days he is widely known for his popular books and magazine columns, where his speciality is to explain economics in straightforward terms while pouring scorn on rival experts such as Lester Thurow and Robert Reich. Mr Krugman is a brilliant writer of economics for non-specialists. Annoyingly, he is at least as good at the real thing. The "new economic geography" has lately been a chief interest, developing as it does from his earlier research in "new trade theory".

It is too soon to say whether this new area of work, like its predecessor, will become a major field of research in its own right. But it is not too early for geographers to be very annoyed that economists are roaming blithely across their land.

In a new paper, Ron Martin, from the geography department at the University of Cambridge, surveys the "new economic geography" with the critical eye of an actual geographer-and makes some interesting points. As will become clear, much of what he has to say applies not just to new economic geography but also to much else of what economic theorists spend their time on these days.

Mr Martin explains that the two main branches of new economic geography-one of them concerned with clusters of activity, the other with regional growth disparities-raise questions that are entirely familiar to geographers.

On clusters, economists start with the idea that centripetal and centrifugal forces are in opposition when firms choose where to locate. The agglomerating forces are "externalities" such as the ability to tap into an established local market for appropriate labour and intermediate goods. The dispersing forces are the costs of congestion, and the bidding-up of prices for land and labour. In models of clustering, transport costs and labour mobility usually have a central role: if transport is cheap and labour is mobile, agglomeration will tend to outweigh dispersal (and the converse).

According to Mr Martin, these and other models "generate a dull sense of deja vu . . . Geographers were busy analysing industrial location in these terms back in the 1960s and 1970s."

Much the same goes for the other sort of models-the kind concerned with convergence (or lack of it) in regional incomes. Again, according to Mr Martin, the economists arrived late for the meeting. Their starting point was the comparatively recent finding that incomes among regions converge more slowly than the standard neoclassical model of growth would predict. (This model embodies diminishing returns, meaning that as you invest more, you get a smaller return-so poor regions should catch up.) New economic geographers are examining the reasons for this slow convergence. Again, Mr Martin observes that their work "merely revives" ideas proposed more than 30 years ago in proper geography.

He goes further. The geographers have not only been there and done that, they have given it up as a fruitless enterprise and moved on. The real problem with new economic geography, they believe, is its obsession with mathematical modelling. The economists sometimes argue that geography came to a halt way back because the mathematical tools of the day were not up to the job. Imperfect competiton features prominently, for instance, and clever maths is needed to deal with it. Now those tools are available-their use in economics pioneered by specialists in industrial organisation and trade-so economists can breathe new vigour back into the lifeless body of geography.

No, not quite, says Mr Martin. It wasn't mathematical backwardness that led geographers away from this approach but the conviction that to rely too heavily on maths was a dead end. Geographers realised that "formal mathematical models impose severe limits on our understanding. Geographers became more interested in real economic landscapes, with all their complex histories and local contexts and particularities . . ."

A specific complaint is that an economist may be happy to use the same model to explain clusters at completely different scales-at the international level, at the scale of core versus periphery within a single economy, among urban concentrations and even within city neighbourhoods. (Yes, that is just like an economist.) Geographers insist that profoundly different processes are at work at different scales. So modern geography is interested in a more "discursive" approach, piling on detail and colour, in "close explication of locally specific and contingent factors", in building models from the bottom up, not from the top down.

Mr Krugman's response is robust. The geographers are often simply anti-model, anti-quantitative, anti-clarity, he has written. The geographical literature uses terms likes post-Fordism-even Derrida gets a look in-and Mr Krugman finds that very off-putting. (So does this column, it must be said.) Mr Krugman's general defence of maths in economics, and his strictures on using it properly, seem right: economic statements are based on models, he has argued, whether you acknowledge their existence or not. Better to know the model you are using, if only to understand its limitations, than to kid yourself you have moved to a deeper, model-free plane. Quite so.

Yet who can deny, as the geographers complain, that at the frontier of research, abstract economic modelling and the real world have moved dispiritingly far apart? A meeting of minds ought to be possible: a middle ground between bottom-up and top-down. One day, maybe, but tempers will have to cool first.

The article by Mr Martin is "The New `Geographical' Turn in Economics", Cambridge Journal of Economics, January 1999. Among Mr Krugman's many writings on the subject are "Development, Geography and Economic Theory" (MIT Press, 1995).

No comments: