Wednesday, March 19, 2014

Whatever Happened To Regional Science?

Frankly, I fear that it is all but dead, even though organizations and even journals continue to exist that have either that name or its close relative, regional economics, in their titles.  However, I am teaching a course called Regional Economics right now, as I have for decades, and there has not been a textbook bearing the title of either "Regional Economics" or "Regional Science" now for several decades.  Yes, there is one by Philip McCann just out in a second edition from Oxford University Press, which I am using, titled "Urban and Regional Economics," but that is what it as been reduced to, a tail on the dog of the definitely thriving Urban Economics, not to mention that it is a book published in Britain.  But as for the broader "regional science," it has gone the way of the dodo, I fear.

To a substantial degree "regional science" was the invention of a single  man, Walter Isard, whose death in 2010 at age 91 may have been the final death knell of the field, even though he had himself drifted from it several decades earlier into another invention of his, "peace science" after he moved to Cornell University in 1977.  He founded the Regional Science Association in 1954, drawing on the older location theory coming particularly from the German language tradition that began with the 1826 publication of Der Isolierte Staat (The Isolated State) by Johann Heinrich von Thunen.  This was followed by the establishment of the Journal of Regional Science in 1958, which still exists (I published a book review in it a couple of years ago), with Annals of Regional Science starting in 1967, and at the behest of Isard in 1971, Regional Science and Urban Economics, perhaps the most respected of these journals today, although this may reflect its having that "urban" part tagged onto it.  Others began as well during this period, but most of them are thoroughly obscure by now, such as Papers in Regional Science and the International Regional Science Review.

Also, in 1956 after Isard moved from MIT to the University of Pennsylvania he founded the Regional Science Department, which began granting PhDs in that subject (I note that he served on the dissertation committee of my major professor, Eugene Smolensky, who got a PhD in econ at Penn).  A sign of the decline of the field  came with the closing of that department in 1993, mourned publicly by Paul Krugman, whom many had thought had given the field a new life with his work on the "new economic geography" a few years earlier, which had been preceded by similar work in RSUE by Masahisa Fujita.  And while their approach received much attention during the following decade, including with the excellent book they coauthored with Anthony Venables on The Space Economy at the end of the last millennium, that effort seems to have spread into international  trade, economic geography, and urban economics, without reviving or sustaining either regional economics or more particularly the inherently multi-disciplinary regional science all that much (Please note that I am not blaming Krugman for what has happened regarding this in the least).

I am not really sure what happened here, although the disintegration of regional science back into its core disciplines in rump form as urban and regional economics, economic geography, as well as related versions in political science and sociology seems to have played a role.  Certainly there are many interesting issues in regional economics in my mind at least, with foundational issues in nonlinear and complexity economics first being addressed certainly in the broader urban and regional economics, with Krugman among others noting this.  Many would argue that the foundational paper in agent-based modeling was in urban economics, the 1971 paper in the Journal of Mathematical Sociology by Thomas Schelling on urban residential segregation, a paper and model that continues to attract much study, although again, it is viewed as being urban economics more than regional economics or regional science.  The link between regional economics and international economics has long been recognized, with Bertil Ohlin's classic work bearing the title International and Interregional Economics, with some arguing that why Krugman was the first to win a Nobel Prize in international trade since Ohlin was due to his also applying the same ideas to regional economics as did the Swedish Ohlin, even if one of the inventors of what he applied, Avinash Dixit, still has not made that trip to Stockholm as has neither Fujita, who preceded Krugman in applying the same ideas.

Maybe it is because regional economics is a sort of meso-field.  With the exception of when people look at multi-national regions such as the European Union or Eurozone, it generally deals with sub-national entities that may or may not be well-defined.  But it also tends to deal with entities that are above the very well-defined entities known as cities, where microeconomic issues tend to predominate and with policy issues strongly linked to political entities that are distinct in the form and focus from others due to their greater density of population and reliance on highly developed infrastructure, such as water and sewer systems that may be of less importance in "regions."  However, as sub-national entities they do not have the institutions that national economies have, such as those that deal explicitly with macroeconomic policies, fiscal or monetary.  To focus on state governments, they do not have monetary policies at all, and at least in the US, their fiscal policies are highly limited due to their facing strict balanced budget rules.  They are macroeconomically subject to what goes on around them at higher levels, but lack the peculiar fiscal issues associated with managing urban infrastructure as well.  Their meso level nature leaves them in a sort of no-man's land, even though such things as export base multipliers and input-output models are highly relevant to them, with Isard in particular  working on the latter as a student at Harvard under Leontieff.  It may also be the decline in the use of the latter with the fall of central planning after 1991 that has contributed to the demise of the broader field of regional science.

In any case, when I look around now, it is very hard to see even regional economics per se as a field in the US, much less the broader regional science, although I do not have a definite reason why this is the case, and I personally continue to think that the field, or at least the idea of it, is still important, maybe even of renewed importance, which makes its absence all the more disheartening.

I close by noting that the late Walter Isard may have invented another field, one now labeled as "transdisciplinary," but has received nearly zero credit for doing so.  That would be ecological economics, which has a thriving journal of that title founded by Robert Costanza.  Costanza and some of his associates view themselves as having coined this term and invented this field, under the influence of Herman Daly, in the late 1980s, and I cannot disagree that they did so independently at that time.  However, in 1972 Isard published a book that I not only bought a copy of at the time but still possess that uses input-output analysis as its central tool, which also inspired Daly in his first paper from 1973 in the JPE on "Economics as a Life Science," this book bearing the title Ecologic-Economic Analysis of Regional Development from Free Press, which promptly dropped into complete obscurity.  It was Isard's last book on regional science.  I note that it initiated what Daly advocated: thinking of the broader system as an input-output one in which there are quadrants of economic-to-economic, our usual I-O matrix, but then ones of economic-to-ecologic, which includes pollution (among other things), of ecologic-to-economic, which includes agriculture (among other things), as well as a strictly ecologic-to-ecologic, with many pure ecologists dating from the work of Eugene Odum in the 1950s using such matrices to study ecological communities.  I find it unfortunate that along with the death of his field of regional science, Isard's role in foreshadowing modern ecological economics has also largely been forgotten.


Noah Smith said...

This is so disappointing. Economic geography is what got me interested in econ in the first place. I still think it's a much more convincing development story than anything the growth or development people have come up with.

genauer said...

It was actually also drawing me closer into macro.

Is there really nobody left?

Maybe searching for "volkswirtschaft regionalentwicklung"

The problem was that Krugman was the wrong guy for this, with his aversion for data and detail.

He never contributed anything valuable.

Noah, remember talking about Krugmans paper about Zipf's law?

With the demographic development going into reverse in most of Europe, some systematic to partial withdrawal might be beneficial.

To let things just run during buildup is much easier

doc said...

I agree wholeheartedly with everything you said, Barkley. I studied under Bill Miernyk (at West Virginia University), who had studied under and worked closely with Isard. WVU had (still has) a Regional Research Institute, although even in the 1970s its work focused fairly narrowly on West Virginia. (There was a fair amount that dealt with the Appalachian region as a whole, though.)

In that respect, I think there is a chance for revival of the field, particularly in geographically large countries (and despite the issues with state governments) with regions that have strongly divergent characteristics. In the US, Appalachia is perhaps the best example, although the mountain west is also relevant. Even in countries like France, the differences between west-south-central France and the Paris region are marked, And with discussion of independence for Scotland, perhaps the British Isles are ripe for a renewal of interest in regional questions.

The relevance of regional issues for Russia, China, and India is too obvious to spend much time remarking on.

So I may be somewhat more optimistic about a revival than you are... said...


There is a fairly new journal only a bit over a decade old called the Journal of Economic Geography that seems to be quite good and has moved up quite a bit in the journal rankings, although still well below RSUE. Maybe that it is a sign of hope. As it is, I think that "economic geography" is probably in better shape than either "regional economics" or "regional science." There are some other journals that have dealt with this stuff, atlhough some of them do so less now. One that was important in the 80s and 90s was Environment and Planning A. Another was Geographical Analysis, the most mathematical of the geography journals.

There was a lot of important in my mind work done in the 80s by a a group of geographers, physicists, and semi-heterodox economists, that was quite rigorous and mathematical that was completely ignored by the economics establishment, and on this matter I do hold Krugman responsible, who has never cited a single thing by these people and has simply denied their existence, putting himself in the role of the person who mathematized and resolved the problem of increasing returns in economic geography and development by using the Dixit-Stiglitz model, although, as noted above, Fujita beat him to the punch with getting littel credit, althoug PK has at least admitted that he did something and then coauthored with him.

Among those doing this stuff were some associated with the Prigogine group in Brussels, most importantly Peter Allen, a physicist, now at Cranfield University. Another physicist was Wolfgang Weidlich of the Stuttgart Institute of Theooretical Physics who became a father of "sociophysics." A geographer was Roger White at Dalhousie, and also the Swedish economist, Tonu Puu at Umea University. And Walter Isard himself published some things along these lines, with many more on this list. They mostly published their stuff in places like Environment and Planning A, Geographical Analysis, Journal of Regional Science, or RSUE, easily ignored by the broader economics profession, particularly when Paul Krugman said these people did not exist. Much of this involved nonlinear dynamics including catastophe and chaos theory,

I reviewed most of that 80s literature in my 1991 From Catastrophe to Chaos: A General Theory of Discontinuities, and expanded my discussion of it in my 2011 Complex Evolutionary Dynamicss in Urban-Regional and Ecologic-Economic Systems: Beyond Catastrophe and Chaos.


As I said in the main post, I do not blame Krugman for the disappearance of regional science, and indeed he supported it at the time of the ending of the program at Penn. I do blame him for contributing to the widespread ignorance about these contributions from the 80s, which seem to have largely vanished from the general consciousness. I would note that some interesting work has followed his uding the D-S model, although its empirical validity is somewhat limited, something even PK recognizes (and did so in his Nobel lecture). Richard Baldwin, an American and son of a famous trade theorist at Wisconsin (whom I knew well), Robert Baldwin, now in Geneva has extended this to international trade models very insightfully along with various coauthors. With more stricly regional models, Ingrid Kubin in Germany has shown how nonlinear dynamics such as chaotic ones can arise in such models, also with various coauthors.

Regarding Zipf's Law, I note that most of the apps for that are for city size distributions, putting this issue more into the strictly urban economics category, in much better shape than regional science. said...


Yes, I believe there is still technically that institute at WVU, although it is very reduced since the excellent Bill Miernyk left. His book on Input-Output Analysis was excellent, and he made original extensions of that particularly into regional economics apps.

I completely agree that regional problems of poorer regions are increasingly important in many countries, and even arguably within the Eurozone in terms of the troubled PIIGS economies. It is not surprising that a poor state like WV would be a place that might still be hosting a remnant of regional science. But I fully agree that we need a revival of the field.