Via “Tyler Cowen”:http://marginalrevolution.com/marginalrevolution/2013/09/ronald-coase-has-passed-away.html. I met him briefly once, at the third meeting of the International Society for the New Institutional Economics, where he gave the “keynote address”:http://www.coase.org/coasespeech.htm. His address was followed by the usual kind of discussion, in which various prominent institutional economists asked self-serving ‘questions’ that were obviously crafted to magnify their role, or further their own specific agenda. Coase, who was then in his late eighties, did a wonderful job of deflating them in a fashion that combined acerbity and politeness. His contribution to economics is sometimes misunderstood. He repeatedly “deplored”:http://coase.org/coaseinterview.htm the way in which the Coase ‘theorem’ had been taken up in the literature – economists had focused on the model in which bargaining would resolve social problems perfectly well in a world without transaction costs, while ignoring his actual point (that such a frictionless environment did not, and never would exist). Obviously, his politics weren’t my politics, nor the politics of anyone else here at Crooked Timber – he’s still worth reading. That he remained so intellectually active – into his 11th decade! – is extraordinary.
{ 13 comments }
Jonathan H. Adler 09.03.13 at 3:17 pm
You make a point that is missed by many of Coase’s fans and critics alike: Coase believed that the question of whether a given government intervention in the economy was justified was ultimately an empirical question. Though he had strong libertarian priors himself, and believed that careful study would vindicate his generally libertarian worldview, the analytical approach he favored does not necessarily lead one to libertarian conclusions or an embrace of free market absolutism. It does, however, counsel against mistaking the world of economic theory for the world in which we live.
JHA
Lee A. Arnold 09.03.13 at 3:39 pm
Coase repeatedly indicated in his interviews that economists weren’t getting his whole idea. It is a profound idea that puts him among the great 20th-century economists, perhaps in league with Keynes. The idea is related to Adam Smith ch. 3 (“the division of labor is limited by the extent of the market”). It goes far beyond the bounds of “New Institutional Economics” or “mechanism design”. Coase’s insight leads to a universal characterization of “increasing returns”.
In essence, institutions are an archetypical “non-rival good”. Things which provide growth — ideas, individuals, innovations, institutions — are ALL reducing the transformation/ translation/ transportation/ transaction costs which are in and under their specific purviews. Returns are enabled, and often increased. Even something so simple as an Archimedean lever works by retranslating the force to be applied, and thus enables a return to effort.
Economists seem to regard Coase’s idea as an adjunct to the more basic ideas of economics. This is an error. There are two coequal sources of growth: 1) specialization and trade (which also involves competition), and 2) trans-cost reduction by the innovation or institution. These are BOTH found in every level of organization. Their contributions to growth may be close to 50/50.
Economic growth theory may be non-quantifiable because it requires two logically different sorts of connections in every level of organization, and these connections are incommensurable: the transaction or transformation whose cost is to be reduced, and the lines of agreement or control to the institution of the cost-saving. It is interesting and it may be significant that one of the primary insights of 20th-century economics was developed by a man who eschewed mathematics.
I think a flow-chart picture might help. I am in the process of constructing a picture of this in one-minute segments. Just completed yesterday, by coincidence, and I was thinking of Coase the whole time: the non-mathematical mechanism of an institution, of any type whatsoever, showing both kinds of required connections:
Chris Mealy 09.03.13 at 4:49 pm
It’s been awhile since I’ve read “The Problem of Social Cost,” but what I thought Coase was getting at was that because transaction costs are almost always high, and the pre-existing distribution rights may be economically inefficient, it’s up to the courts to reassign rights to increase the total product. Which is basically asking the legal system to be economic planners. It seems like that’s how Posner treats Coase anyway.
Chris Mealy 09.03.13 at 4:52 pm
I also remember A. W. Brian Simpson’s “Coase v. Pigou” Reexamined making me laugh out loud.
Peter Hovde 09.03.13 at 5:48 pm
When Coase did get empirical, however, he really bent into knots trying to portray the evidence as consonant with his libertarian perspective, as in his essay on the system of private lighthouses in Britain, which he characterizes as being like a market system, only with fixed prices (and forced purchase of the service, and no choice between competing providers).
Bruce Wilder 09.03.13 at 5:55 pm
Coase did not have the vast, boring, impenetrable output of many academic economists, of our latter-day. He didn’t take up the “hard math”, to make his thoughts seem esoteric or important; he chose to think about important topics, and he made his thinking clear.
Still, I never liked “transaction costs” as a heuristic. It seems to act like a thought blockade, rather than a proto-insight. The core problem is the control of error under uncertainty: that we know enough to organize a production process, but we don’t know everything, we don’t even know for certain what we don’t know, and so risk dominates our own and others’ incentives. The private incentive is to control what you can control, and push off the rest: costs and risks. Institutions, as Lee said, are the public good counterpart that contains that centrifugal force toward a war of all against all, institutions truncate risks and makes them manageable. Why not just say that? One reason is that it is not the philosophy of a curmudgeonly conservative libertarian.
The neat symmetry of The Problem of Social Cost falters, if reliable feedback for control can be constructed for one attribution of legal rights, but not its mirror. If I need to pollute the air a bit to run my railroad, but my neighbors have to be compensated to allow that, that can be setup, in a way that leads me to pay in proportion to the damage I do, and making the effort to control the pollution generated by the process of running the railroad, a production process concerning which I am well-informed, able to act, and positioned to learn. If my railroad and I have the right to pollute, and my neighbors want to bribe me to control it, the problem of feedback and control is a bit dicier. My neighbors don’t have control of the process, or the private information it generates, and are not in a position to learn.
If Coase spent his life being misunderstood by lesser minds, maybe he had some responsibility for setting up that state of affairs.
Henry 09.03.13 at 5:59 pm
dsquared had a good post on the problems with Coase’s account of lighthouses, way back in 2002.
Peter Hovde 09.03.13 at 6:05 pm
That was pre-CT, right? Is the post currently accessible?
Henry 09.03.13 at 7:13 pm
Not available, but if you have access this later article has some of the same counter-arguments.
ben wolfson 09.03.13 at 8:00 pm
Henry, is this not it?
Peter Hovde 09.03.13 at 8:10 pm
Looks like it-thanks Henry and Ben.
James Wimberley 09.03.13 at 11:21 pm
Henry VIIôs system for running Trinity House, which he set up in 1514 as lighthouse regulator, was this:
Doesn´t this match Crooked Timber beautifully? The bloggers are the Elder Brethren, the commenters the Younger Brethren. All that´s missing is the figurehead royal Master, with her rather jolly personal flag.
Ebenezer Scrooge 09.04.13 at 10:50 am
The Coase Theorem reminds me of the Modigliani-Miller result (i.e., capital structure is irrelevant.) They’re both always wrong, but showing why they’re wrong is always a productive exercise.
Comments on this entry are closed.