« Who wants cryonics? | Main | "Whatever is Needed" »
Taliban v. Coase
Sadly, the Taliban are succeeding where Coase (and the Pakistan government) have failed.
The mountain of white marble shines with such brilliance in the sun it looks like snow. For four years, the quarry beneath it lay dormant, its riches captive to tribal squabbles and government ineptitude in this corner of Pakistan’s tribal areas.
But in April, the Taliban appeared and imposed a firm hand. They settled the feud between the tribes, demanded a fat fee up front and a tax on every truck that ferried the treasure from the quarry. Since then, Mir Zaman, a contractor from the Masaud subtribe, which was picked by the Taliban to run the quarry, has watched contentedly as his trucks roll out of the quarry with colossal boulders bound for refining in nearby towns.
Posted by Alex Tabarrok on July 14, 2008 at 08:14 AM in Current Affairs | Permalink
Comments
Alex,
I'm sure you are aware that only the first one-third or so of Coase's
paper ("The Problem of Social Cost") is about the case where contracts
are costless to make, negotiate, and enforce (his example is of an overflow
of cattle from a single farmer's ranch onto a single farmers farm).
The rest of it is a review of various court cases and, near then end,
a discussion of a high contracting cost example involving a railroad
and a multitude of farmers up and down the track that are potential
"victims" of fire damage cause by sparks from the trains (this is Coase's
high contracting cost example, with many transactors to illustrate a
situation to contrast with the one rancher, one farmer example that started
his paper). It is in this case that Coase discusses the key point of his
paper, which is that the problem of social cost is a reciprocal problem.
In the high contracting cost case, if an authority (say, a judge, or the
taliban for that matter) imposes full liability on the so called "polluter"
without holding the so called "victim" to a "reasonable man standard",
then the social efficient allocation will not result if victims behave
unreasonably in the face of the imposed liability on polluters (i.e., if
moral hazard arises).
So when you imply that Coase "failed", I'm not sure what you are talking
about. I don't think that Coase failed to address cases of high
contracting cost that could be improved upon by outside interventions.
Posted by: indiana jim at Jul 14, 2008 8:40:19 AM
IOW, (Afghan tribal squabbles = High transaction costs) -> Coase is not challenged.
Posted by: Alex J. at Jul 14, 2008 9:20:59 AM
Where does Coase talk about co-operation coerced by the very real threat of summary execution? Alex's "I'll write a cute title to this post, facts be damned" approach ain't helping this blog.
Posted by: bartman at Jul 14, 2008 9:29:11 AM
"challenged" may carry a different meaning; my only point is only that Coase did not "fail" to talk about high contracting cost environments.
Posted by: indiana jim at Jul 14, 2008 10:54:40 AM
Depends on how far you're willing to stretch the definition of "transactions costs" I suppose.
Posted by: Michael F. Martin at Jul 14, 2008 11:25:14 AM
Michael,
It is straightforward; no stretch required.
Posted by: indiana jim at Jul 14, 2008 12:49:31 PM
The Coase Theorem is false: it is based on contradictory assumptions.
Bartman, there are a large number of people who, following David Friedman, use the Coase Theorem as if it applied to coercive bargaining.
As usual, defenders of the Coase Theorem invoke "high transaction costs" in a way that defines "transaction costs" so broadly as to make the whole theory unfalsifiable. There are many kinds of costs that are not usefully analogous to a broker's fee: extortion, for example, isn't just a voluntary fee for a transaction one can simply decline to participate in.
But the economists who try to arbitrarily exclude coercive situations from the Coase Theorem are also wrong. Externalities and coercion often cannot be distinguished except by sophisticated tort and property laws (at least) that violate a condition of the Coase Theorem, namely that bargaining is supposed to produce an efficient outcome under any prior allocation of legal rights. In physical reality externalities and coercion cannot be readily distinguished. The guns toted by the Taliban are just particularly effective externalities: they inflict the highest costs at the lowest cost to the emitter of the externality.
Mancur Olson's approach to this problem is far better than the Coaseians'. The story of the Taliban and the quarry is yet another classic case of a stationary bandit out-competing groups of roving bandits, per Olson. The Laffer maximum extortion of the stationary bandit is far lower than the combined Laffer maxima of the roving bandits. The quarry can operate economically only if those extorting money from it are stationary rather than roving.
Posted by: nick at Jul 14, 2008 8:26:01 PM
You're right, Jim.
It's too bad that "transactions costs" was the choice of words. I understand that Coase had a broader view of the meaning of these words than they conjure up for most hearers.
If we had to substitute a different phrase for them, one that was easier for most people to grasp without having read the paper, what could we choose? It would have to be something that captured administrative costs of bureaucracy, both regulatory and private forms of fraud and abuse, and good 'ole fashioned service charges from agents necessary to completing the transaction. Beyond that it would have to capture the costs of gathering information and the time needed to process it through whatever institutional mechanism was in place to facilitate the exchange.
Actually, "transactions costs" is about the simplest two word formulation. But if you want to capture more of the dynamics of the process, I would suggest "asymmetries."
Posted by: Michael F. Martin at Jul 14, 2008 9:47:53 PM
Nick; you wrote:
"Externalities and coercion often cannot be distinguished except by sophisticated tort and property laws (at least) that violate a condition of the Coase Theorem, namely that bargaining is supposed to produce an efficient outcome under any prior allocation of
legal rights."
This is entirely 100% correct. But the "coase theorem" is only the
subject of the first 35% or so of Coase's paper. The remainder is
decidedly about the more interesting case of high contracting costs.
The logic of Coase's discussion is unassailable. If you are trying to
suggest that any time one invokes transactions costs, there are no
operational propositions, this is false. I predict that you will not
buy milk, eggs, bread, etc. on a regular basis from a grocery store that is 500 miles
from your home. If you do this, this implication of transactions cost
theory has been refuted.
Posted by: indiana jim at Jul 15, 2008 3:36:17 AM
But the "coase theorem" is only the subject of the first 35% or so of Coase's paper. The remainder is
decidedly about the more interesting case of high contracting costs.
The main conclusion of the last 65%, namely that transaction costs are the key to diagnosing and solving externality problems, is deduced from the conclusion of the first 35%, that in the absence of transaction costs bargaining over externalities will be efficient, a conclusion which as I have shown is false (follow the link in my post above). In reality institutional and market inefficiencies are due to many kinds of costs, including problems caused by coercion, that are not usefully analogous to broker's fees and cannot be solved just by more efficient bargaining, contracting, or transacting. Many important economic and political problems, such as ensuring that bargaining is voluntary in the first place, require better tort, property, or related solutions ex ante rather than, or in addition to, better contracts and more efficient transactions.
The logic of Coase's discussion is unassailable.
Logic that contradicts itself assails itself. By discovering the contradictory assumptions of the Coase Theorem I have quite thoroughly assailed the "unassailable."
Posted by: nick at Jul 15, 2008 4:02:40 PM
The coase theorem is most definitely NOT false just as the law of demand is not false. All coase is saying in his social cost paper is that the parties will find a mutually-beneficial bargain if the private benefits of the possible bargain outweigh the costs of bargaining (coase himself never used the term "transaction costs"), so get over it
Posted by: enrique at Jul 15, 2008 9:13:26 PM
Nick writes "The main conclusion of the last 65%, namely that transaction costs are the key to diagnosing and solving externality problems, is deduced from the conclusion of the first 35%"
The main conclusion has escaped you Nick, but being the devoted teacher, I will repeat it for you one last time: the problem of social cost is R E C I P R O C A L. Got it? It is not merely that "transaction costs are the key to diagnosing and solving externality problems".
Do you think that Coase got his Nobel undeservedly? I do not, because I can read.
Posted by: indiana jim at Jul 15, 2008 10:26:13 PM
Nick,
Jim has a point in that Coase was aware of the non-dyadic components to the transactions costs that you consider coercive.
Jim,
Nick has a point in that although Coase suggests (largely by inference) that private contracts and property rights are a better way to allocate scarce resources than central planning by government agents, he never really spells out (at least not in The Problem of Social Cost) a coherent theory of voluntary exchange. Not every dyadic exchange is voluntary; not every voluntary exchange is dyadic.
Posted by: Michael F. Martin at Jul 15, 2008 10:43:15 PM
Michael,
"Blessed are the peacemakers"
Posted by: indiana jim at Jul 15, 2008 11:09:03 PM
the problem of social cost is R E C I P R O C A L.
No, the problem of social cost is NOT necessarily reciprocal. This false conclusion of Coase and his followers is based on the following false premise:
All coase is saying in his social cost paper is that the parties will find a mutually-beneficial bargain if the private benefits of the possible bargain outweigh the costs of bargaining
which I have shown to be false. An "offer you can't refuse", for example, is not mutually beneficial, nor are the outcomes based on the more subtle coercions often associated with externalities. Such ex ante coercions or threats of coercion create assymetries in social cost problems.
Coase was aware of the non-dyadic components to the transactions costs that you consider coercive.
In what sense could Coase be said to be "aware" of them when he didn't properly account for them, and indeed when his Theorem (admittedly he himself did not call it a Theorem) is shown false when they are taken into account?
Posted by: at Jul 16, 2008 6:13:50 PM
No, the problem of social cost is NOT necessarily reciprocal...
This was my message, I forgot to fill out the form, sorry for the confusion.
Posted by: nick at Jul 16, 2008 6:16:39 PM
Nick,
As I've said before, Coase knew that transactions costs would never be zero, and that some component of those costs would be borne by non-parties to the transaction. What you're talking about doing is extending the Coase Theorem, not disproving it. You have a valid point, but you're not going to get very far making it if all you do is point out how Coase didn't address it in the Problem of Social Cost. Chances are pretty good that you wouldn't have found your point if the Problem of Social Cost had never been written.
Posted by: Michael F. Martin at Jul 16, 2008 11:56:41 PM
Nick,
As I've said before, Coase knew that transactions costs would never be zero, and that some component of those costs would be borne by non-parties to the transaction. What you're talking about doing is extending the Coase Theorem, not disproving it. You have a valid point, but you're not going to get very far making it if all you do is point out how Coase didn't address it in the Problem of Social Cost. Chances are pretty good that you wouldn't have found your point if the Problem of Social Cost had never been written.
Posted by: Michael F. Martin at Jul 16, 2008 11:57:10 PM
Coase's main point it IS that the problem of social cost is R E C I P R O C A L. It may not, as Nick suggests "necessarily" be so in ALL cases, but that does not change that fact that
this IS Coase's main point. It IS valid to be concerned, as Michael correctly points out, when Coase's framing is applicable and when not. But as Michael also correctly points out: "Coase knew that transactions costs would never be zero, and that some component of those costs would be borne by non-parties to the transaction." This is why Coase has a statement indicating that "nothing could be more anti-social" than to preclude any action that would impose any harm on anyone else. (I'd give you the exact quote, but I'm not in my office at the moment and can't look it up from here easily.)
Posted by: indiana jim at Jul 17, 2008 8:33:25 AM
By the way, I like Indiana Jim's emphasis on the "reciprocal" nature of externalities, becuase Coase himself emphasizes this throughout his original paper. In fact, when I teach Coase's ideas to my students, I like to separate the Coase theorem from the idea of reciprocal harms. Although both ideas are related, the ideas about reciprocal harms is to me, the far deeper and more original idea.
Posted by: enrique at Jul 18, 2008 4:05:19 AM
By the way, I like Indiana Jim's emphasis on the "reciprocal" nature of externalities, becuase Coase himself emphasizes this throughout his original paper. In fact, when I teach Coase's ideas to my students, I like to separate the Coase theorem from the idea of reciprocal harms. Although both ideas are related, the ideas about reciprocal harms is to me, the far deeper and more original idea.
Posted by: enrique at Jul 18, 2008 4:05:19 AM
Flyff gold
flyff Penya
flyff money
buy flyff penya
cheap flyff penya
cheap flyff gold
Posted by: aion kina at Mar 18, 2009 10:23:40 PM
