Monday 14 January 2008

The Armchair Economist

Aaron Schiff has been blogging on his reading of Steven Landsburg's book, The Armchair Economist: Economics and Everyday Life. He writes
I've been reading The Armchair Economist by Steven Landsburg. I think it’s the original "pop econ" book, published more than 10 years before the Freakonomists et al got in on the act. If you liked the other pop econ books and haven't read Landsburg’s yet, I recommend it. I like this book because it really made me think more deeply than, say, Freakonomics did.
First of all I have to agree with two points Schiff makes. The first, that Landburg's book is well worth reading. Despite all the new "pop econ" books that have some out since "The Armchair Economist", I think in many ways it is still the best. Secondly, his view on Freakonomics. I don't really like it, its just seems too glib, and smart for its own good. I think Ariel Rubinstein makes a good point, in his review of the book, when at one stage he asks "What have we learned about Levitt? He is a smart guy with connections in the municipality. What is the connection to economics? None." And the book is supposed to be about economics.

Anyway back to "The Armchair Economist". I agree with Schiff in that it made me think more about the basis ideas of economics. But I think he misses the best part of the book, the chapter on the Coase Theorem. The Coase Theorem is one of the most misunderstood ideas in all of economics. Deirdre McCloskey reckons that only,
Something like a dozen people in the world understand that the "Coase" theorem is not the Coase theorem. (I'll adopt the convention of putting quotation marks around the non-Coasean "Coase" theorem.) One of this select group is Ronald Coase himself, so I suspect we blessed few are right.
The group would be much larger if more people read chapter 9, "Of Medicine and Candy, Trains and Sparks: Economics in the Courtroom", of Landsbury's book. Let me see if I can explain.

Landsbury opens the chapter with a discussion of the famous Bridgman v. Sturges case.
Bridgman made candy in the kitchen of his London home. He got along well with his neighbors, including Dr. Sturges, who lived and practiced medicine in a house around the corner. In 1879, Dr. Sturges built a new consulting room at the end of his garden, adjacent to Bridgman's kitchen. [Bridgman produced candy.] Only after the construction was complete did the doctor discover that Bridgman's machinery made noise-so much noise that the consulting room was unusable. Sturges brought suit in an attempt to shut down Bridgman's business.
The ruling was in Sturges's favour. He got the right to demand that Bridgman shut down his machinery. The judges explicitly referred to the effects their decision would have on the production of goods and services when justifying their decision. The point however is that their decision had no effect on the production of candy or medical care.

What Coase pointed out was that as long as Bridgman and Sturges could negotiate, the decision of the court didn't matter as far the allocation of resources is concerned. If Sturges had the right to stop Bridgman's machinery but Bridgman valued running the machines more than Sturges valued stopping them, then Bridgman could buy the right to run the machines from Sturges. Thus no matter what the court decided, the resource would end up in the hands of whoever valued it most highly.

The court's decision matters to Bridgman and Sturges since it determine who pays who. But it doesn't matter for the allocation of resources, which is what matters to economists. This gave rise to what is normally called The Coase Theorem:
It applies whenever the parties to a dispute are able to negotiate, to strike bargains, and to be confident that their bargains are enforceable. Under these circumstances, the Coase Theorem says that the allocation of property rights, or the choice of liability rules, or more generally any distribution of entitlements (a formulation that includes both property rights and liability rules) has no effect on the ultimate allocation of resources. Judges' decisions don't matter.
Its easy to find examples of when the Coase Theorem doesn't apply because negotiation is either impossible or prohibitively expensive. This can happen when there are many people who need to be negotiated with. In a case like this, the court's decision does matter. Whenever the court's orders are unlikely to be undone by subsequent negotiations, then how the court rules does matter.

Let us suppose that the court's aim is to allocate resources in an economically efficient way. Then how should the court rule? Before Coase, the answer would have been make whoever "causes" the problem liable. Coase argued, however, that it makes no sense to say that one party causes the problem. For example, if a railroad runs tracks through farmland and the trains throw off sparks which occasionally ignite the surrounding crops, then farmers will suffer damage. But does the train cause the problem or the does the farmer cause it? Coase's view is that both parties are needed for the problem to arise. Without the trains, there are no sparks and thus no crops are damaged, but without the crops, they cannot be damaged, no matter how many trains are run. So we see that both the trains and the crop are needed for damage to occur.

Landsburg summaries things thus far as:
And so we come to the flip side of the Coase Theorem. When circumstances prevent negotiations, entitlements-liability rules, property rights, and so forth-do matter. Moreover, the traditional economist's prescription for efficiency-making each individual fully responsible for the costs he imposes on others-is meaningless. It is meaningless because the costs in question result from conflicts between two activities, not from either activity in isolation. The traditional prescription blinds us to the fact that either party to a conflict might be in possession of the efficient solution, and that the wrong liability rule can eliminate the incentive to implement that solution.
The big question is what should the courts do in this situation? Landsburg's advice to judges is
First, we can offer a note of reassurance: If you are trying a case in which the opposing parties are able to negotiate and enforce contracts, then your decision does not matter and you cannot be wrong. Subsequent negotiations will lead to an efficient allocation of resources that is entirely independent of what you decide.

Second, a note of caution: Do not attempt to decide a case by deciding who is at fault. Even if you think that you can make sense of this notion, there is no reason why it should lead to an efficient decision. The costs of damage should be borne by the party who can prevent the damage more cheaply, not necessarily by the one who would be labeled the "perpetrator" by misguided common sense.

Third, a note of condolence: It might be very difficult for you to tell who can prevent the damage more cheaply. Suppose you announce in court that the trains will be liable for spark damage unless farmers can prevent the damage at low cost, in which case the trains bear no liability. Do you then expect the farmers to reveal that they can prevent the damage at low cost? Of course they won't, and unless you are an expert in both farming and railroading, you are unlikely to know where to place the burden.

Fourth, a suggestion: Try to make it easier for the parties to negotiate. If they can, then we are back in the situation where you can't go wrong.
Even Deirdre would have to be happy with this. To end, let me quote what Coase has written about how he views the Coase Theorem:
... I tend to regard the Coase Theorem as a stepping stone on the way to an analysis of an economy with positive transaction costs. The significance to me of the Coase Theorem is that it undermines the Pigovian system. Since standard economic theory assumes transaction costs to be zero, the Coase Theorem demonstrates that the Pigovian solutions are unnecessary in these circumstances. Of course, it does not imply, when transaction costs are positive, that government actions (such as government operation, regulation or taxation, including subsidies) could not produce a better result than relying on negotiations between individuals in the market. Whether this would be so could be discovered not by studying imaginary governments but what real governments actually do. My conclusion: Let us study the world of positive transaction costs.
Seems like good advance.

No comments: