### There He Goes Again

Paul Krugman’s latest venture into self-parody starts with a recent paper on the cost of air pollution, which finds that said costs are big and heavily concentrated in a few industries. Krugman then links to a New York Times article surveying Rick Perry’s past clashes with the EPA. With no further argument, he concludes that

Today’s American right doesn’t believe in externalities, or correcting market failures; it believes that there are no market failures, that capitalism unregulated is always right. Faced with evidence that market prices are in fact wrong, they simply attack the science.

Where to begin?

First: Krugman tells us that

Externalities like pollution are one of the classic forms of market failure, and Econ 101 says that this failure should be remedied through pollution taxes or tradable emissions permits that get the price right.

Yes, that oversimplification is indeed sometimes taught in Econ 101. What Krugman doesn’t tell you is that it gets corrected in Econ 102, where students learn that taxing the source of an externality might or might not be efficient policy, depending on the available alternative remedies. It would, for example, be tragic to tax coal plants out of existence if it proved substantially cheaper to clean up their emissions after the fact, or to relocate the victims of those emissions.

I’m tempted to say that this is just Krugman being true to his principle that economics stopped advancing somewhere around the middle of the 20th century (something I certainly believe about music), so that just as Keynes had the last word on macroeconomics, Pigou had the last word on externalities. (Pigou, writing in 1920, is the source of the Econ 101 analysis; Ronald Coase, writing forty years later, advanced us to the level of Econ 102.) Except I know that Krugman knows better, because he once wrote a pretty good piece on green economics for the New York Times, where he made the Coasian argument tolerably well.

Second: the Times article is about Rick Perry’s past criticisms of the EPA, whereas Krugman is citing a brand new paper from the current issue of the American Economic Review. Apparently, his complaint is that Perry failed to account for this research before it existed. Perhaps he’s overestimated the import of fast neutrinos.

Third: I do not think it is beyond Krugman’s analytical powers to recognize that there is no inconsistency between the positions that a) some industries should be more heavily taxed and b) the EPA is often wrong.

Fourth: it is rarely the case that a single journal article is sufficiently definitive that it ought to dictate policy to the point that any dissension would be laughable. (Krugman’s comment on the whole affair is “Hahahahahahaha”.) One possible exception to that rule is Coase’s paper on social cost, which was quite definitive indeed in its refutation of Krugman’s argument from “Econ 101″.

Fifth: Even if we all agreed that the EPA is a paragon of economic efficiency, some of its policies might still be objectionable on some other grounds. Or does Krugman now want to retract every column and every blogpost in which he has appealed to compassion, equity, justice, fairness, etc., etc.?

Sixth: Krugman concludes that

What this tells us is that we are not actually having a debate about economics. Our free-market advocates aren’t actually operating from a model of how the economy works; they’re operating from some combination of knee-jerk defense of the haves against the rest and mystical faith that self-interest always leads to the common good.

What this tells me is that we are not actually having any sort of debate about policy. Krugman isn’t actually operating from a model of how government works; he’s operating from some combination of knee-jerk defense of redistribution and mystical faith that government always serves the common good.

After all, even if we all agreed that stricter environmental policies are desirable in principle, it would not follow that stricter environmental policies are desirable in a world where those policies will be formulated and enforced by politicians and bureaucrats with decidedly parochial agendas.

Krugman says:

Today’s American right doesn’t believe in externalities, or correcting market failures; it believes that there are no market failures, that capitalism unregulated is always right.

It would indeed be naive to believe that there are no market failures. It would be equally naive to believe that there are no government failures, which is what Krugman seems to be assuming.

The choice, all too often, is not between market failure on the one hand and government success on the other; instead it’s between market failure on the one hand and government failure on the other. That calls for some hard thinking about the costs and benefits of the various policy options, thinking that is not advanced by pointing one’s finger and saying “Hahahahaha.” Economists can contribute to this discussion with an honest attempt to illuminate the tradeoff. Tradeoff! Now there’s a concept Krugman ought to recognize. It is, after all, at the core of Econ 101.

Share/Save

#### 84 Responses to “There He Goes Again”

1. 1 1 Nickolaus

As Don Boudreaux often points out, Dr. Krugman (economist) and Mr. Krugman (blogger/shill for the Democrats) often do not see eye to eye.

2. 2 2 Brandoch Daha

I do not think it is beyond Krugman’s analytical powers to recognize that there is no inconsistency between the positions that a) some industries should be more heavily taxed and b) the EPA is often wrong.

You don’t?! Oh, yes you do. You know Krugman better than that.

3. 3 3 Bennett Haselton

Re: “it gets corrected in Econ 102, where students learn that taxing the source of an externality might or might not be efficient policy, depending on the available alternative remedies. It would, for example, be tragic to tax coal plants out of existence if it proved substantially cheaper to clean up their emissions after the fact, or to relocate the victims of those emissions.”

Well, I haven’t taken Econ 102, but couldn’t you still achieve efficient policy by taxing the source of an externality if you tax exactly the right *form* of the externality?

So in your example, rather than taxing “air pollution”, tax “air pollution that doesn’t get cleaned up and actually reaches people’s lungs”. Then, the coal plant would in fact clean up the emissions after the fact, or pay people to move away, if either of those options were cheaper than reducing the emissions.

4. 4 4 Ken B

An excellent post. As for music though, I guess I won’t bother recommending the minimalists (I assume you have heard and disliked them) but seek out Robert Simpson’s symphony 9 from 20 years ago. He is very much an heir of Sibelius & Bruckner, and his 9th is a masterpiece.

5. 5 5 Seth

“It would be equally naive to believe that there are no government failures, which is what Krugman seems to be assuming.”

Thank you.

6. 6 6 John Faben

The link to Coase’s paper should go to http://www.thebigquestions.com/coase.pdf At the moment it has a forward slash in place of the final dot.

Incidentally, what is the legal status with hosting copies of papers obtained from JSTOR, or other online articles on your own site? I’ve noticed quite a few people seem to do it, and none of them have problems, but is this just because the journals don’t bother to do anything about it? Or is there some clause allowing you to do this? If not – why doesn’t someone just host an entire copy of the JSTOR archive for free somewhere?

7. 7 7 Phil

Great!

May I make a request for a blog post about market failures? You’ve given some examples in the past but an in-depth treatment with unassailable examples would be very helpful for those of us who continue to believe, naively or not, that a market failure is an oxymoron (at least in a world with fully defined property rights).

8. 8 8 Bill

Wouldn’t think a guy as bright as Krugman has shown himself to be in the past would be subject to the Nirvana Fallacy, but what the hell do I know?

Also, I like the Kling and Schultz perspective, “Markets fail; we need more markets.”

9. 9 9 Josh s

And then way later in econ (300 level?) you discover that the incentives for government actions often lead in a direction quite different from efficiency-maximizing policy.

10. 10 10 Harold

I don’t think you are wrong, but also Krugman is not totlly wrong either. He jusy has not actually made his case. One must try to see the main point, and not get too lost in minor diversions. As I see it, the point is that Krugman thinks Perry does not believe in regulation because the market will always be better, and Krugman disagrees.
First, the Econ 101 is a reasonable first point to get the point across. The debate highlighted is between no regulation and some sort of regulation. Econ 101 suggests this should be taxes on pollution. The more thorough approach of Coase suggests it is more complicated, but will still need some other regulatory framework. The fact is that efffcient contracts between polluter and pollutee are not going to happen on their own. Some form of Government enforcement will be needed. I believe Krugmans econ 101 is a good enough summary here to get his point across, and the more thorough approach of Coates does not substantially alter the argument.
Second, the brand new paper is new in matters of detail, but hardly new in principle. This does not alter his criticism of Perry, which is that he does not believe in market failures.
Third, Krugman did not say the EPA is never wrong.
Fourth, his point is that there are externalities, which is supported by more than a single paper – the cited one is an example, not the definitive one on which all policy should be based.
Fifth, Krugman does not say that EPA policies are never objectionable.
Sixth, Krugman concludes that “this” tells us that our free market advocates are not operating from a model of how the economy works. This is where he fails. This seems to be based on assumption that the free market advocates do not accept the existence of externalities. He may be right in some cases, but he has not shown it here. I have a feeling that Perry is not arguing from a nuanced, Coatsian view, but that Krugman more or less has summed him up correctly. I may be wrong, but I don’t remember Perry saying “sure, there are externalities – we should find the most efficient way to correct for them” or anything similar.

11. 11 11 Jerry

For practical purposes, there are no externalities for believers in this type of free market.

In order for there to be an externality would require each and every harmed individual to be able to *prove* harm from each and every source that harmed them–AND they would be required to prove “how much harm” each one did. For practical purposes, this is impossible because each plant produces pollutants that affect hundreds of thousands of square miles–with a lot of overlap between pollution sources. The pollutants are NOT tagged–so they generally can NOT be traced back to the true origin.

Thus, the only way to effectively control the producers is to impose federal regulations. Their pollutants have been shown to impact people across state borders–but individual states do NOT have the authority to interfere with the choices of other states other than through the federal govt.

12. 12 12 PrometheeFeu

@Jerry:

There actually are alternatives. For instance, environmental groups could raise funds and pay companies to install and maintain air-scrubbers. They could also try to impose a cost on companies that refuse to install air scrubbers by organizing boycotts.

You could imagine very large class-action suits against companies that release pollutants in the air.

There are plenty of possibilities apart from Piguvian taxes and federal regulations.

13. 13 13 Patrick R. Sullivan

‘I have a feeling that Perry is not arguing from a nuanced, Coatsian view, but that Krugman more or less has summed him up correctly. ‘

Compared to whom, Obama (who thinks it’s a ‘settled matter of economics’ that deficit spending is the way out of recession and that in private enterprise ‘profits eat up overhead’)?

Krugman HAS said that he has never read the Public Choice literature, nor has any plans to do so.

Btw, it’s ‘Coase’.

I second Harold’s critique of the OP – Krugman’s aim in this article is not a sophisticated and nuanced discussion of environmental policy, but simply a somewhat crude, reasonably easy-to-understand, simplified take-down of what passes for policy discussion among people like Perry these days, which is way cruder and less sophisticated and lacking any coherence.

15. 15 15 Nick

“. . . economics stopped advancing somewhere around the middle of the 20th century (something I certainly believe about music) . . .”

Clearly you haven’t listened to As I Lay Dying. It is the canonical music for doing abstract algebra.

16. 16 16 Mike H

The NYTimes article is not Krugman pushing Econ 101 vs Econ 102, It’s Krugman pushing Econ 101 vs Perry’s Econ $-\infty$. I think 101 wins in this instance.

And Econ 102 doesn’t say, like Perry’s $-\infty$, that 101′s externality taxes doesn’t work. It just says you can’t always obviously identify the right target for the taxes. However, does anyone really believe this for air pollution?

17. 17 17 Steve Landsburg

Mike H:

It just says you can’t always obviously identify the right target for the taxes.

No, it says far more: That sometimes any tax is counterproductive.

18. 18 18 Silas Barta

“But why should *I* have to pay the coal-burners to stop? They’re the ones who suck!”

19. 19 19 Tony Cohen

Quick query for readers. Which do think is more likely:

1. Krugman’s critique is unfair because because Perry ascribes to the ‘Econ 102′ model OR

2. Krugman’s hardly nuanced broadside more accurately reflects Perry’s take on minimal/no taxes and a fairly unaware perspective on externalizes.

20. 20 20 iceman

Tony Cohen: Your question would be a good and fair one if PK’s article was merely intended as an attack on Perry (assuming anyone really cares). But true to form it has to morph into a straw man swipe at all “free market advocates” or the entire “American right”. At that point it starts to matter what Econ 102 actually says (since there might be some real economists and other non-clueless people in the mix), and whether stopping at Econ 101 serves to inform or disinform the general audience.

21. 21 21 Chris Koresko

I think this discussion is missing an important piece of background info, which is (if admittedly vague memory serves) that Texas has seen recent large improvements in environmental quality brought about by cooperative efforts between regulators and industry, characterized by a willingness on both sides to be flexible and seek efficient solutions. Under the current Administration that has been replaced by a new rigidity on the regulatory (EPA) side, which is causing large economic losses without producing significant environmental benefits.

22. 22 22 Harvey Galumpky

Harold,

I really liked your post. I would have (I hope) made many of the same arguments, but there can be no question that you stated them better.

One thing about citing Coase as the solution to the pollution problem as I understood Coasian (sp?) solutions, is they become less and less likely as the number of people involved increases (i.e. transaction costs are far from zero). While it’s fair to say that government solutions to pollution aren’t the only theoretical solutions (ie people could be moved etc.), in this particular case the problem appears to be so far ranging that it doesn’t appear practical to think that the parties involved will sort it out.

Does anybody have a better real world example of addressing a pollution type problem than the cap and trade used to addressed acid rain? If so, I’d be really interested in reading a paper about it.

I did not pay for the content, and only read the abstract of the referenced pollution paper, but the last sentence of the abstract reads: “The largest industrial contributor to external costs is coal-fired electric generation, whose damages range from .8 to 5.6 times value added. If the research is solid (I can neither confirm nor deny), this isn’t just some small pricing ineffiency were talking about here. The damage done by the industry is bigger than the value the industry provides. As an example, the price for coal fired electric generation would run contrary to discussion earlier on this blog about how much information is transmitted through price and call the premise into question (I remember eggs, or organic farming or something as the subject at the time).

23. 23 23 Steve Landsburg

Harvey Galumpky:

One thing about citing Coase as the solution to the pollution problem as I understood Coasian (sp?) solutions, is they become less and less likely as the number of people involved increases

I conclude that you don’t understand Coasian solutions.

Suppose a coal plant creates emissions that blacken the windows of houses, or the lungs of citizens, in the surrounding neighborhood. A Coasian solution might be for the homeowner to coat his windows with an anti-soot film, or to wear a gas mask, or to move away. None of these solutions requires collective action and none is in any obvious way precluded by transactions costs.

The cheapest solution to this problem might be to tax the coal plant (which has the unfortunate side effect of leaving the world with less energy) or it might be to avoid taxing the coal plant in order to give the neighbors an incentive to move. That, in a nutshell, is the essence of Coase.

24. 24 24 Harvey Galumpky

Steve,

I thought I acknowledged that government doesn’t have to be the answer, but I also added “in this particular case the problems appear to be so far ranging that it doesn’t appear practical to think that the parties involved will sort it out.”

The study abstract claims that the health costs of coal fired electric generation are somewhere between .8 and 5.6 times value added. It would seem to be a little beyond window coatings. But as I mentioned before I didn’t pay to read the whole paper.

If you have a paper that I can read that has a real world example of a pollution problem that was solved without government intervention I’d be interested in the link. An air pollution solution would be even better, as the only example that comes to mind is cap and trade for acid rain. That problem didn’t sort itself out.

25. 25 25 Steve Landsburg

Harvey Galumpky: Here is a real world example of a (noise) pollution problem that was solved without government intervention: When my neighbor recently had a noisy party that threatened to keep me awake, I acquired a pair of (non-government-supplied) earplugs.

Here’s another: When the BP oil spill made it impossible for shrimpers to continue operating in parts of the Gulf of Mexico, many shrimpers found other jobs. Not all of them were assisted by the government.

26. 26 26 Harvey Galumpky

Steve L:

Thank you for the response.

The noise case is a nice Coasian(?) solution, but it’s an example of a static event between two parties, and my post was about the challenge of finding Coasian solutions without govt intervention when a large number of parties were involved. (lots of interesting issues and caveats raised about the Coasian noise problem in this blog here: http://makeanysense.blogspot.com/2010/08/silliness-of-coase-theorem.html

The BP Oil Spill clearly involves a lot people, but I don’t find the example compelling. Here’s why. As I remember things, there were a number of claims made against BP that were not getting resolved and the logjam didn’t begin to break until the Government got involved, forcing BP to set up a $20B compensation fund. I did a quick google search to check my memory on this, and I learned that Government payouts for the BP oil spill are still ongoing, and per the link below the current state of affairs is: “It has paid around$7 billion to compensate fishermen, hoteliers and cruise ship owners, mainly through the $20 billion fund it created under President Barack Obama’s direction, and expects to pay out another$7.4 billion, according to its regulatory filings.”

There are other articles with slightly different numbers, but this one is the most current based on my very quick searh. All articles I scanned had payout value in the billions of dollars with hundreds of thousands of claims fulfilled. If I had to guess, I’d say the vast majority of the BP oil spill was resolved under government assistance.

27. 27 27 Steve Landsburg

Harvey Galumpky: You’re talking about the question of who pays who. This is quite different from the question at hand, which is “What is the efficient allocation of resources?” I.e. is it more efficient to tolerate oil spills or not to tolerate oil spills?

Having the shrimpers go elsewhere (which they did, with no prompting from the government) might (or of course might not) be the efficient solution. If so, the efficient solution was found with no help from the government. The directions of the various payoffs are entirely beside the point.

(PS: The blog post you cite seems to have been written by someone who is entirely unfamiliar with either economics in general or the Coase theorem in particular.)

28. 28 28 RF

Well, the idea that Republicans don’t believe in externalities is absurd. Their platform is full of opposition to behaviors they claim produce externalities (premarital sex, homosexuality, drug use, etc.) Democrats, on the other hand, think that we can provide universal health care coverage without any costs at all. You know who also doesn’t believe in externalities? Google. I’m typing this in Chrome, and every time I type “extenalities”, it gets marked as not-a-word. Also, perhaps I’m being pedantic, but Krugman incorrectly uses “value-added” as a noun phrase.

Steve Landsburg said
“It would, for example, be tragic to tax coal plants out of existence if it proved substantially cheaper to clean up their emissions after the fact, or to relocate the victims of those emissions.”
Well, that’s just plain silly. Krugman suggests taxing pollution, and you respond that taxing coal plants that don’t produce (net) pollution isn’t efficient. Huh?

“None of these solutions requires collective action and none is in any obvious way precluded by transactions costs.”
And none of them are optimally efficient. Which was surely what he meant by a Coasian solution.

“When the BP oil spill made it impossible for shrimpers to continue operating in parts of the Gulf of Mexico, many shrimpers found other jobs. Not all of them were assisted by the government.”
Again, that’s not optimally efficient. Amelioration is not the same thing as solution.

Jerry said:
“In order for there to be an externality would require each and every harmed individual to be able to *prove* harm from each and every source that harmed them–AND they would be required to prove “how much harm” each one did.”
That sentence sort of got away from you, didn’t it?
This argument is addressed by Coase.

“Thus, the only way to effectively control the producers is to impose federal regulations.”
I don’t think you’re familiar with Coase.

PrometheeFeu
“You could imagine very large class-action suits against companies that release pollutants in the air.”
Tort law is government interference that transfers wealth from one party to another. So it’s not really all that distinct from taxation.

29. 29 29 Jerry

RF:

I don’t think you understood my response.

In order for there to *be* an externality requires proof of “cause and effect”–i.e. the *specific* power plant caused the damage. With multiple plants having overlapping “damage zones” outside the state, the power plants effectively have no externailities.

Coase says the people should move because they have no way to force the power plants to change (state borders and lack of proof to show which–if any–power plants are causing the problem). So people move. But nobody replaces them (for the same reason people left). Which means NO demand for power–so no need for the power plants. Also, Coase begs the question in this type of situation. Given MILLIONS of people having to move to avoid the power plant damage, to WHERE are they to move? The power plants impact farms and other businesses and people who are not able to move unless they give up their “tied to the land” jobs or professions or businesses–how do farmers move their land?

30. 30 30 Steve Landsburg

Jerry:

Coase says the people should move because they have no way to force the power plants to change (state borders and lack of proof to show which–if any–power plants are causing the problem)

It really would be much better if you didn’t try to make authoritative statements about things you clearly don’t understand.

31. 31 31 Harvey Galumpky

Steve L,

Again thank you for your response.

My writing is not clear. I wish it were. It has been a very long time since I last sat in an econ class. One of the key points I remember about Coase is basically he argued that whenever it was practical for affected parties to forge private agreements among themselves, they would have strong incentives to use the least costly solution to the problem.

My point about the BP Oil Spill was not that Government intervention is optimally efficient. My point was that there were so many parties involved that the parties, as point of fact, were not working it out among themselves. My intention in pointing to the dollars and the number of cases involved in the BP Oil spill, was to try to get an idea in the form of a stand in metric of just how far apart the parties actually were from finding an optimally effient solution (I already knew who was at fault and which way the damages should flow). The idea that shrimpers should just have found other jobs, aka loser pays, apparently isn’t optimally effient either, otherwise there wouldn’t have been a need for billions in payouts.

Coase is a pratical man. He understands that for some problems, becuase of the wide variety of activities involved and the large number of people affected, there is no practical way to negotiate private solutions. In such cases, Coase has suggested that regulators should try to mimmic solutions that people would have adopted on their own if negotiations had been practical. I’ve read that he is in agreement that both taxes and tradeable permits satisfy his criterion of concentrating damage abatement with those who can accomplish it at least cost. In the case of electric generation, those with inexpensive ways of reducing emissions will find it attractive to adopt them, thus avoiding carbon taxes or the need to purchase costly permits. Others will find it cheaper to pay taxes or buy permits.

So my original question stands, do you have a real world example of a pollution problem involving a lot of parties where they got together and resolved the issue without government intervention (preferably an air pollution problem)? If so, I’d like to read the paper as it could provide clues to alternative approaches. Cap and trade has a track record of working with respect to reducing acid rain.

32. 32 32 Steve Landsburg

Harvey Galumpky:

So my original question stands, do you have a real world example of a pollution problem involving a lot of parties where they got together and resolved the issue without government intervention (preferably an air pollution problem)?

My original answer stands: You are asking the wrong question. The right question is: In the absence of govt intervention, will or will not people take actions that lead to a more efficient outcome? It is of no importance whether they take those actions collectively or individually. There are lots and lots of examples of people moving away from polluted areas. In some cases, surely, that outcome was efficient and in others it wasn’t. It makes not the slightest difference to anything interesting whether they moved away separately or all together.

33. 33 33 Harvey Galumpky

Steve,

Hmmm. At least I think I’m starting to understand some of the difference in our thinking about the question. So thank you. I am learning, and therefore grateful.

You wrote: “The right question is: In the absence of govt intervention, will or will not people take actions that lead to a more efficient outcome?”

With the strong caveat that I did not pay to read the original paper that spawned the Krugman article, that spawned this blog post, that spawed our discussion, after reading the abstract of the paper, a first pass answer to your question would be: Not in the case being discussed.

As mentioned before the abstract reads: “The largest industrial contributor to external costs is coal-fired electric generation, whose damages range from .8 to 5.6 times value added.” In this case the industry, through the use of an externality in the form of air pollution, is inflicting more damage than the value it adds at market prices.

Bottom line, the consumers who use coal generated electricty are paying much too low a price for it. If the research is solid, in this case the market isn’t getting the price right, and efficient outcomes aren’t appearing on their own.

34. 34 34 Silas Barta

Once, a murderer came into my home and tried to kill me. So, I ran like hell. Problem solved, no government required.

See, what a lot of people don’t “get” is that it’s just as much my fault for being alive when the murderer wanted me dead, as it is the murderers fault for (nearly) killing me.

Sure, fat ol’ government could hobble in and do something clumsy like “ban murder”, but that would prevent us from ever finding out if the gains to the murderer outweighed the loss to me. And so clearly, it’s an inefficient outcome.

The efficient outcome would be for people to constantly pay murderers not to kill them. I know this because that’s what the clear-thinking economists told me. Then again, maybe they were distracted by loud noise…

35. 35 35 RF

Jerry said:
“I don’t think you understood my response.”
No, I don’t think I did. Which was my point: whatever your point is, you’re not expressing it very clearly. My understanding is that an externality is any cost resulting from a decision, borne by someone other than the person making the decision, regardless of whether anyone is aware that the cost exists, or of how large it is. If a tree falls in the forest, and no one hears it, it can still be an externality. I don’t recall anything in the definition of “externality” about it having to be known. Can you point me to an example of where “externality” is so defined

36. 36 36 Steve Landsburg

Silas Barta: What you seem to have missed is that the ENTIRE POINT of Coase’s paper is that you have to look at these things on a case by case basis. Ergo, your analysis of one case tells us nothing about what’s efficient in another.

37. 37 37 Steve Landsburg

Harvey Galumpky:

Bottom line, the consumers who use coal generated electricty are paying much too low a price for it.

No, that does not follow, even granting all your assumptions. Nor will saying it one more time cause it to follow.

38. 38 38 Harvey Galumpky

Steve,

So are you suggesting that when the externalized social cost of producing a good is shown to be greater than the price of that same good, that is not an example of market failure and people are not underpaying for the good? Please elaborate.

Coase wrote: “Under perfect competition, once government has assigned clearly defined property rights in contested resouces and as long as transactions costs are negligible, private parties that generate or are affected by externalities will negotiate voluntary agreements that lead to the socially optimal resource allocation and output mix regardless of how the property rights are assigned” (Ronald H. Coase, “The Problem of Social Cost,” The Journal of Law and Economics, October 1960).

The case in question has neither clearly defined property rights (who owns the air?) nor does it have neglible transaction costs (too many people involved). Not sure why an optimal solution would be expected to spontaneously emerge.

39. 39 39 Steve Landsburg

Harvey Galumpky:

I don’t know how many ways to say this. You have managed to find one paragraph in Coase’s sixty page paper that happens to have nothing to do with what we’re talking about.

Try this example (from the part of Coase’s paper that *is* relevant): Train tracks run through farmland. The wheels throw off sparks that start fires. That’s an externality. If we tax that externality, the railroads will run fewer trains, which sounds like a good thing. On the other hand, if we *don’t* tax the externality, there might be enough fires to make it worth the farmers’ while to stop planting crops so close to the tracks. That’s also a good thing.

Nothing in theory can tell you which of those two good things is the better of the two. Depending on many particulars, it might be more efficient to tax them or not to tax them.

In particular, if the trains are untaxed, it does not follow that we’d get a more efficient outcome by taxing the trains in order to bring the price in line with social cost. We *might* get a more efficient outcome, or we might get a less efficient outcome.

This part of Coase’s paper, incidentally, is based on a real world situation and probably satisifies your demands for an example of such.

40. 40 40 RF

@Steve Landsburg
Perhaps I’m not thinking this through enough, but it seems to me that if taxing trains DOES bring the cost to the RR owner more in line with the social cost, then that is a more efficient situation. Suppose the cost of sparks flying and setting fire to crops is A, the cost of not planting crops is B, and the cost of not running trains is C. If A > C, then the farmer will plant fewer crops, without any government intervention. So at the point where we’re contemplating taxes, we’ll be comparing B to min{A,C}. If B > min{A,C}, then the value produced by the trains is greater than its externalities, and Harvey Galumpky’s condition that “the externalized social cost of producing a good is shown to be greater than the price of that same good” is not satisfied. In this situation, for that condition to be satisfied, the “externalized social cost of producing a good”, i.e. min{A,C}, must be greater than “the price of that same good”, i.e. B. So Harvey Galumpky is saying that if min{A,C} > B, then that is “an example of market failure and people are … underpaying for the good”. If we tax trains, and the RR owner responds by stopping the train runs, then society has a net gain.

You say “In particular, if the trains are untaxed, it does not follow that we’d get a more efficient outcome by taxing the trains in order to bring the price in line with social cost.” If B > min{A,C}, then taxing trains will move the price of trains AWAY from the social cost. So the only time we would tax the trains to bring the price in line with social cost is when min{A,C} > B, and if min{A,C} > B, then reducing the amount of train runs is indeed a more efficient outcome.

What in this analysis do you disagree with?

41. 41 41 Mike H

@Silas – have you been reading Terry Pratchett?

42. 42 42 Ken B

@RF: I disagree with the part where the only choice is to tax or not to tax. I’d say that meshes with Steve’s epitome of Coase: you need to consider the concrete situation in front of you. There might be other options — as there have been in all the examples Steve cited.

A university has a problem: students walk diagoinally across the lawn and this hurts the grass. The SPCG (Socitey for the prevention of cruelty to grass) demands action. Should we tax the walkers? Should we tax the university? Install guard towers and machine guns? Perhaps, just perhaps, we should build a sidewalk on the diagonal.

43. 43 43 Ken B

Silas Barta wrote: “The efficient outcome would be for people to constantly pay murderers not to kill them.”

In fact we do. More interesting (and ironic) from a logical point of view is that YOUR proposal is that we should. One of the ways we do this is by paying them negative amounts upon arrest and conviction.

44. 44 44 Silas Barta

@Steve_Landsburg: What you seem to have missed is that the ENTIRE POINT of Coase’s paper is that you have to look at these things on a case by case basis.

People *already* “look at things” on a “case by case basis”. If Coase hasn’t given a *general* rule for helping resolve these cases, then what’s the point?

Remember, people are using the *exact* same “case-by-case” reasoning that you used in the murder case, to explain why they don’t use Coase’s “solution” even when they know about it. See if you can detect the parallels between:

“Why should I have to pay a murderer not to kill me? That’s extortion?”

and

“Why should I have to pay a coal plant not to dump toxins down my lungs? That’s extortion!”

For extra edification, learn about the similarites between extortion, rackets, and Coasean bargaining here.

45. 45 45 Ken B

Silas Barta: “People *already* “look at things” on a “case by case basis”. ”
Silas Barta’s Evil Twin: “Why should I have to pay a coal plant not to dump toxins down my lungs? That’s extortion!”

46. 46 46 Silas Barta

47. 47 47 Silas Barta

note: My previous comment was in reference to the Ken_B post directly above it.

Also @Ken_B: In fact we do. More interesting (and ironic) from a logical point of view is that YOUR proposal is that we should. One of the ways we do this is by paying them negative amounts upon arrest and conviction.

Well, it wasn’t my proposal, but in any case, you’ve just equated pre-Coasean and Coasean solutions. If you’re allowed to count punishment of wrongdoers as “paying negative amounts via Coasean bargaining”, then people had hit on Coase’s insight long before he did! Indeed, this is exacty what people are advocating when they want legal penalties for burning someone’s crops, or Pigovian taxes on pollution.

And once again, Coase’s defenders try to salvage their position by reverting to non-Coasean or anti-Coasean arguments. (Steve_Landsburg, for his part, does the same thing when you point out the implication that we should pay the “Danegeld” instead of banning murder.

48. 48 48 Ken B

@Silas: Of course it is your proposal. You propose gov’t action to prevent or deter murder in favour of (positive) payments. I assume you do not favor pre-emptive assassination (you are not WillA after all) so you probably favor deterence and punishment. Which I am pointing out are negative payments. (I suspect that you see these differently because you are thinking in terms of blame not consequences.)

More abstractly we DO bribe people not to kill you, routinely. We do this through various indirect means such as education, charity, exhortation, welfare. And importantly enforcing the law, which protects the incipient murderer and stops his slide into such as extremis that he would kill you for the change in your pocket.

49. 49 49 Silas Barta

@Ken_B: Of course it is your proposal. You propose gov’t action to prevent or deter murder in favour of (positive) payments.

Okay, I misunderstood; I thought you were referring to the “pay people not to murder” when you said “my solution”. And indeed, I do favor government/private-security bans on (certain types of) anti-social behavior.

Which I am pointing out are negative payments. (I suspect that you see these differently because you are thinking in terms of blame not consequences.)

Oh, please, not this load again. “Look at me, I’m so enlightened, I use Coasean solutions that free me from barbarically thinking in terms of blame and only think about what’s *good* for everyone!” Sorry, this isn’t the first time I’ve heard that confusion.

For one thing, sorry, but you are also thinking in terms of blame with your pseudo-Coasean solutions (or rather, your anti-Coasean hacks to salvage his solutions from the realm of absurdity). It’s just that you don’t use the word “blame”, but some identical concept that allows you to forget the whole “bargaining” thing and just inflict disutility on performers of an entire class of behavior so as to avoid rewarding it.

In other words, you came to the _exact same_ policy conclusion people had come to millenia before Coase: bargain with people who have an honest disagreement with you, but punish those who hurt others. So in what non-trivial sense are you not thinking in terms of “blame”? If it walks like a duck…

Second of all, the reason I reject its classification as a “payment” is that the whole freakin’ _point_ of Coase’s argument is that all of the transfers in his solution are Pareto-optimal, making the resulting re-allocation of resources efficient (in the economic sense). So you’d be contradicting the original argument to start counting forced (i.e. _non_-Pareto-optimal) payments as the same kind of payment that leads to Coase’s result.

We do this through various indirect means such as education, charity, exhortation, welfare. And importantly enforcing the law, which protects the incipient murderer and stops his slide into such as extremis that he would kill you for the change in your pocket.

Right: the solution everyone came up with *before* Coase, and which Coase ends up reverting to in all the cases that actually, you know, _matter_.

50. 50 50 Steve Landsburg

Silas Barta:

the whole freakin’ _point_ of Coase’s argument is that all of the transfers in his solution are Pareto-optimal, making the resulting re-allocation of resources efficient

This extraordinary misunderstanding goes a long way toward explaining a lot of your more bizarre recent postings.

51. 51 51 Ken B

Silas hyperbolically attributes this thought to me:
“Look at me, I’m so enlightened, I use Coasean solutions that free me from barbarically thinking in terms of blame and only think about what’s *good* for everyone!”

But I don’t think that is what is implied by the point I made (I am not addressing the question wheter it is a good guess or off the wall.) It seems from your commehnts and examples that you are seeing the coal dust producer as inherently to blame. (I admit this is only an inference but it seems required by the logic of your argument and your tone.) Now that may be true in many cases. But perhaps the factory was located in Siberia, 100 miles from anyone, and you moved there 100 yards down-wind. An extreme example no doubt, but it it illustrates exactly the sort of details of the case at hand you that are possible and that in your formulation you choose to ignore.

52. 52 52 Steve Landsburg

RF: I don’t have the energy to work through your example so I’ll give you a fresh example; you can translate my variables to yours if you wish.

Consider a competitive firm that causes pollution.

I. At the firm’s private optimum, in the absence of a tax, it earns a surplus of A, while imposing pollution costs of B. Net social gain: A-B.

II. Now suppose we make the firm pay a tax equal to the pollution it causes. As a result, they produce less, earning a profit of C while imposing smaller pollution costs of D. (That is, D < B.) The government collects D in tax revenue. Net social gain: C-D+D = C.

Because this is the social optimum, we know that C > A-B.

Now suppose the neighbors can move away, eliminating all pollution damage, for a cost M with the property that D < M < B.

1) If we impose a tax, the neighbors have to live with pollution costs of D. They'd rather do this than move away, so they stick around. Net social gain, per II) above, is C.

2) If we don't impose a tax, the neighbors, if they stay, have to live with pollution costs of B. Since M < B, they'd rather move, which they do. Net social gain: A-M (the firm's surplus minus the cost of moving).

If C is bigger than A-M, we maximize social gain by imposing the tax. If C is less than A-M, we *sacrifice* social gain by imposing the tax.

Thus if we assume that D < M < B, and that C > A-M, then you get more gain by *not* taxing than you get by taxing.

If we assumed different inequalities then we’d get a different result. That’s Coase’s whole point: The result can be anything at all depending on what inequalities happen to hold, and those are going to differ from one case to the next.

53. 53 53 Silas Barta

@Steve_Landsburg: I’m _so_, _so_ sorry for that _complete_ butchering of Coase’s exposition. Please replace “Pareto-optimal” with “a Pareto-improvement”. Better now?

@Ken_B: Yes, there are situations where coal burning should be allowed. No duh. Nevertheless, when the full implications of Coase’s solution are appreciated, its proponents inevitably revert to pre-Coasean methods of determining who should be punished, thereby operating identically to someone who uses the “blame” concept to decide that murderers should not be bargained with. So my points stand.

54. 54 54 Steve Landsburg

RF: Let me add, in case it’s not clear, that the same exact structure applies to the trains/sparks/farmers problem; make the firm a train company, change pollution to sparks, and change “move away” to “move your crops”.

55. 55 55 Steve Landsburg

Silas Barta: If you have any interest in understanding what Coase’s point actually was instead of just continuing to propagate your made-up cartoon version, you might want to look at my reply to RF above; this captures the whole point in a nutshell.

56. 56 56 Silas Barta

Sorry, Steve_Landsburg, none of your points there were new to me, and I remain unconvinced that “maybe, paying off murderers is the social optimum — we just don’t know until we let the transactions play out”.

57. 57 57 iceman

Silas Barta: It seems the flaw in your analogy is that unlike the murderer, we’re assuming the coal plant provides some social benefits as well, which warrants a closer look at the particulars of the trade-off involved. (I assume we don’t really have to consider the utility gains to murderers.)

BTW undoubtedly “pre-Coasian” people sometimes figured out efficient solutions, but identifying them as such still provided a valuable insight for judges and policymakers.

Ken B: I suppose one could argue that someone still has to pay for the sidewalk. But that’s all a judge etc. will determine, and the good news is it’s cheaper than guarded towers.

58. 58 58 Steve Landsburg

Silas: I’m glad that you’ve figured out that paying off murderers is not a social optimum. I hope you’ll take the next step in your intellectual development and figure out that nobody thinks otherwise.

59. 59 59 Steve Landsburg

Silas:

@Steve_Landsburg: I’m _so_, _so_ sorry for that _complete_ butchering of Coase’s exposition. Please replace “Pareto-optimal” with “a Pareto-improvement”. Better now?

No, not the tiniest bit.

Two questions:

1) You wrote: all of the transfers in his solution are Pareto-optimal

Since I haven’t the foggiest idea what this means, perhaps you can clarify in the context of the farmers/trains/sparks example. Exactly what transfers are you talking about? What resources do you view as transferred?

2) Is it your assertion that in the farnmers/trains/sparks example it is always efficient to tax the trains? If so, we ought to be able to settle this with a little arithmetic. If not, then of course you agree with Coase.

60. 60 60 Silas Barta

@iceman: It seems the flaw in your analogy is that unlike the murderer, we’re assuming the coal plant provides some social benefits as well, which warrants a closer look at the particulars of the trade-off involved. (I assume we don’t really have to consider the utility gains to murderers.)

Actually, yes we do. We can’t be judgmental or use the concept of “blame”, right? I remember an important story about a famous economist rejecting a paper after the first few lines simply on the basis that it ignored the benefit of crime to criminals. (Though I agree that even criminals benefit from the existence of low-crime environments.)

So, in terms of whose utility we get to ignore, where do murderers end and coal burners begin?

@Steve_Landsburg:

1) Sorry, I meant all of the *bargains*: X pays Y not to do Z; Bob moves farther away from Alice, etc. That is, each step that Coase suggests would happen in the absence of transaction costs. I’m sure my intended meaning must have been _so_ hard to figure out in that context, you might have needed a 150+ IQ to guess at it, you were completely justified in claiming my point was 100% wrong.

2) No, that is not my assertion, and no, it doese not follow that I agree with Coase. For Coase’s reasoning applies just as much to murderers as it does to sparky trains, where we clearly see all of the problems, and indeed inefficiencies, with such “bargaining”. (And as one of the linked blogs stated, it’s hard to see where bargaining ends and extortion begins.)

And so every problem you see with a Coasean solution to murder — i.e. all the long term incentive effect problems, debasing of property rights, the need for there to be good, enforceable rules to begin with, the stupidity of paying the Danegeld, etc. — applies just as much to the Coasean solution to sparky trains (in at least some cases).

Now, let’s turn it around: Do you see why the “problem” or murderers is not simply a matter of getting transaction costs low, and that you shouldn’t just buy out whatever rights to murder they might have? If so, then you agree (with me) that Coase hasn’t provided even a theoretical solution to any non-trivial case, and his actual position regresses to the solution people already used long before him: negotiate with the reasonable, punish the unreasonable.

61. 61 61 Ken B

It’s a side point but since it came up …. Steve is wrong on a point. Lots of people DO approve of paying murderers (positive amounts) not to kill us. The usual term for these people is “appeasers”. Once upon a time, back about 1938, appeasement was widely considered the moral high-ground (I am not kidding about this.) Indeed in 2003 I read various pieces arguing we ought to *subsidize* Saddam Hussein so he wouldn’t be aggressive and wouldn’t represent a threat.

62. 62 62 Ken B

@Iceman: “I assume we don’t really have to consider the utility gains to murderers.” Actually some months ago Steve argued that we do.

63. 63 63 Silas Barta

@Ken_B: That’s not a side point, and indeed represents a problem with Coasean bargain that some folks are overlooking.

64. 64 64 Harvey Galumpky

Steve,

Time does not permit a full response. In very brief summary if you think I missed the point in Coase that externalities/nuisances etc. are symmetric conflicts between the interests of the two parties and best fit solutions are not always obvious, I selected the wrong quote.

If you believe, as Coase himself indicates, that the paper is really about the importance of studying the world of transaction costs, my selection wasn’t so bad (there are better) as it points precisely to the preconditions required to expect the parties to arrive at an efficient solution on their own.

The obvious obstacle in the way of Coase’s thought experiment about sparks and trains is transaction costs. If there are many farmers and many railroads, or private information about costs or how much care the parties are taking, bargaining can break down and an inefficient outcome will likely occur.

We know we live in a world with transaction costs, because according to Coase, a world without transaction costs would be a world without firms. Firms are still around.

So I’m writing from the point of view that we already know that because of the significant transaction costs involved in coal fired electrical generation, there is very little likelihood an efficient solution will emerge under any approach (including letting the parties work it out, taxes, or cap and trade or whatever other option you want to propose), and probably the best we can hope is something more efficient. My bias is that something needs to be done. The question is what?

If you assume I don’t understand that taxes and caps may make matters worse, I can see why you don’t think my quote is relevant. But given the caveats I’ve apparently too often repeated in this specific case, RF did a much better job of describing the back of the envelope math going on in my head than I ever could (and translated it back to trains!). I think the assumptions of this case greatly increase the likelihood that a tax or a cap won’t make matters worse, but better. I’ll need to think about your response to RF a bit, but in step one A-B is negative (too oft repeated caveats apply) which is probably very unusual (and probably couldn’t happen in a world of no transaction costs), and really the point that started the conversation.

65. 65 65 Steve Landsburg

Silas:

Do you see why the “problem” or murderers is not simply a matter of getting transaction costs low, and that you shouldn’t just buy out whatever rights to murder they might have? If so, then you agree (with me) that Coase hasn’t provided even a theoretical solution to any non-trivial case,

Wait a minute. Because I agree that there is no “Coasian” solution to this particular case, I must agree that there is no “Coasian” solution to any non-trivial case?

And this even after I’ve pointed to a non-trivial case in which there is such a solution?

Color me baffled.

66. 66 66 Ken B

Baffled clashes with your complexion; stick to earth tones.

67. 67 67 Silas Barta

@Steve_Landsburg: Please read my posts carefully if you wish to reply to them. Let’s compare my post and your response, and I will highlight the difference:

Me: If so, then you agree (with me) that Coase hasn’t provided eve a theoretical solution to any non-trivial case,

You: Wait a minute. Because I agree that there is no “Coasian” solution to this particular case, I must agree that there is no “Coasian” solution to any non-trivial case?

See the difference there? Between Coase providing a solution, vs. a Coasean solution? A Coasean solution is when one party buys out the other’s right to such-and-such or otherwise rearranges their life so as not to be hit by the externality. These existed before Coase, believe it or not.

So yeah, the solution in this situation does exist, though Coase did not provide it (in the sense of originating it as a novel idea people hadn’t thought of before).

Long before Coase, people followed a policy of “bargain with reasonable people, but punish anti-social people”. Bargain when the rights are ill-defined, don’t bargain when someone torches your crops. So it’s not that people before coase didn’t know you could bargain away externalities; it’s that they put the sparky train into the arsonist category, the same category you might put, say, a murderer.

But wait: why do you think there’s no Coasean solution to murderers? Why can’t we simply pay their Danegeld? Problem solved, right? What’s the big deal? Could that “big deal” be the very consideration that people aren’t impressed with the idea of paying someone who’s burning your crops?

68. 68 68 Silas Barta

Sorry, the last sentence should read, “Could that “big deal” be the very reason that people aren’t impressed with the idea of paying someone who’s burning your crops?”

69. 69 69 Ken B

I think Silas Barta simply does not understand what a “Coasian solution” is. He seems to think it means one that leaves the person afffected by the externality happy. He wants a ‘solution’ to the plight of the affected sufferer not relizing that 1) the situation can be symmetric and who is the sufferer is not always so clear and 2) Coase is talking about maximizing total social gain not the gain of either of the parties to the dispute seperately.

70. 70 70 Steve Landsburg

Silas: The problem with the murderers is that if you pay one off, another will show up. This is a nearly trivial observation, and it’s clearly stated in every textbook treatement of externalities/Coase with which I am familiar. (See, for example, page 441 of the 8th edition of my own textbook, which contains the sentence “Unfortunately, Jack has 13 identical cousins, all named Jack, each of whom is prepared to present Jill with the same threat”.)

It’s very commendable that you thought of this issue all on your own, but that doesn’t make it an original discovery, and you do look rather ill-informed when you keep repeating it as if it were something we didn’t all already know.

And of course this issue does not arise in the train example. Paying off murderers calls forth more murderers, but moving your crops does not call forth more trains. You look even more ill-informed when you keep trying to call attention to your great discovery in a context where it is not relevant.

These facts remain:

1) In the trains/sparks example, Pigou’s analysis says that you always get a more efficient outcome by taxing the trains.

2) That analysis is incorrect. You sometimes get a more efficient outcome by taxing the trains, and you sometimes get a more efficient outcome by not taxing the trains.

3) Although 2) is not a deep or difficult point, Pigou had it wrong and Coase had it right. That counts as a contribution.

You are right, of course, that Pigou and others always knew that people would take steps to protect themselves from externalities. What he missed was that because Pigovian taxes disincentivize those efforts, Pigovian taxes can be counterproductive.

Which of points 1), 2), 3) do you not understand or think is wrong?

71. 71 71 Silas Barta

I disagree with this: but moving your crops does not call forth more trains

Not only is it wrong in this specific example[1], it completely ignores the scope of the incentive effects. If people can get paid for doing anti-social things, then they will increase their anti-social activites, massively decreasing utility and productive capacity all around. Harvey_Galumpky even gave you a nice link that explains this very articulately.

This point bears directly on your glib suggestions that people “just” move away or buy out coal burners’ rights: it doesn’t work as a solution, for any reasonable person, if you are to look at anything beyond the immediate wake of the event. If “owning” farmland means that anyone can come by, torch it, and build railroads through, who farms? Who eats?

And if you already “get” this point, why don’t you act like it, such as by making arguments that don’t imply endorsement of extortion?

Yes, _sometimes_ it makes sense to bargain, such as when it’s an honest disagreement over who deserves the right to do what. Is *this* such a sometimes? If yes, where do you draw the line between that and extortion, and on what basis? Because your latest, half-hearted attempt based on “this doesn’t draw forth more trains” sure as heck doesn’t work!

[1] If every farmer has to “just put up” with trains like that, then I’ll send through a train specifically designed to throw sparks off as far as possible until he starts paying me to stop or abandons his land, let me run more railroads through it. At what point did I become a plain-vanilla arsonist rather than an _enlightening_ example of the miracles of low transaction costs?

72. 72 72 Silas Barta

@Ken_B: Yep, there is a perfect symmetry between your whiny desire to live and a murderer’s desire to kill you, and anyone who doesn’t recognize that you are both EQUALLY at fault for this conflict is a moron who just hasn’t fully assimilated the deep insight in Coase’s ground-breaking, blame-transcending treatise.

73. 73 73 Steve Landsburg

Silas: Either we’re not communicating or you’re just being an ass for the sake of being an ass; it’s getting harder to tell which. I can think of absolutely no reason why moving a farmer’s crops will call forth more trains. In the regime where trains are not taxed, the railroads have no reason to care whether the crops are there or not. Moving the crops does not reward the railroads.

This leaves us with three options:

a) You’ve completely minunderstood the example and you think that moving the crops somehow rewards the railroads. In that case, you need to reread the example.

b) You see some mechanism I don’t see by which moving the crops rewards the railroads and calls forth more trains. If so, you need to spell out what that mechanism is.

c) You just enjoy being an ass.

If b), please be explicit about the missing mechanism. If your response makes it clear that we’re in case c), I’m cutting off this discussion.

74. 74 74 Silas Barta

@Steve_Landsburg:

It’s b), and I explained the mechanism in the footnote. I shall repeat it again here:

If every farmer has to “just put up” with trains like that, then I’ll send through a train specifically designed to throw sparks off as far as possible until he starts paying me to stop or abandons his land, let me run more railroads through it. At what point did I become a plain-vanilla arsonist rather than an _enlightening_ example of the miracles of low transaction costs?

IOW, if the railroad has the right to burn the farmer’s crops, they can leverage this for extortion or to just seize his land by driving him away, and encourages other railroad to act similarly around other land.

This leaves us with three options:

a) You don’t read footnotes on principle. If so, you need to start.

b) You don’t see why being able to get away with burning someone’s crops could lead to inefficient outcomes, even after I explain it to you. If so, you need to spell out where my explanation is in error.

c) You just enjoy being an ass. If that’s the case … well, I guess it’s never gotten in the way before, has it?

75. 75 75 Steve Landsburg

Silas:

Thanks for clarifying what it is that you’re not getting.

If every farmer has to “just put up” with trains like that, then I’ll send through a train specifically designed to throw sparks off as far as possible until he starts paying me to stop or abandons his land, let me run more railroads through it. At what point did I become a plain-vanilla arsonist rather than an _enlightening_ example of the miracles of low transaction costs?

The farmers in this example move away. They do not pay the railroads to stop throwing off sparks, because of the problem of collective action. The sparks affect many farmers, and it is in no farmer’s interest to pay off the railroad to stop. This free rider problem is an example of a high transactions cost, which prevents bargaining between the farmers and the railroad. Given that high transactions cost, the railroad gains nothing from running another train.

I had assumed, perhaps incorrectly, that you’d actually read Coase and therefore assumed you knew that the high transactions costs were a critical assumption in this example. I hope it’s clearer now.

(More generally, one of the key points of Coase’s paper is that you need high transactions costs to make any of the examples interesting — but that’s okay, because high transactions costs are often the real-world norm.)

Edited to add: “Moving” in this case does not mean abandoning the land; it just means not planting crops up quite so close to the tracks.

Edited to add further: When the railroad is taxed to reduce the externality, the outcome might (or might not) be inefficient. When the railroad is required to reimburse farmers for the damage (as opposed to paying taxes to a third party), the likelihood of an inefficient outcome increases.

76. 76 76 Silas Barta

@Steve_Landsburg: The farmers in this example move away. They do not pay the railroads to stop throwing off sparks, because of the problem of collective action. The sparks affect many farmers, and it is in no farmer’s interest to pay off the railroad to stop.

Yes, I’m aware of that. The point remains that IF you grant the RR the legal right to burn crops, then they can *deliberately* burn just the right crops to perform extortion. (Yes, Steve_Landsburg, even if the example doesn’t begin by positing that this is what they want to do.)

So, the initial judgment of who owns what makes a big difference in how much one side can “tax” the wealth of the other, as the link I pointed you to explained. This then affects efficiency of the final arrangement: if the RR has to buy the buffer zone, farming doesn’t instantly become unprofitable, thus preventing the large-scale loss of wealth all around.

In any case, I think you’re changing the subject. The question was of where Coasean bargaining becomes extortion. I’m still waiting for you to differentiate the two and explain why you don’t take your arguments seriously enough to apply them to murderers, who are in exactly the same position as an errant RR that likes to torch crops.

77. 77 77 Steve Landsburg

Silas:

The point remains that IF you grant the RR the legal right to burn crops, then they can *deliberately* burn just the right crops to perform extortion.

Except that the example stipulates that extortion is immpossible because of high transactions costs. I am quite unclear on why you keep choosing to ignore this.

78. 78 78 RF

@Ken B
You’re simply not getting the point of my post. There are no “other options”. All the “other options” are analyzed into one option. Either you accept the situation that results from tax, or you accept the situation that results from no tax. Those are the only two choices. Your “other options” are irrelevant. The UNIVERSITY can install guard towers, the UNIVERSITY can build a sidewalk, but those are not GOVERNMENT options. If “we” are the government, then we have only two choices. You are not identifying other choices that WE have; you identifying other PEOPLE who bring with them additional options.

Steve Landsburg said:
“I don’t have the energy to work through your example so I’ll give you a fresh example; you can translate my variables to yours if you wish.”
Ummm… it’s not my example, it’s yours. Remember? http://www.thebigquestions.com/2011/10/03/there-he-goes-again-3/#comment-32718 Every time I disagree with one of your examples, are you just going to switch to another one?

“Now suppose the neighbors can move away, eliminating all pollution damage, for a cost M with the property that D < M < B."
Again, I think that you people are just not understanding my point. In the train example, I identified the cost of the train as min{A,C}. Because that's what a cost is. A cost of an act is the LEAST EXPENSIVE response to that act. When we're calculating a cost, we should FIRST consider all options, and THEN identify the lowest-cost option. Once we've calculated the cost, there are no "other options". "Options" are a bound variable within the cost. All the "other options" have already been calculated into the cost. If moving away is less expensive than staying there and developing lung cancer, then the cost of the pollution is the cost of moving away. You can't simply identify some situation that hypothetically could result from an event, and then declare that that is the cost of the event. That's nonsense. It's logically impossible for the cost of the pollution to be greater than moving away, because the cost of the pollution is DEFINED to be the lowest-cost response to the pollution. If the people are rational actors, then there would NEVER be a situation where M < B, because as soon as moving away became less expensive, they would move away (and if we're not assuming rational actors, then Coase is completely inapplicable, no?). Your entire argument depends on hypothesizing an impossible situation. Your argument boils down to "Suppose I hypothesize a situation where people are acting inefficiently. And suppose I now compare 'the people spontaneously start acting efficiently' to 'the people continue acting inefficiently, and the government steps in and starts taxing people'. The latter can be less inefficient than the former, so taxes don't always improve efficiency."

79. 79 79 RF

As for the whole “paying murderers” issue: the Coasian argument is that each assignment of property rights is equally efficient. Saying “we could have had a different assignment of property rights, and it would have been as efficient” is NOT the same thing as “We can CHANGE our assignments of property rights any time we want, and it won’t result in any inefficiency”. If we’re paying someone not to kill, then we’ve basically said that he “owns” the life of everyone. Then a Coasian solution would have everyone pay him not to kill them, but what would they pay him with? If he owns the lives of everyone, then he owns everything they have, so they have nothing to pay him with. So Coase doesn’t really apply. If only one person has money, it’s rather hard to defend the premise that there are no transaction costs.

As for the trains, it’s not really on point to posit additional railroads adding more lines through the field. Coase says that we COULD HAVE assigned property rights to any ONE of the railroads. He doesn’t say that we can assign property rights to ALL of them.

80. 80 80 Steve Landsburg

RF:

the Coasian argument is that each assignment of property rights is equally efficient.

This is not only false but the exact opposite of the truth.

Please be careful about making inaccurate statements in an authoritative tone that might confuse readers who are trying to learn something.

81. 81 81 Ken B

Silas: You had an exchange with Steve in which he made the point that you seem to be arguing that if there is ONE situation where we would not want bargaining then there is NO situation where we would want bargaining. You deny this allegation but then in your latest response to me you do exactly that once again. My point was about “coasdean solutions” in general and yet you respond as if they were all the murderer case.

82. 82 82 Harvey Galumpky

Steve,

You wrote: “Extortion is impossible because of high transaction costs.”

The British law about sparks and trains that Pigou and then Coase explored has some clauses about who pays when. The law read that farmers pay for spark fires (and the Coase insight was why that makes sense), but, the law also read that railroad pays if the the damage was over 200 pounds, that the railroad pays if the railroad is shown to be negligent, or as set up in the Silas case, that the railroad pays if trains are enhanced to throw sparks extra far.

Without the extra clauses, a train operator with sufficiently deep pockets potentially has a big incentive to exploit high transaction costs that farmers face by running trains that spew sparks extra far onto farmland as Silas posits, perhaps not necessarily to run extra trains Silas suggested, but another possibility would be to buy farmland at a discount. When the farmers move away, one could easily imagine a railroad purchasing the land at a depressed price and then profiting by turning the sparks back down and turning the crops back up.

It isn’t that high transaction costs protect the farmer by removing incentive from the railroad. They don’t. The extra clause in the law, ensuring that railroads act responsibly, is designed to reduce the high transaction costs confronting the farmer. Without the responsibility clause in the law, extortion is a possibility, and there is no reason to believe moving away is an efficient response to extra crop sparks. We can only say moving away is probably efficient if the transaction costs are low and the parties can work out the terms of it themselves. Part of the Coasian insight has to not only include the idea that common law, in this instance farmers paying for RR spark fires, is not always as environmentally inefficient as portrayed by Pigou but also that common law aids efficient solutions by minimizing transaction costs. Coase doesn’t just tell us to be wary of regulation, he tells us that regulation can reduce transaction costs and improve efficiency. It’s a two edge sword, nonetheless, there are two edges.

In the pollution case, the free rider problem prevents collective bargaining with the coal fire generation plant, and the individual has no ability to make a claim because we only know statistically that particulate matter causes lung cancer, not that an individual case can be pinpointed to a specific generator. In such a world of high transaction costs and ill defined rights, there is no reason to believe an efficient solution will emerge. Which is why Coase gave us the essential preconditions. If anything the Coase question is: How do we minimize transaction costs so that efficient solutions can emerge? Coase himself said regulation is a possibility.

83. 83 83 Ken B

An observation. I think Steve has been marvelously patient in this thread, and very generous with his time towards one or two of the more confused posters. It’s clear he cares about helping people get it right, and I expect this shows up on his student reviews.

I do apologize for this unwonted bit of generosity, but it needed to be said.

84. 84 84 iceman

The extortion examples are interesting, but even there I don’t think it follows that no one will farm or eat or there will be a “large-scale loss of wealth all around”. Assuming the RR or whoever it re-sells the land to is a rational being, the ‘highest and best use’ of fertile land may well tend to remain in food production rather than, say, duplicative parallel rail lines. Of course the farmers may get screwed in the process (it does seem we tend to assume they were there first or didn’t move their crops closer to the rails etc.), but this is just a wealth transfer (so Pareto-hyphens don’t seem to apply). I would say the issue there is that a failure to uphold property rights provides an incentive for socially destructive behavior.

@Silas: “So, in terms of whose utility we get to ignore, where do murderers end and coal burners begin?”

Maybe the ‘rational being’ notion provides another point of distinction that can be applied to the extortion cases. I knew I was wading into a “deontological” pool on the murderer example, and that SL has written about the need to account for all utility changes. One of the first things that intrigued me about economics was its reluctance to make judgments about the sources of utility, but I confess I’m more partial to that in cases of “victimless” crimes…like calculating the costs and benefits of the drug war (see The Armchair Economist). Even in the types of “trolley scenarios” outlined in TBQ, we were (I think) deliberating around the sacrifice of one life to save others, not any direct pleasure involved with pushing someone to their death. And at the end of the day if we’re seeking real-world application, our system (which can ultimately determine who pays the Coasian price) does draw sharp distinctions around a presumedly basic right to life.

1. 1 Who Ignores Significant Scientific Advances?
2. 2 (Market = bad, Govt. = good) = false dichotomy « EVOLVE
3. 3 There He Goes Again « Daniel J. Smith
4. 4 Maggie's Farm
5. 5 Stopped Clocks at Steven Landsburg | The Big Questions: Tackling the Problems of Philosophy with Ideas from Mathematics, Economics, and Physics
6. 6 Air Pollution and Alleged Market Failures | Institute for Energy Research
7. 7 Air Pollution and Alleged Market Failures Jobs | Real-time Offshore Oil Rig Job Listing | Offshore Job Guide
8. 8 Air Pollution and Alleged Market Failures | | ENVIRONMENTAL POLLUTIONENVIRONMENTAL POLLUTION
9. 9 DYSPEPSIA GENERATION » Blog Archive » There He Goes Again
10. 10 Regulation as Wishful Thinking « Politics & Prosperity
11. 11 Air Pollution and Alleged Market Failures
12. 12 Quotation of the Day…