Coase, Invariance, and LPE
There’s an ongoing symposium on the Law and Political Economy (LPE) Project’s blog on Alyssa Battistoni’s new book Free Gifts. Both Battistoni’s opening piece, as well as Madison Condon’s piece, begin with Ronald Coase’s 1960 paper, “The Problem of Social Cost”. This piqued my interest because I recently gave a conference talk about the so-called “Coase Theorem” in which I was trying to correct an error that is common, and that lies at the heart of a recent book about climate change by moral and political philosopher Joseph Heath. Let me quickly state the error, and then return to Battistoni’s discussion.
A Coasean Error
Restricting the focus to pollution, the Coase Theorem is the conjunction of two propositions:
- Efficiency Claim: In a world with complete markets, well-defined rights, and no transactions costs, the result of bargaining between polluters and pollutees will be Pareto efficient. That is, once bargaining ends, it will not be possible to make one party better off without making another party worse off.
- Invariance Claim: No matter the initial distribution of rights to pollute and rights to be free from pollution, bargaining of the sort described above will result in the same level of pollution. Give the rights to the polluters (so that pollutees will have to offer bribes to polluters not to pollute), and bargaining will result in \(x\) tons of pollution. Given the rights to the pollutees (so that polluters will have to pay a fee in order to pollute) and bargaining will result in \(x\) tons of pollution.
The Efficiency Claim is not controversial. It’s basically another way of stating the First Fundamental Theorem of Welfare Economics. One can accept it without accepting the logically distinct view that pollution policy ought to aim at Pareto efficient allocations (or to any given efficient allocation), or that the status quo distribution of legal rights to pollute are ethically defensible.
The Invariance Claim, however, is false. It holds only when an individual’s marginal willingness to pay for abatement at any given pollution level does not increase as they become richer. But it is a standard result in environmental economics that environmental quality is a so-called “normal good”—as people become richer, their willingness to pay for a marginal unit of pollution abatement increases (compared, that is, to what they would be willing to pay for the same change in environmental quality if they were poorer).
Regarding this problem for Coase’s Invariance Claim, Steven Medema, probably the world’s leading authority on the Coase Theorem, writes:
Coase, for his part, later waved aside these objections on the grounds that income effects “will normally be so insignificant that they can be safely neglected,” but “normally” is not sufficient to rescue a “theorem”. (p. 1065)
A very nice 2001 paper by Richard Howarth drives this point home in the context of climate change. Howarth uses an overlapping generations model that is calibrated to the parameters and functional forms of William Nordhaus’s DICE model. Howarth’s model enables one to explore the results of bargaining between different generations when rights to pollute and rights to be free from pollution are allocated in different ways.
Howarth finds that when using Nordhaus’s standard parameterizations, the longrun stock of carbon dioxide in the atmosphere, as well as long run temperature increase, are invariant to the initial distribution of rights. This is Coasean invariance in the wild (or, if you like, in a wild model). Thus, regardless of whether the present generation is endowed with the right to pollute as much as it wants (what Howarth calls the “Pollution rights” scenario), or whether future generations are endowed with the right to be free of all pollution if they want (the “Climate rights” scenario), the amount that each generation will emit after intergenerational bargaining ceases will be the same.
However, Coasean invariance disappears when the model is changed to reflect the possibility that climate change will be more harmful than (that vintage of) the DICE model assumed. When the model is programmed to treat a 3 degrees C increase in global mean temperature as being ten times worse for gross world output than the baseline DICE calibration, invariance no longer holds:
In a “Climate rights” scenario, present-day polluters must pay future pollutees for each ton of CO₂ they wish to emit. Suppose they do this by paying into a trust fund that will offset the damages. The trust fund in turn increases the capital stock, which increases pre-damage future GDP. Higher future GDP means that future generations are wealthier, and since greater wealth leads to a greater valuing of environmental quality, future generations value it more highly in monetary terms. As a result, they charge higher prices to present polluters who would like to buy rights to emit. In Howarth’s “High Damage” setting, these forces conspire to radically lower both the pollution level and temperature increases in the Climate rights scenario as compared to the Pollution rights scenario.
From what I can tell, Coase’s invariance claim is never questioned in Battistoni’s book, and it is not questioned in Heath’s book either. Indeed, invariance underlies Heath’s central claim that one can rely on (Pareto) efficiency considerations alone to determine “whether or not to regulate, and if so how much” (p. 202). Howarth’s analysis shows that, in at least some contexts, there is no single pollution level that is “the” efficient level, since the level that would result from frictionless bargaining is in part a function of how rights are initially allocated.
Coase 4 LPE?
I am neither a scholar of law and economics nor a scholar of the progressive competitor discipline of law and political economy (LPE). So I freely admit that I know precious little about the impact Coase’s 1960 paper had on the development of law and economics and on scholars affiliated with it. But this week I’ve read Coase’s paper carefully twice, and I noticed something that I was surprised not to see mentioned in Battistoni’s writing. It is something that seems only very rarely mentioned in the secondary literature on Coase, but which I think is worth flagging and pondering.
Relatively early in his paper, Coase writes:
The economic problem in all cases of harmful effects is how to maximise the value of production. (p. 15)
Coase repeatedly declares that “the economic problem” with externalities is that they consist in a failure to “the value of production”. Virtually always he means that externalities arise because parties cannot enter into mutually advantageous bargains, which in turn means that net benefits for each party are left on the table. Of course, this would not happen in a Coasean paradise in which markets are complete, rights are well-assigned, and bargaining can take place without friction or incomplete knowledge. But we don’t live in that world, and Coase knows it. Indeed, he spends the bulk of his paper asking what should be done when “maximizing value production” cannot be left to the magic of the market either because rights are not well-assigned or well-enforced, or else because of some other impediment to effective bargaining.
It’s in this context that Coase takes aim at Pigou, who (Coase claims) reflexively assumed that rights not to be harmed should be given to pollutees and thus that polluters should be made to pay a fee or fine every time they wished to pollute. Coase says (to paraphrase), “Not so fast.” Given all the imperfections of the real world, it’s at least possible that one could better “maximize the value of production” by legally permitting polluters to pollute. Maybe, for example, it’s much cheaper for pollutees to move than it is for polluters to pay a fee, so that on balance things are better when pollutees must bear the costs of pollution, rather than polluters.
But there’s a subtle shift here. Back when the topic was the Coase theorem, “maximizing the value of production” meant “exhausting mutually beneficial bargains”. In the second part of the paper, however, it takes on a more abstract connotation; there, it involves the general idea of “weighing up the gains that would accrue from eliminating these harmful effects against the gains that accrue from allowing them to continue” (26). And while Coase continues to articulate quantitative examples in this second part in terms of comparing monetary gains with monetary losses, he concludes with this:
In this article, the analysis has been confined, as is usual in this part of economics, to comparisons of the value of production, as measured by the market. But it is, of course, desirable that the choice between different social arrangements for the solution of economic problems should be carried out in broader terms than this and that the total effect of these arrangements in all spheres of life should be taken into account. As Frank H. Knight has so often emphasized, problems of welfare economics must ultimately dissolve into a study of aesthetics and morals. (p. 43; my emphasis)
Yet this runs contrary to Battistoni’s suggestion that one of Coase’s complaints about Pigou was that “Pigou had imbued the positive science of economics with normative evaluation” (p. 125). And from what I can tell, Battistoni’s reading of Coase is inaccurate. Coase’s complaint was that Pigou was a one-trick policy pony (“Tax, tax, tax!”), not that “Pigou’s analysis relies on an unspoken and unjustified moral framework” (p. 127).
In any case, the passage of Coase’s that I quoted just above would seem a perfect one for Battistoni to highlight as she presses her driving concern that:
The problem comes when the “optimum amount” of pollution and the appropriate level and distribution of social cost are decided largely as a function of value production, under conditions of class and market rule, articulated with other forms of domination … (p. 143)
I’d be interested in how (or whether) her use of Coase’s ideas would change if the passage I’ve flagged were to figure in her discussion.
If the Invariance claim is false; and if the proper approach to externalities in the real world is to evaluate them and their policy responses by “examin[ing] the effects of a proposed policy change and…decid[ing] whether the new situation would be, in total, better or worse than the original one” (p. 43); and if this deciding should be responsive to the best ethics rather than to some narrow economic conception of “value production”; then in at least some nearby possible worlds “The Problem of Social Cost” kickstarted a very different intellectual tradition than it did in this world.
Still, it’s not hard to understand why it had the effect it did here, given that certain powerful and vested interests find quite congenial the narrow economic conception of value production that Coase routinely employed. And to be fair, “The Problem of Social Cost” is a long paper that is not much more fun to read than the corresponding chapter in Pigou’s The Economics of Welfare. So at least some readers can be forgiven for not getting to (or not being awake enough to register) Coase’s remark about the essential role of ethics near the very end. But possibly it could be interesting to ponder what might’ve happened if early eagle-eyed readers outside of the Chicago School orbit had stressed and explored that remark in contributions to the field that became law and economics.