One of the recurring themes of my Dorf on Law posts has been the meaninglessness of the concept of "economic (or Pareto) efficiency." I have attempted to explain why the claim that economics is a positive science -- which carries with it the claim that one can make neutral claims, based on so-called economic reasoning, free of the taint of mere normative moralizing -- is ultimately wishful thinking, at best, and venal deceit, at worst.
There are several ways in which this broad point can be made. The most obvious is to analyze the theoretical structure that supposedly backs up the claims of scientific rigor in economics. It is often useful to analyze the assumptions (especially the unspoken ones), but even better, it is helpful to see how the standard theoretical structure is itself based on naked moralizing. As a colleague once put it to me, in summarizing a law-and-economics paper that attempted to justify the simplistic version of utilitarianism on which so much (bad) scholarship relies: "The authors aren't just arguing that everyone should agree with them about utilitarianism. They're arguing that your mortal soul is at risk, if you don't believe in utilitarianism." Of course, the authors in question never mentioned morality, and they surely would have been deeply offended by the suggestion that they were closet moralists, but part of the deep (valid) critique of mainstream economics is that it is a thinly veiled normative case for leaving the distribution of wealth alone.
As satisfying (and even, I am hesitant to admit, fun) as it can be to play with the theoretical underpinnings of economics, most people have not studied economics at the level necessary to see why the devastating arguments against mainstream theory are so devastating. It is necessary, therefore, to offer telling examples to show that "economic thinking" does not provide useful guidance in thinking about policy questions. Because of the underlying emptiness of orthodox economic theory, all applications of that approach to real-world questions end up providing no useful guidance to policy.
Recently, for example, I noted a vivid example regarding the relative costs of imprisonment. Harper's reported that the average cost of holding a prisoner at the Guantanamo Bay prison is $900,000 per year, while the cost of holding a prisoner in a SuperMax prison is about $65,000. Even so, the "economic case against Gitmo" is hardly a slam dunk. Anyone who wants to defend the continuation of Gitmo can simply say that there are unmeasured benefits of keeping terrorism suspects out of the United States. Their opponents can then claim that the cost of Gitmo should include the cost to our global reputation, which can manifest itself in dollars-and-cents costs (heightened security needs, rebuilding targets of terrorism) as well as other important costs. There is no right answer, at least as provided by an economistic approach.
I came across a similar example about twenty years ago, when I read an article by two professors (who were, I think, on the faculty of a military college), who had worked out a cost/benefit analysis, comparing the military academies with the alternative institutions that train future officers. (For the latter, think of the movie "An Officer and a Gentleman.") They found that the non-academies produced graduates whose careers were generally more successful than academy graduates, as measured by ultimate rank achieved and so on. They also found that the cost of training someone at West Point or Annapolis was something like ten times the cost of training someone in a nameless military training program.
Again, a slam dunk, right? Hardly. The publication that ran the original article soon ran a series of letters, one of which was especially notable. There, an academy graduate argued passionately that the benefits of running the traditional academies could not possibly be measured in the cramped way that the pointy-headed professors had adopted. The continuation of the Long Gray Line, the tradition of the Army/Navy game, the connection to the past, were all important -- and, by implication, presumptively greater than any cost.
At the time, I scoffed at that argument. It struck me as an example that liberals can use to say that "economic reasoning" provides a solid answer, whereas mushy-headed sentimentalism caused people (certainly including politicians) to continue to support the wrong policy choice. But of course, the truth is that there is no right or wrong policy choice here, either, based on this kind of reasoning, at least as an objective matter. Once again, the outcome is determined by what one includes and excludes.
An even clearer example of this problem showed up recently in the op-ed pages of The New York Times. A few Sundays ago, the ethicist Peter Singer, in "Good Charity, Bad Charity," provided a classic economistic case for the proposition that some charitable donations are clearly morally superior to others. Singer walked readers through a numerical example, comparing money that could either be donated to build a new wing on an art gallery, or to reduce the incidence of an eye disease that afflicts children in poor countries. He concluded by saying that, even putting extremely unlikely weights on what people care about, the donation to reduce disease must be morally superior to the donation to help display art in a more pleasing way.
I confess that I found Singer's article more than a bit appealing. That is hardly surprising, of course, because even though I am recovering from my training as an economist, I am still an economist. Singer's brilliance was in making it seem that there are some questions to which the answers are so obvious that it really does not matter what assumptions one makes, within the broadest range of reasonable possibilities. There really is a right answer.
Again, the letters flowed in. The Times printed a few, and the first two struck me as particularly important. The first letter-writer pointed out that (like most such economistic arguments), Singer's method of formulating the question created a slippery slope, providing no principled stopping point for guiding charitable giving. Singer's approach ultimately supports the mindless people who fault high-profile philanthropists for not giving all of their money away. At the very least, it argues for a strict hierarchy of giving, which will result in zero giving for less morally worthy charities, so long as the money available for charitable giving is finite.
For my purposes here, however, the most important letter was the second one, which reads:
The point is that a proper cost/benefit analysis must take into account the cost of diverting funds from "better angels"-producing charities. I sincerely doubt that the letter writer would be comfortable recasting his argument in such crude cost/benefit terms, but that is actually the point. Even casting anti-economistic arguments in economistic ways still undermines economistic thinking.An effective attack on the scourges that Peter Singer cites, and so many other endemic diseases, will be mounted only by an ethical society with an educated, broadly aware populace. It is a society’s cultural institutions — its libraries, museums, universities — that provide the foundations for such a society. A rich cultural heritage enables us to develop the “better angels of our nature.”Can anyone doubt that Andrew Carnegie’s gifts to establish a system of public libraries have been far more meaningful to society than if he had given that money to fight a particular disease?It is not a choice of giving to one or the other; neither is a “bad charity.”
In this case, the problem is not just unmeasured costs and benefits of Gitmo or military academies, but the unmeasured costs of failing to "produce moral people." Singer presumes the existence of people who even care about other people, which is the necessary condition for mainstream economists' claim that they can provide the positive analysis, and the normative analysis is someone else's job (actually, everyone else's job). If we do not invest in our precious stock of moral people, however, we will have no idea what to do with the output of economists' reasoning.