The Burned Books Fallacy

As I said in earlier posting, seeing if a concept is transferable from one kind of spontaneous order to another can help highlight an idea's absurdity (or its brilliance).

Two things I want to look at are the broken windows fallacy, which Paul Krugman and many other Keynesians have in fact been promoting as true (and which Keynes himself argued to be true), and stimulus money.

Let us start with the idea of stimulus. Let's say that you wanted to stimulate literary growth, that you wanted more literature produced. What is the best way of going about this? One can try to stimulate either supply or demand. The former would put the power in the hands of the government doing the stimulating; the latter would put the power in the hands of the average person. Let us look first at the former.

The government can either try to stimulate literary production in general, or engage in more targeted production. If the government tried to do the latter, you would have politicians from the different states insisting that their writers were the best, and should therefore receive the most stimulus dollars (let's say, for argument's sake, that the government is going to pay a certain amount per page of writing produced). If you decided to just rely on the NEH, you would have money going primarily to those the NEH liked, and those are writers who are not necessarily who the average reader would like. Or even who the average professor of literature would like. Of course, this doesn't even address the problem of quality, even among the best. Toni Morrison consistently creates great works of literature, but if you were paying her by page produced, I have little doubt that over time her quality would decrease. If you then want to avoid all of these problems, maybe just paying everyone to write would be the best way to stimulate the growth of literature. Wouldn't the fact that more people are writing increase the likelihood that more good works are produced? Not necessarily. If it is likely Toni Morrison would produce fewer quality works, what makes anyone believe anyone else would? The result would be a writing bubble of very low-quality work, with both the average quality decreasing and the highest quality decreasing as well. As a result, stimulus money would in fact have the opposite result intended. In the first, most likely scenario, those already writing would get the money and continue writing -- to the extent they increased production, it would be to the detriment of quality. In the second scenario, the desired result of more literature produced wouldn't be accomplished, as what we would get would be mostly large quantities of garbage. In fact, it would become increasingly difficult to find any actual quality.

Now suppose instead that the government decided to give the consumer the ability to purchase what they wanted. We can suppose that they mail out $1000 debit cards that can only be used on books written by living authors. Here at least we would get some idea of what people want. But such a program would in fact be pretty similar to the Cash for Clunkers program, which didn't boost long-term demand for cars, but only boosted short-term demand. There was then a slump in car sales immediately after the program was over. Again, those who already have books out would benefit, and new writers (entrepreneurs in the economy) would receive none of the benefit. In writing as in economic activity, it is often the upstarts who are driving innovation. Thomas Pynchon's great novel was Gravity's Rainbow, released in the 1970s. He is still writing now, but everyone agrees that that novel is the best he's ever written. Stimulus money for literature would go not to the person writing the great novel that will be coming out next year, but to authors who have already been established. Those already established would benefit, not the innovators. So while it may be true that stimulus money given to the consumers would be better, it's still going to create distortions favoring the status quo.

Now let us look at the broken windows fallacy so popular among Keynesians today. This is, to apply it to literature, the theory that if we just had a big book burning, that would stimulate people to buy more books. Or maybe if Thomas Pynchon, Don DeLillo, Toni Morrison, Frederick Barthelme, and Cormac McCarthy were to all die in a plane crash during a hurricane, other literary writers would have a chance to be noticed. Both ideas are idiotic. It does not benefit literature to lose any of these people. If they write more, that will increase the quality of literature, making it more likely that more writers will write better works. If anyone is stimulated to buy more books, in the first scenario, that is taken away from future (and different) book sales, creating the cash for clunkers problem mentioned above. And getting rid of the best authors in the country will do nothing to make anyone else any better. (Yes, the book burning example is closest to the broken windows fallacy as typically conceived; the latter example merely takes this idiotic idea to its logical conclusion.) And Krugman's latest idiotic idea, as reported here, is that more regulations will actually result in economic growth. This is like arguing that if we only forced all writers to write everything in iambic pentameter, we would get more writing. What will really happen is many writers will go find something else to do.

In regards to that earlier posting, the concept of labor being fungible may in fact be more Marxist than either neoclassical or Keynesian, though the fact that when you treat labor (or capital) as an aggregate, you necessarily treat it as homogeneous and, thus, necessarily treat it as fungible. The math necessarily treats labor or capital as perfectly interchangeable, even if the author then goes on to elaborate that they are, in fact, not. However, how can one come to a right conclusion when one necessarily starts with either a false foundation (in the math) or from contradictory premises? So the average Keynesian may not explicitly argue that labor and/or capital is fungible, but insofar as they are using equations with labor or capital in it, they necessarily base their economics on it.