July 2014

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Blog powered by Typepad

« Chris Coyne on Doing Bad By Doing Good (March 2014) | Main | Teaching Economic Transitions »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Using a specific example...let's say that the show Firefly was prematurely canceled. Resources were diverted away from the show even though there was sufficient demand. The problem was that the demand was "latent". It was only revealed after the show was canceled. Many people bought the DVD and campaigned on behalf of the show.

As I've argued before...price theory is too narrow. There are certainly situations where we can correctly determine Hayek's "solution" without using prices. The producers of Firefly could have clarified the demand by creating a crowdfunding campaign. This would have allowed each and every fan of the show to decide for themselves exactly how much they were personally willing to contribute/sacrifice in order to keep the show alive. There wouldn't have been one price...there would have been a continuum of "prices". People would have been paying vastly different amounts for the same exact product. It's the same thing with the non-profit sector.

Price theory, as it stands, is not a great prophylactic because it's got a giant hole in it. It doesn't account for the vast majority of situations where we sacrifice in order to try and keep something alive. A better theory would be something like "input theory"...or "positive feedback theory"...or some other better name. Maybe "flowcilitation theory"?

"The central conception of price theory is that of an equilibrium adjustment with respect to relative prices and relative production."

This statement from Henry Simons does not mean what some Austrians might think it means. It simply means comparative statics.

In the old days dynamics meant changes in the data (Simons's second kind of change). However, this was not explained; these were simply posited as shocks.

Simons did not teach anything different than completely static price theory, in my view. Of course, this is good enough to prevent people from making many errors. But it also produces Simons's positive program for achieving perfect competition -- a bad error in its own right.

What Buchanan learned from Knight was the standard stuff prior to Samuelson, a critical attitude of mind and an opposition to behaviorism and positivism.. The process ideas of the later Buchanan did not come from Knight, Viner or Simons. They came from Hayek and Shackle.

Alchian is a mystery to me. Coase has the LSE cost-tradition background but really he was a "student" of Edwin Cannan and Arnold Plant who stressed the application of price theory rather than its formal development.


According to mainstream economics, price is determined by the negative feedback of both supply and demand interacting with one another. As the supply of a good goes up relative to demand, the price goes down; as demand for it goes up relative to supply, the price goes down.

Underlying the supply of a given good, there are producer costs affecting how much will be produced. The law of diminishing returns (negative feedback) says that you will get less and less value out of something over time. You can only get so much work out of people, at incrementally lower levels. You can only get so much work out of a machine. Crops can only be grown at a certain 2-D density.

Underlying the demand for a given good, we have the law of diminishing marginal utility (negative feedback), which affects how much demand one has for a given good at a given time. I will likely pay more for the first apple than for the second. I will probably not buy more apples than I can eat before they rot. Thus, my demand for a given good decreases the more of that good I have.

Again, this is equilibrium economics. From this we get our supply and demand curves, and from their intersection, we get the market clearing price. This is the equilibrium price.

The problem is that with only negative feedback, we have only half the equation. What if there is positive feedback? Let us look at what would happen in a positive feedback-only economy.

The positive feedback version of the law of diminishing returns is the law of increasing returns. How can we have increasing returns? If the marginal cost of producing something decreases with each unit created, we would have increasing returns. Software is a good example of this. So would be e-books (paper books abide by the law of diminishing returns). However, there is another example of this outside of knowledge/information-based products, and that is the agglomeration economies of cities. The existence of spillover effects and mass production creates increasing returns (at least for a while, in the latter case). Thus, increasing urbanization and increasing virtualization result in increased positive feedback.

The positive feedback version of the law of diminishing marginal utility is the law of increasing marginal utility. While it is true that individual consumers must necessarily abide by the law of diminishing marginal utility, groups do not. Fads cause more and more product to get purchased. Of course, there are various mechanisms that can drive a fad, from those that drive fashion fads to those that drive things like housing bubbles. The former may be accurately described by things like Keynes' "animal spirits," but the latter likely have underlying mechanisms more accurately described by Austrian Business Cycle Theory. Whatever the case, if you can get a lot of people buying something, you can get positive feedback.

As pointed out above, negative feedback results in equilibrium. Positive feedback, however, results in boom-bust cycles and multiple equilibria. If you had a positive feedback fire, it would get hotter and hotter over time, until it suddenly became too hot and burned itself out. This is the description of a bubble, be it a housing bubble or a fad.

But note that negative feedback models, which give us the equilibrium theory of prices, is based on viewing people as individuals. The individual producer has diminishing returns; the individual consumer has diminishing marginal utility. However, if we view the economy as being a group of people, producers can have increasing returns and consumers have increasing marginal utility. This can cause one to come to the conclusion that a future of abundance is possible (and if an abundance economy is possible, socialism is possible). Such people would have to ignore the fact that increasing returns are not infinite, and in fact result in not just boom, but bust as well. But one can nevertheless see why they would make such an error.

However, if we understand people as being what they in fact are, both individuals and social, we have to consider both negative and positive feedback processes simultaneously.

With both kinds of feedback, supply is determined by both diminishing and increasing returns, demand is determined by both diminishing and increasing marginal utility. Both interact. Thus, the supply of a good cannot reach an equilibrium point, but must always fluctuate in ways that are incalculable and unpredictable. For the same reason, demand cannot reach an equilibrium, but must always fluctuate in ways that are incalculable and unpredictable. Thus, we would not have the smooth curves we see when we draw supply and demand curves, but lines that would instead look more like the graph for a given stock over time. Worse, these two complex processes then have to interact to create the price of that good. Thus, price is inherently incalculable and unpredictable. Without the ability to calculate what the price "should" be, if there is not any such thing as a market clearing price, dreams of even market socialism must fall away.

Bipolar feedback drives a large number of complex, creative processes. The economy is but one kind of such a process. As we can see, we have bipolar feedback (increasing vs. diminishing returns/marginal utility) underlying both supply and demand, which are themselves bipolar feedback processes. This is complex, creative process on top of complex, creative process. The economy thus becomes increasingly incalculable, the more such processes we realize are part of the economy.

This discussion is very interesting in light of some key passages from Knight’s (1924) “The Limitations of Scientific Method in Economics”:

“…needless confusion has been caused by the unfortunate use of the terms “static” and “dynamic.” These are highly objectionable because they have in mechanics a definite meaning unrelated to the main issue in economics. The problem of conditions of equilibrium among given forces – “statics” in the proper sense – is often important in economics, but is after all subsidiary, as indeed it is in physical mechanics. The larger question is that of whether the forces under given conditions tend to produce an equilibrium, and if so how, and if not what is their tendency; that is, it is a problem in dynamics. This type of problem has been too largely passed over hitherto, leaving a fatal gap in the science. The crying need of economic theory to-day is for the study of the “laws of motion,” the kinetics of economic changes.” (pg. 133 in the The Ethics of Competition Volume).

Mario, this seems to me to be quite different from your interpretation of Simon’s view of dynamics. Also, this passage comes after three important discussions. One, a treatment of economic theory and applied economics that reads very much like Mises’ distinction of theory and history. Two, a sustained critique of the data of human action as a basis for a predictive science. And three, a discussion of the subjective nature of the data of human action and the necessity for the data to be communicated accurately for intelligent thought to be possible. This included what looks like an account of equilibrium as perfect intercommunication between buyers and sellers, and frictions being aberrations in the communicative process on pg (130):

“Other considerations produce aberrations from the result of perfection intercommunication among buyers and sellers who know the goods and their wants, but they are to be understood only as producing aberrations from the fundamental tendency and hence in subordination to it. They are of the nature of friction…”

I find it hard to read this and not see the seeds of a very subjective and process-based view of economic activity. That said, I’m not claiming that this is the Knight that Buchanan received, as I am under the impression that Knight’s thoughts on this changed substantially. On this point however, I am treading out of my depth. If someone would care to elaborate on these differences in Knight and on the Knight / Buchanan relationship - I'd be quite interested. Perhaps there is something of interest in Buchanan's archives on this?

But there is one more curiosity in this Knight.

“If the term economics were to be interpreted in the literal sense, as covering all behaviour which involves the adaptation of means to ends and the “economizing” of means in order to maximize ends, then economics would be an almost all-inclusive science.”(pg 132)
…”But the study of this vast field falls into certain natural divisions….”
[skipping to the next paragraph]
…”The most important division is that between political organization, on the one hand, …. And on the other hand what has come to be called the economic organization.”

Knight then goes on to suggest that economics has in practice come to be restricted to the study of the latter. This would seem to imply that an economic approach to the study of politics is a logically unexplored area of study. As such, this stands as a very early Chicago argument for economic imperialism of which public choice would then be a subset.

Pete, I don’t however think this provides a deep complication to your distinction between the Knight/Viner/Simons tradition of Chicago Price Theory from the Friedman/Stigler/Becker version. Knight’s view of economic actors and the process of choice is categorically more human than what will come to be the view of Friedman/Stigler/Becker’s agents.

I had a similar reaction to Mario's (above) regarding the statement:

"The central conception of price theory is that of an equilibrium adjustment with respect to relative prices and relative production."

But I was more intrigued by the two-pronged approach to describing price theory as focused on:

"(a) the arrangements under which the system will be in equilibrium and (b) how departure from the equilibrium arrangements will set in motion forces operating to restore equilibrium"

This is, I would argue, exactly the opposite of Misesian price theory. The final state of rest (FSR, "equilibrium") is an imaginary construction that we might wish to use in order to analyze the market structure and the free (though imperfect, and thus disequilibrium) allocation of resources. But the point core to Austrian price theory is anti-(b): the FSR is neither reached nor fixed. It is an ever-moving imagined "target" that only serves as a means for analysis. It is not the "departure" from FSR that sets in motion processes of equilibration, but the market as a process continuously shift toward the moving target of FSR.

I would agree that (b) perhaps sounds like a disequilibrium market process (if we for a moment disregard Mario's comment), but it is more of a tatonnement correction (or, perhaps, a Schumpeterian post-destruction equilibration) than a process in constant disequilibrium. I don't think this is splitting hairs, but actually a fundamental difference in perspective with implications for the approach, methods, and implications of one's thinking. Simons starts with equilibration, then considers deviations from it and (with Mario) calculates the characteristics of the new equilibrium. An Austrian would start with disequilibrium and the market's crude price approximations through entrepreneurial bidding, and then use imaginary constructions such as PSR, WSR, and FSR to explain the real prices - and the foreseeable effect of endogenous change and exogenous distortions. The latter cannot be calculated since there is no fixed point and all we can measure ("data") are price approximations under changing preferences and varying and dispersed information.

Come to think of it, I'm not sure I see a theoretical similarity that goes deeper than the use of similar-sounding wording.

My statement about the difference in meaning of "dynamics" in the earlier part of the 20th century and in the latter comes from Roy Weintraub (I think). It is certainly obvious in John Bates Clark.

Knight's use of the term "dynamics" in 1924 seems out of step with the usage at the time. But it does not matter much because Knight identified *scientific* economics with statics. Beyond that we are in a world not completely amenable to scientific analysis.

But the proof of the pudding as far as Henry Simons is concerned is to look at what Simons taught -- standard static stuff. And I reiterate that he would not has his positive program for laissez-faire if he were not a statics guy.

I admit that the statement from Simons "how departure from the equilibrium arrangements will set in motion forces operating to restore equilibrium" sounds like process but it is not. This is just the old Marshallian thing about stability of equilibrium. I am saying that Simons was probably committing the error of mistaking stability analysis for process analysis.

All in all, I do not think we are going to find price theory "treasures" in the old Chicago School.

Mario: As you probably know very well, Alchian thought he was criticizing both Lester and Machlup in his famous 1950 paper. Under Machlup's influence, Penrose wrote an article criticizing Alchian. Hayek, by way of contrast, liked it. In the hands of Fritz Machlup, Austrian price theory was about weak rationality and not system-level filters. Today, of course, we think to these two types of argument as complementary. Later, Becker made an Alchian style argument without realizing that it depended on the existence of *some* "rationality" *somewhere* in the system. Kirzner criticized him without, as I recall anyway, acknowledging the how much truth there really was in Becker's argument.

Pete,

Simon's vision of price theory shares a striking similarity to that of Walras. In his Elements Walras adopted a two-pronged approach to price theory - to first spell out, in each of his models, the "theoretical" or "mathematical" solution, i.e., the equilibrium solution given the underlying data, and second, to trace out the "practical" solution, or the process of tatonnement,so as to verify that the realized prices on the market correspond to or for all practical purposes are identical to these equilibrium prices. Simon's point (a) corresponds to Walras' theoretical solution, whereas his point (b) more or less corresponds to Walras' practical solution.

The price theoretic vision sketched out by Menger and then developed by Bohm Bawerk, Davenport, Fetter and Mises (among others), in my understanding, is radically different. The focus in this tradition is on explaining realized prices as they are formed from moment to moment on the market. There is, in other words, no divide between statics and dynamics in this system; no analysis of prices as they would be in a hypothetical static state (or ERE) followed by a proof of the fact that real world prices tend to approach these static prices. Instead, all price theory is dynamic; an analysis of choice in a world of unceasing change. This, I believe, is what Mises is getting at in his first quote.

The role of the long run equilibrium construct (a world of non-action as Mises terms it) in this Mengerian tradition is one that is purely conceptual and methodological. It is a tool used by the economist in his endeavor to explain realized prices and has no empirical relevance.

So I guess, in sum, I would argue that the Mengerian/Misesian vision of price theory differs fundamentally and radically from the one spelt out by Simons.

Alchian was the link between Vienna and Chicago. We pretty much established that at a recent Liberty Fund colloquium on Alchian. Henry Manne, who couldn't participate, has detailed the Alchian/Mises connection.

Alchian's point in his 1950 article is not to deny rationality, but merely to argue that a system of profits and losses ensures that actors will move in the right direction. It is not so different from Mises' weak form of rationality that men have purposes.

Jerry gets credit for seeing one of the key bridges I would contend, but I believe the conversation got derailed in other comments. I never claimed Simons was an Austrian market process theorist, but I also think the Mises quote (second one) shows something that is missed by some of the word games being played in some comments ... Do prices tend toward costs? Do we see the equimarginal principle at work or no?

And the "quiz" is intended to provoke papers, not blog comments so all those making comments write them up into a journal article :)

The comments to this entry are closed.