It all seems to come down to whether it is form over function or function over form. In the short run, with proper analysis, function can be described in terms of form, but in the long run, will not serve. Hence why those who operate under the illusion that their understood form is the pinnacle of understanding will use force to make sure it continues. Then again, there are those who perceive a passing of the old form with the next emerging construct, and fight for its implementation, flaws and all, on a, as yet, unprepared function.

This pretty much encapsulates the conservative/liberal dialectic, as the “liberal-progressive” becomes the next “conservative-establishment”. The net result is ham handed passing of the baton by Statists, the ultimate reckoning of the two in their endeavors to use force on the masses.

What is a fourcast?

To some extent I agree with Callahan’ observation of how obsessive the neoclassicals have really become with mathematic apparatus, yet his own vision of the proper method in economics is actually no better and leads nowhere. In fact, what he proposes is not at all economics, but uncritical and essentially naive play with words.

No one doubts that it’s individuals who act. Yet the most important and interesting problems in economics go beyond this basic fact. Take the theory of value, for example. It’s clearly not enough to argue, as Callahan does, that it’s “creative, intelligent agents” who by their purposive actions influence the height of market prices. As an economist, your task is to explain the structure of those market prices, not simply to point out that they’re products of human action. The questions to be answered are, for example: what factors influence the formation of prices? What is the exact nature of the obvious interdependencies that exist between the demands for various consumers’ goods and factors of production, for instance? In order to derive eternal economic laws you have to abstract from the concrete content of individual human actions, to take this content as given, and proceed to analyse the exact nature of these interdependencies.

It would also be of no help whatsoever to argue that prices are formed according to the considerations of marginal utility. That it is according to the law of marginal utility (together with the human action axiome, to be sure) that the schedules of demand and supply are formed and, therefore, the market prices, is simply not true. Virtually no producer attaches any utilities to his products. The only thing he cares about is the money prices he expects to obtain for the sale. But how to explain those money prices? It’s obvious that one of the two essentials elements which are supposed to determine the market price cannot be applied in cases where products are valued not according to their direct marginal utility. The basic demand/supply analysis can provide a reasonably good picture if applied to primitive conditions of a barter exchange economy. If I have a cow and attach direct utility to the services the possession of the cow gives me (I’m an avid milk drinker), but I also can evaluate this utility against the utility of the services of a horse, say. Here, if we imagine an individual with a reverse order of preferences, we may reasonably assume that a possibility for a mutually advantageous trade exists. Entirely correct! And you may even analyse the whole thing using marginal utility reasoning. But this kind of reasoning breaks down completely if the seller, say, doesn’t attach zero utility to his products. In this case, how one’s supposed to arrive at a watertight explanation of market prices? By assuming the money prices as given, or what?

I argue that Callahan’s method to look at human action alone and stay there, does not help us any further in our efforts to understand the complex pattern of market interdependencies. Neoclassicals, at least, try to construct general equilibrium models in order to think through such problems. Callahan and Austrians want to freeze economics before it gets really interesting and useful.

Callahan’s vision of proper economics is to freeze economics at the most primitive stage. It is not enough to say that “[p]eople act in the market in order to profit, in the broadest sense of the word: they exchange because they feel they will be better off after the exchange than they were beforehand.” It’s clear that they want to profit but the problem is: how can we utilize this piece information for our analysis of prices.

Bob

Callahan’s article did cover the bases, but it’s likely to fall on deaf ears if brought to the attention of mathematical economists. Here’s why:

“Of course we work with cause and effect! Whenever the econometric equations are out of balance, it means that a causal sequence is in play which implies a transition from the old equilibrium to the new one.”

And then, the math lesson begins. The neat-o part of the contention in quotes above is that, once the Austrian critique is brushed aside, the brush-aside itself serves as a prelude to the process of solving the equations themselves.

“Okay. Let’s assume that we’re analyzing a market out of equilibrium, as so many markets are. [No need to check the accuracy of any econometric equation(s) now, now is there?] As the lecture notes show, the equation we’re discussing takes six steps to solve. Before moving on to the next one, you should ask yourself which step corresponds most closely to the typical out-of-equilibrium market you’re likely to be analyzing. How would you decide which step corresponds most closely to the economic reality you’re observing?”

The interesting implication of this kind of answer to praxeologists’ criticism of econometrics is that it throws out the kernel of what was supposed to make econometrics scientific in the first place. If economic reality does not correspond to the equation (so the old mathematical economists reasoned) then the hypothesis which inheres in the equation is untenable. The “mid-stage approach” throws out what used to be seen as verification of hypothesis.

It looks like a kind of compromise peace has been established. Mathematical economists declare that the old econometrics was too “postivistic” and praxeological economists have decided that mathematical economics is just another way of describing the economy. I suppose that the next stage in the compromise peace would be for mathematical economists to let in praxeological economics as “a kind of proto-neural network analysis in disguise” and for praxeological economists to conclude that “mathematical economists implicitly assume cause and effect anyway, and they’re all deductivists at heart too.”

Since Sir William Petty in the 17th century we are still waiting for econometricians (if I spelt that word correctly) to come up with reliable predictions.

Some decades ago we would told that computers would do the job, then it was more powerful computers, then more powerful computers – and it still does not work.

The above is an empirical attack on empiricism (or the start of one) – one would then produce lots of stats to justify one’s conclusion. One would of course produce the stats first and then produces the conclsion (although, normally, the conclusion is written before the “investigation” but people do not like to admit that).

My old economics master at school (what Americans would call High School) many, many years ago was an ex Univerisity lecturer (and a leftist). He made the mistake of doing an econometric study of econometric models – and found that their record of prediction was pathetic (and often fudged after the event to make them look good) – they have not really changed (in spite of better computers), but his study was still the end of his life in academia.

Economics is not empirical. I am not a mathematician so I might be considered biased against number crunching (as I think of it) – but both Mises and Rothbard were very good at maths (especially statistics) and they both came to the conclusion that it was just a different subject.

It is somply not true (as the Econmetrics Society would have it) that “Science if Prediction”.

Natural science may well be – the study of the forces of nature (classically physics). But not all “bodies of knowledge” are like the natural sciences.

Some subjects are about the human choices and the mathematical methods of physics are not suitable for these subjects.

It is (I hope) starting to be well known what a botch their effort to make mathematical methods made of history – a subject concerned with what human beings choose to do and what the consequences were.

Economics and history are different subjects, but in fact mathematics has more of a role in history than in economics.

Indeed most economists (as Mises was fond of pointing out) are in fact economic historians – carefully collecting data which simply tells them what has happened (i.e. the history of events).

They do not know (by collecting data) WHY it happened or what is GOING to happen – and when they play with models to try and find out they are not even being good mathematicians (they are just wasting their time).

“Empirical evidence” can be found to support any theory (no matter how absurd) – remember the “economists” (or rather some people who worked in a University “Economics” department) who found that a higher minimum wage caused more job growth? They compared New Jersey to Penn and found that the State (New Jersey) that had the higher minimum wage had the higher job growth and the lower unemployment “therefore” the minimum wage (indeed a higher mimimum wage) was good for jobs.

I am sure that their mathematics was perfect. And I am equally sure that their conclusion was utterly absurd.

“Empirical economics” is based on stuff like this.

Economics is a logical deductive science (science as a body of knowledge not as a natural science) it is about the logic of human choice – not the mathematical forces of nature. And is based on the fact of human choice (of the reasoning “I”). Errors in economic theories (and everyone is capable of error – including Mises) are to be found by the logical examination of a theory to find flaws in the reasoning. They are not to be found by collecting “data”.

Data (historical events) are stuff that we toss at each other in political debates. “Wages were higher in the United States than in any other country before either unions or government were of any great importance”, “ah but they are even higher now, so statism must be a good thing”.

It is a matter of debating points (which, I freely, admit I indulge in when on blog threads) – not serious economics.

Reductivism (the effort to break things down into things that such subjects as physics can deal with with their equations) will not work with economics – humand beings are BEINGS (not just collections of atoms).

“Market forces” are in fact millions of human choices. Neither predetermined nor random – but chosen.

In short “agency rules O.K.”.

I wish I had a copy of the Landsburg book referenced handy.

Landsburg is telling students that they must not think of prices as being determined by the actions of individuals

That may be what he meant, and perhaps a greater reading of the context surrounding the quote would make it obvious, but just reading the quoted line doesn’t suggest to me a meaning at all resembling that claimed.

cynical said:

I once had an economics professor tell me that he was upset that his students “didn’t seem to care about the economic reasoning behind the mathematical equations he had assigned them”. Then, in the next breath, after he and I had discussed the lack of economic reasoning in time-series forecasting, he just shrugged and said “well, if they seem to work”. Talk about abandoning cause and effect.

What he’s really wishing for is someone with mathematical intuition or insight to take up economics. This wish isn’t very realistic because anyone with such a gift winds up in the mathematics world anyway.

He should watch what he wishes for, though. He might have a future Ann (or Neal) Koblitz on his hands.

Alan Dunn –

the problem with neo-classical economics is not in mathematics (which is generally fine), but in the fact that the fundamental assumptions these models are based upon do not correspond to the reality.

The common (and most damning) assumption is that the utility scores of different actors can be somehow compared, averaged, or summarized.

Remove that assumption and you end up with only qualitative (and NOT quantitative) predictions the austrian economics has to offer – there’s little to be said beyond what is already said.

And, no, the analogy with quantum mechanics is not really applicable – QM yields very good probabilistic predictions, but has a significant problem of irreconcileable incompatibility with the general relativity theory, which also yields very good predictions. There’s also a small fact (which physicists tend to avoid discussing, as it makes them understandably uneasy) that QM allows multiple physical interpretations ranging from ridiculous to mystic. The entire body of work on strings and LQG attempts to resolve these problems, so far unsuccessfully.

I think the way to develop Austrian economics further is to combine it with the evolutionary theory (it already contains hidden assumption of the “natural selection” of unsuccessful economical practices, and propagation of successful ones).

The value of the evolutionary theory is in the fact that it does provide an objective and measurable utility function for the replicators: their survival. It can be argued that economical practices (business models, forecasting methods, technologies, etc) can be considered replicators.

I think one of the biggest problems with all schools of economics is that they consider humans, but not species of ideas, as the actors (this is the very same kind of “wrong level” mistake which plagued biology until the biologists recognized that the biological evolution works on the level of genes, not organisms or groups — and thus finally were able to explain altruistic behaviours, cooperation, and many other things which are still puzzling the economists).

Paul — absolutely, they should.

However, a careful logical analysis of praxeology yields one interesting observation: it assumes, essentially, that there’s evolution among economic actors. Bad ones are not imitated, and their methods die out.

If this wasn’t the case, the range of admissible outcomes would inlcude a populace quite rationally pursuing the goal of self-destruction. Why not? After all, death wish is a subjective utility, and it is conceivable that everyone in some society may have it as their top priority. So a global suicide could be quite Pareto-optimal – everyone gets what they wish. Thus to make sense of praxeology we must state (or assume, as the case is) that there’s some commonality between subjective utility functions – such as survival.

There’s also another, well established, fact, that improving efficiency (including using cooperation) via evolution does not require rational thought on the part of the actors. The fact that altruism and cooperation are common among animals (and even bacteria) – and come entirely from the absolutely unthinking (how can a molecule think?) behaviour of their “selfish” genes, makes one to wonder if rationality is overrated.

The more we learn about how our brains work, the murkier the whole concept of rationality becomes. It seems to be an arbitrary, ever-shifing line in sand delineating what we think makes sense from what we don’t understand yet.

“Austrian economics is the economics of people viewed as creative, intelligent agents.”

Then Austrian economics is doomed because “no one ever lost money under estimating the intelligence of the American public.”

I’m wondering about Gregory Chaitin’s concept in Algorithm Information Theory that For every axiomatic system, there is an infinity of undecidebale propositions?

If true, it seems to me, that the growth of “intelligence”, even seeking models of equilibrium, always results in greater complexity and therefore, greater levels of “chaos”. The attempt to create certainty would result in uncertainty, certainly an aspect of heisenberg’s concept, and bringing to mind Godel’s Incompleteness Theorem. But Godel pointed out that in a sufficiently complex system, the truth of the formal statement could not be proven from within the system.

This would seem to mean that for every axiomatic system that provides equilibrium, there must be a degree of force applied to make it work. If we accept classical econometric models as predictive, it is as much a matter of faith as one’s belief in a complete and consistent God. “God” as the masses accept “Him” is a matter of force through warfare, at the cost of destruction of an infinity of alternatives. Classical economics is no less a matter of faith, and constant attempt to apply the economic model to the “gaps” discovered within that model parallel the “God of the gaps” developed by religionists.

Chaitin’s idea would seem to suggest that axiomatic models are not only incomplete, but infinitely incomplete, making perfectability in the sense of equilibrium unattainable.

“What is truth?” Becomes not a question of eternal form, but of constant and changing interaction. Equilibrium, in thermodynamics, can represent heat death. It would seem that the uncertainty relations literally built into the fabric of our thinking serves to insulate us from the complete equilibrium we desire. Life may be sustained by the simple fact that “we can’t get there from here”.

If an economic theorem cannot be stated without the use of mathematics, it is probably irrelevant. If it cannot be stated by use of mathematics, it is probably invalid.

All this bears on the expression of the concept, not on its foundation. Mathematics is useless for proving or disproving any economic concept, but in a few cases may be helpful for purposes of observing its operation.