Ludwig Von Mises was an Austrian economist of the twentieth century. He died in 1973, one year after I was born. I didn’t know him personally. So I don’t mean my title in the way that Murray Rothbard didn’t like Alan Greenspan:
[See Murray Rothbard, “My Expulsion from the Ayn Rand Cult,” Liberty, 3, 1 (September 1989), pp. 27-32.]
One of the things that Von Mises is best known for is his work developing (at the same time as Hayek) the Austrian theory of the business cycle. This theory points the finger at active monetary policy by central banks as playing a role in creating booms and busts. I believe there is probably a lot of truth to this.
Modern economists, for instance of the rational expectations school, view counter-cyclical interest rate place as a little too costless, and it is quite clear that they do not properly account for speculation. (A major open problem in the rational expectations school is to explain why it is that the volatility in corporate earnings does not account for the volatility in share prices.)
Champions of the Austrian school give important public addresses advocating for political causes in which I deeply believe. They are among the strongest voices advocating such causes. Were they to succeed in electing Ron Paul as president and dismantling the Federal reserve, no one will cheer more loudly than I.
What then, the reader might ask, is there not to like about Ludwig Von Mises. I have to answer sheepishly, a little embarrassed sounding like the Randroid that I am, I cannot stomach his epistemology. One’s epistemology is one’s theory of knowledge. Von Mises’ epistemology is called Praxeology
Can one do economics?
Von Mises’ view of Praexology is developed in a famous and much recommended book called Human Action. He also wrote a number of other works clarifying aspects of his philosophy. Von Mises felt that economics could never be a science. I will summarize his view with a quote from a collection of his writings published posthumously: [Money, Method and the Market Process [1990] ] “Economic history can neither prove nor disprove the teachings of Economic theory.” Anyone who thinks that I am taking this radically out of context should look here.
“Economic history can neither prove nor disprove the teachings of Economic theory.” A fair-minded application of this principle can be very painful. Maybe there are lots of competing economic theories: Austrian, Marxist, Keynesian, Neo-Classical. You name it. Let’s say you like the Austrian theory better than the others. How do you prove you’re right. [Hint: you aren’t allowed to use economic history.]
As an example, people frequently tell me that Austrian economics is in fashion because
Peter Schiff predicted the housing crash. But if I take Von Mises seriously then Schiff did a naughty thing. If Austrian economics, as a theory, could be systematically used to make predictions, then if these predictions always failed to come true, you could use economic history to disprove the theory. If they always did come true, economic history through the scientific method would be helping to prove the theory. The only way to really support Von Mises is by viewing Austrian economics as a theory that cannot make predictions. [Joke: Of course it can make predictions. You just shouldn’t expect them to be correct.] This is exactly what Woods is asserting in the article linked to above. You don’t use Austrian economics to predict history. You wait for history to happen. After it does, you use Austrian theory to know who to blame.
I think the reader can clearly see why this is unsatisfactory. I want to take Austrian Business Cycle theory seriously. I want to know how the size of a bubble might depend on the size of the central bank interventions that preceded it. No dice. That would be a quantitative not a qualitative conclusion. You would have to use math somehow and for Von Mises this is verboten. He says it can’t be done.
Now, of course it is true that there can be limits to human knowledge or nonsensical
questions but normally to prove this you have to do something formal (which might involve
math.) Von Mises just asserts it. Of course, he points out that all existing theories fail to make accurate quantitative predictions which is still true and a major problem for mainstream economics. And he also points out difficulties in interpreting history (it’s not a controlled laboratory experiment) which are frequently a problem for modern statistics. But it’s like the way in The Princess Bride, Buttercup expresses her fears to Wesley that they may not survive the fire swamp. Wesley correctly responds, “You’re only saying that
because nobody ever has.” That’s really the only reason Von Mises is saying it too.
I would like to point out one quantitative result in economics that predates Von Mises. It is
the law of supply and demand. You can hang out at ebay where the supply curve, demand curve, and price are all observable and the law holds. Indeed, this observation is almost the proof that the law holds. In the markets in which it holds, the price is set by some mechanism approximating an auction. In the limit, the approximation and hence the explicit choice of mechanism don’t matter. There are perhaps more aspects
of economics, aspects of macroeconomics, that can be made precise in the same way. Von Mises says no, but he’s just being unhelpful. Human knowledge can expand.
Getting to know you
The wikipedia article on praexology linked to above gives an objection Von Mises has to economic modeling. (I apologize I don’t have a more direct quote but the wikipedia article seems to be accurate and refering to ideas in “Human Action.”) It says that this kind of modeling does not work because the action of human beings in simple situations cannot be used to predict their actions in complex situations. (Honest, I’m not making this up. This is what Von Mises believed.) It would come as news to eharmony.
In fact, it violates day-to-day common sense. We try to get to know people. This means we
observe how they act in certain situations. Based on this, we decide if we want them to be
our friends. We do this by making predictions, not always accurate but still valuable, about
how they might behave in more complex situations.
Now, I don’t want to leave anyone with the impression that mainstream economists actually know how to predict the behavior of people in complex situations from their behavior in simple situations. They can make egregious errors which undermine their whole way of doing economics. I’d like to give an example.
The rational expectations flavor of modern macroeconomics is founded on the idea
that agents are maximizing their expected utility and that this utility function is concave
and even bounded. In laymen’s terms, they (rational expectations macroeconomists)
say that people get less happiness from their last dollar than their first dollar and that they can get only a finite amount of total happiness. This statement makes sense to them because they believe that happiness has well-defined units. (Otherwise, taking the expectation value wouldn’t make sense.) This is their model for how people act. I’d like to demonstrate that it is too restrictive as anyone finds out by asking the economists about it.
First you ask them, why should utility be bounded, and they tell you about the
St. Petersburg paradox. They posit a lottery with infinite expectation value and observe that no one will pay infinitely much for it. (It’s really true experimentally. I’ve yet to meet anyone who’ll sell me his soul for the right to play the St. Petersburg lottery against me.)
So then you say all right, but you start to pay attention to what is happening at your local
gas station. People are paying above expectation value for the right to play lotteries.
That isn’t actually consistent with the explanation about St. Petersburg. (They seem to
value their last dollars which they might win less than the few dollars they already have.)
You ask your favorite rational expectations macroeconomist about this and he explains,
“those people are just stupid.” This is no way to model people’s behavior. You have to
acknowledge that they’re surprise-loving agents and are not maximizing expected utility.
This alone probably says you have to overturn the whole of modern macroeconomics.
But the difficulty is that if you were to approach the followers of Von Mises with this and
tell them: “look mainstream economics is all messed up at the level of the choice of utility
functions. Please help me fix this.” In that case, they don’t help. They just tut tut and say,
“I told you so,” and then they insinuate how much smarter they are than you because they
aren’t trying to do the impossible. (Which they haven’t really proved is impossible. You might learn something just from analyzing data made public by the Hoosier lottery. )
Why should I like Ludwig Von Mises? Does he like me?
One thing that followers of the Austrian school of economics often complain about is that they aren’t in the economics establishment and the economics establishment doesn’t listen to them. It is hard not to sympathize with this complaint. After all, I’m not in the economics establishment, so nobody takes my views on economics too seriously. Maybe I’m in the math establishment, but even that’s debatable, stuck as I am at IU.
One of the virtues of math is that it really doesn’t matter too much if you’re in the establishment. Solving a math problem is basically an objective achievement. You can do it without anyone’s liking you. Success constitutes carrying out a number of formal steps and it really doesn’t matter so much who did so or what his lecture style is like.
In Austrian economics, math is not allowed. Formalism is not allowed. You write essays to get your points across and the ways of judging them are inherently subjective. If you really want to be taken seriously by anyone as an Austrian economist you should get made a fellow of the Ludwig Von Mises institute! The Austrians have
an establishment of their own!
It isn’t entirely clear to me what one has to do to join the Austrian establishment. I think it involves a lot of networking. You should make your way through the humanities departments of certain institutions and make friends with pre-existing Austrians. Perhaps it helps if you’ve bought helium balloons or sidewalk chalk for the right political causes. It probably doesn’t help to write a pub critical of Ludwig Von Mises and getting a Ph.D. in math might be a bad idea, too.
So here I am with a job that lets me think about anything I want and I find myself not infrequently thinking about economics. I even have some ideas. Now I can’t really talk about my ideas with Austrians because, first of all, ironically, I’m not in the establishment, and second of all, anyway, the ideas involve math. They’d be unlikely to want to talk to me, as what I’d really like to do are things that Von Mises thought impossible.
But there’s a funny thing about mainstream economists. It’s true I don’t share their politics. But not all of them are so political. Some may not even want to be running the Fed. (OK, maybe they’d all like to be running the Fed, just a little.) Still you find some of them
who are really truth seekers. They are puzzled at the poor record of their models in making predictions about reality and really want to make it work. They have a great respect for and interest in math, which may even seem to be a little excessive at times but is still endearing. They even have a touch of respect for me in an oddball mathematician kind of way.
But the most endearing thing about the mainstream, relatively apolitical, truth-seeking
macroeconomists is that being in the establishment isn’t so necessary for talking to them.
In fact, I could get their attention just by proving the right theorems. Much as there are things wrong with their methodology, there are still objective technical achievements that can get their attention. Appreciation of math is the greatest aphrodisiac. So that’s it, dear readers. I can’t help it. I like the mainstream, relatively apolitical, truth-seeking
macroeconomists. I like them quite a lot actually. I like Ludwig Von Mises, not so much.
Can you see, looking from my point of view, how this might come about? Flame away.
So do you really think that if you rigorously proved that the tenets of Austrian economics are true, the people at the Ludwig von Mises institute would not make you a fellow?
Yes. Because one of the tenets is that you can’t prove them rigorously.
But that’s just the epistemology, right? Isn’t the epistemology outside the actual economic tenets?
Let’s look at it that way. That means I could prove the economics rigorously.
Maybe I could make the Austrian theory of the Business Cycle part of mainstream economic thinking through a mathematical revolution.
The Ludwig Von Mises institute would probably not be impressed. They cling bitterly to their epistemology. They would say I’m just an evil mainstream economist who invented the new school of Neo-Austrian economics.
Well, don’t let that stop you from proving it! I’d actually like to see what the mainstream economists would do then! I’m not sure they are as pure of heart as you suppose. I suspect that their math adulation has its limits.
There are a lot of mainstream economists. Not all of them are the same.
This kind of dovetails with my belief that for most of those who subscribe to Austrian Economics it’s treated more like a religion than it is like an economic theory.
Dave
You know, I think this is true about most people who subscribe to anything.
F. A. Hayek, an Austrian economist, rejected the Misesian views on methodology in the last third of his career. He defended the methodological views of Karl Popper and his students, which I also defend. I’ve criticized the methodological views of Austrian economists for 30 years, while at the same time admiring much of the economic theory. The theory doesn’t depend on the methodology in the way that Austrians think it does, and as Hayek eventually discovered. You can completely eliminate the methodology without touching the economics at all.
That would be good. There are many Austrians who cling to their methodology. It doesn’t make them look good.
I hope you don’t mind the ramblings of an armchair Austrian with negligible mathematical skills (Among the serious Austrians, I think Bob Murphy’s the go to guy for math).
Keep in mind that Alfred Marshall would discard his mathematical economic solutions if he could not translate them into a systematic and logical verbal theory. David Friedman espouses the same thing, so Austrians aren’t completely alone in this way of thinking. Rothbard used limited equations to illustrate his theories, so it’s not as though math has no place in the Austrian school either. I suspect that the real point Mises was trying to make is that you cannot, as some seem to believe, invent a great calculating machine which can logically make all economic decisions for the market. The key elements of the market are determined by the subjective valuations of individuals. These valuations vary, not only from one individual to the next, but from one moment to the next.
You can extrapolate general tendencies based on an imaginary evenly rotating economy (ERE) or make a statement about a frozen moment in time, but I think Mises was just trying to point out that it’s a mistake to neglect the difference between reality and these mythical constructs (which are only tools for exploring and understanding the general direction of an economy under certain circumstances). You never actually reach the level of equilibrium in those constructs; you only see a tendency toward them. The uncertainty that exists outside of the ERE means that errors will be made and preferences will constantly change, keeping things from ever arriving at a final resting point.
Mises was not opposed to making predictions. He was certainly happy to speculate on the coming financial crisis in the years leading up to the Great Depression. He didn’t seem to have a problem with predicting likely trends, but he did seem to have a problem with the notion that this could be done in a quantitative versus a qualitative manner. You might be able to say that Joe prefers product A to product B, but you can’t put a numerical value on the degree of that preference and, even if you could, this preference would be meaningful only for a specific moment in time.
I’m not sure that you are so far outside of Austrian theory in your desire to explore a link between the size of central bank interventions and the size of a bubble. To say that more of A gives you more of B is a qualitative statement and the Austrians assert qualitative economic relations all the time. Rothbard frequently stressed that a demand curve ALWAYS falls as it progresses along the X (price) axis.
It was also not unusual for Rothbard and Mises to criticize and correct the theories of their predecessors, so I’m not so sure you’d be rejected by the Austrians if you managed to effectively do the same in proving a greater usefulness of mathematical economics. Continuing development is an important part of Austrian theory. If you’re looking for the truth, instead of trying to be a mouthpiece for statists and thieves, I suspect that your ideas will be met with respectful and thoughtful debate, not arbitrary dismissal. If your ideas stand up to the debate, I suspect that they will find acceptance.
To say more of A gives you more of B is qualitative but to try to say something about the derivative is quantitative.
My epistemology is that often qualitative statements must be derived from quantitative ones, especially when
there are competing forces pulling in different directions.
I’m still not certain that you are really in disagreement with Mises. I think you are just reading more into his statements than was intended. In the footnotes for chapter 9 of “Man, Economy, and State,” Rothbard discussed the Misesian approach to what are commonly termed “economic models”.
“It will be noted that we have avoided using the very fashionable term “model” to apply to the analyses in this book. The term “model” is an example of an unfortunate bias in favor of the methodology of physics and engineering, as applied to the sciences of human action. The constructs are imaginary because their various elements never coexist in reality; yet they are necessary in order to draw out, by deductive reasoning and ceteris paribus assumptions, the tendencies and causal relations of the real world. The “model” of engineering, on the other hand, is a mechanical construction in miniature, all parts of which can and must coexist in reality. The engineering model portrays in itself all the elements and the relations among them that will coexist in reality. For this distinction between an imaginary construct and a model, the writer is indebted to Professor Ludwig von Mises.”
In other words, play with all the math you want when you are applying it to an imaginary construct, such as the evenly rotating economy, but recognize the limitations of this math in predicting real-life outcomes (as human action becomes an element). In the real world, all of the assumptions can change rather quickly. New products may come on the market, lowering demand for an old one. Perhaps a public fervor for “green” products will drive capital to a less efficient fuel source. Maybe sudden public distaste for foreign products will shift capital to producers of more expensive domestic goods. Some swing in the political winds may instill a greater desire for thriftiness, making it more popular to save money. In the real world, there is no end to the possible changes that could occur outside the realm of any consistently quantifiable models.
One interesting thing about the line that economic history can neither prove nor disprove economic theory is that, as ironic as it is, it can’t possibly not be true.
Do you actually disagree with him on that?
Human history will never lend itself to the kind of mathematical precision that some of the hard sciences do. However, the laws of logic, applied to correct axioms about what human tendencies are, do lend themselves to that kind of precision.
I believe that economic history provides useful empirical evidence. I grant you that it is not
gathered experimentally and therefore its use is somewhat problematic.
Logic is always correct but is only as good as its axioms. When you try to treat a
quantitative world with only qualitative axioms, you have some problems.
Pingback: In Response to Nhkatz (and other simple arguments against Missesians) | Voluntary cooperation
Here’s a response to most of the arguments in this post.
http://www.pmabry.com/?p=135
Chaos Theory also states that deterministically chaotic systems like economies are not subject to precise mathematical treatments. You also can’t predict the weather more than two weeks out. It’s just not possible. Complex systems are initial condition dependent and nobody (save God) was around to see the initial conditions so we’re doomed to uncertainty because He’s not telling. Atheists are especially doomed because they don’t even have God.
Mises intuited this way before the mathematicians said the same thing much more rigorously and with robust general theories that apply way beyond economics. I don’t think that the Austrians have a problem with chaos math. Do you? Do you think that economies are not complex systems that are initial condition dependent and can only be predictable in very limited fashion?