Why I am not an Austrian Economist

Murray Rothbard

A friend whose opinions I respect greatly, and who I hope will one day write some of them down as posts here or elsewhere, forwarded me a great critique of Austrian Economics. The article is written by Bryan Caplan at George Mason University, which university is just outside Washington DC and is known for freer market thinkers. I present a brief summary of the article, which is really more of a critique of Mises/Rothbard Economics than of Austrian Economics, below. Unfortunately I must leave a lot of discussion out to save space - the reader is encouraged to consider the whole article for a fuller picture.

Ludwig von Mises

Foundations of Microeconomics
Mises/Rothbard dismiss the use of utility functions because they reject the idea of cardinal utility. However, this rejection is spurious because utility functions are merely a useful tool to summarise ordinal preferences and they do not constitute a belief in cardinal utility.

Mises/Rothbard reject the concept of indifference curves because they believe that human preferences can only be revealed by human action and such preferences cannot be known if they are not revealed. In other words, if a person is truly indifferent between two options, that person can make no decision between the options; no action is taken and the indifference is unobservable. Caplan argues that preferences can and do exist even if they are not revealed by action and therefore that indifference between two options is knowable even if it is not revealed, thus the use of indifference curves is justified.

Mises/Rothbard reject the assumption that human preferences, as described through supply and demand curves, are continuous. Hence Mises/Rothbard reject the use of calculus in economics. Logically then, Mises/Rothbard's supply and demand functions should be discontinuous - unjoined points on an XY plane - which would almost never intersect. A market price could almost never be established where quantity supplied and quantity demanded are equal.

Results and applications of Austrian Economics
Mises/Rothbard consider the economic calculation problem to be the decisive argument against the economic feasibility of socialism, even though empirical evidence does not exist to support such a claim in the face of other cogent arguments against the economic feasibility of socialism:

1. Work effort
2. Innovation
3. Underground economy

Rothbard does not accept the existence of positive externalities, but he does accept the existence of negative externalities as an odd consequence of his above discredited utility theory. Caplan argues that this distinction is nonsensical because "negative externalities often don't violate property rights, and positive externalities can".

Mises/Rothbard correctly argue that unemployment is caused by excessive real wages and that inflation is not a long term solution to the unemployment problem. However, Rothbard says that "wage rates can only be kept above full-employment rates through coercion by governments, unions, or both." Caplan argues that this is a typical generalisation and ignores the "market-based impediments" to downward wage adjustment:

1. Damage to employee morale,
2. Formal contracts specify wages, and
3. Wage renegotiation can be expensive - it takes time to bargain and renegotiation risks the loss of mutual goodwill between employer and employees.

Caplan also takes issue with the following conclusions of Austrian Economics:

Monetary expansion distorts the structure of production unsustainably
The Austrian Business Cycle Thoery explains the "sudden cluster of business errors"
The Austrian Business Cycle Theory provides the best explanation for why downturns hit the capital goods sectors especially hard
Only the Austrian Theory can explain the existence of inflationary depressions (stagflation)

Quantitative methods in economics
Mises/Rothbard say that economic history can only illustrate economic theory, that economic theory is the only 'true' form of knowledge and it is impossible to use economic history to 'test' properly economic theory. Caplan argues that economic history can be used to test economic theory, but is separately important for the measurement of effect sizes.

Mathematics and statistics have emerged over the last 50 years as the main drivers of economic research, even though their track record in delivering results has been poor relative to qualitative economic theory. Caplan argues that Austrians beleive that the failure of mathematics in economics should lead to the removal of mathematics from the discipline altogether.

Conclusion
In sum, Milton Friedman spoke wisely when he declared that "there is no Austrian economics - only good economics and bad economics," to which I would append: "Austrians do some good economics, but most good economics is not Austrian. Bryan Caplan