Greg Ransom has posted selected quotes from Hayek that deal with the issue of whether macroeconomics is science or pseudo-science. His answer is that it is the latter. Why is this?
As I understand it, the idea of measuring only aggregates of individual human acts (that of earning an income, for instance, translates over a country into GDP) can help us understand history, but Hayek believes that no amount of “macro modeling” can allow us to predict, with any sufficient precision, future GDP or unemployment.
Hayek considered it wrong-headed to recycle a method borrowed from physics, where the Law of Large Numbers allows for reliable prediction. That Law simply states that if individual units behave in a certain random but “tractable” way, aggregate statistics or measurements of what are called “mass phenomena” can be predicted by a theory. Such predictions can then allow the scientist to reject wrong theories, without necessarily accepting a “right” theory. Such predictions can also be the basis for “control,” in the sense that we can control a rocket to land on the moon.
But Hayek believes that such prediction or control is impossible in economics where we “endeavor to explain the structure of society, [because] we do not have to deal with true mass phenomena. The events which we must take into account in any attempt to predict the outcome of particular social processes are never so numerous as to enable us to substitute ascertained probabilities for information about the individual events.”
What Hayek seems to be saying is this: Economic aggregates cannot explain or predict aggregates. So that when a macroeconomist builds a model, and all he has are aggregate or average measurements on both sides of his equations, all he has is spurious correlation, and when he uses that model to espouse “policy,” he commits the fallacy of post hoc ergo propter hoc. The retort of modern macroeconomists is that once they have given macro its proper microeconomic foundations, including dealing with such phenomena as expectations (even allowing for irrational ones), it is a “robust” theory, and if the statistical testing proves favorable, then that’s it (Woodford gives a survey based on this view). Who is right is perhaps a still-unfolding (and never-ending) story within the economics profession.
But what is the alternative to modern macroeconomics? One is to confine economics to a sub-part of history, as suggested by McCloskey, and give up, with all humility, on prediction and control of society. The other, proposed by Hayek, is to rely on what we know from microeconomics and institutional economics (which combines history with theoretical developments from microeconomics, such as game theory), to come up with Hayek’s “events which we must take into account.” Hayek seems to agree with McCloskey, because such “taking into account,” is (only) for the purpose of “explaining the structure of society.” (I summarize McCloskey’s 1984 paper here.)
There is one feature shared by McCloskey’s reasoning and Hayek’s. Hayek would claim that his explanation, based on microeconomics, is enough to warn ourselves of what things are futile, such as economic “fine tuning,” which is the playground domain of Keynesian economics. McCloskey adds that economics can still predict many things, but not those that violate the Law of Zero Profit in the Long Run. This Law is of course a fundamental one in microeconomics.
To sum up, the Hayekian view is that we cannot predict and control the individual and society. But can we perhaps make some tentative conclusions on how human beings might respond to certain changes in institutions or economic policies?
My guess is that there are such conclusions, but they are not arrived at through theories alone. Rather, they may be found by a careful check of whether the economic stories we tell “square with historical facts.” I would also hazard an opinion that so far, the stories of institutional economics have fared better than those of modern macroeconomics.