Tuesday, August 25, 2009

Serious Person Syndrome

Paul Krugman:
I do have one qualm, though, which isn’t really about Bernanke, but rather about the broader symbolism of the reappointment — namely, it unfortunately seems to be a reaffirmation of Serious Person Syndrome, aka it’s better to have been conventionally wrong than unconventionally right.

This sounds right and important. And it need not only apply to reappointments, but to theories and whether they are accepted or not. Absurd statements in conventional packings - accept. True statements in unconventional packings - reject. But this is not just a complaint, more a call to investigate and change. Investigate whether there really is such a bias and if so, try to change the conventions so that they function better as a filter to distinguish between good and bad theories.

Is it possible to do this? And how? First, to believe the Serious Person Syndrome I would like to have something more than anecdotes to base it on. For instance an experiment in which individuals played a game in which every round some people win or lose based partly on ability and partly on luck. Moreover, there would be some "conventional" strategy and unconventional strategies. Maybe based on focal points or just norms or conventions established before the game by the instructors. One could then examine whether the players assigned less blame to a player who was playing an unsuccesful but conventional strategy, than an unsuccesful unconventional strategy.

Vague words you say? What is the empirical proof? Well, I recall a paper where the author believed to have found a sub-optimal conventional strategy in American football. The existence of this was explained by coaches trying to avoid blame for losing a game and there was less blame involved when chosing the sub-optimal, but conventional, strategy.

Maybe the reader can give more examples and create better experiments?

Thursday, August 20, 2009

Macroeconomics as convoluted fiction

One successful strategy in microeconomics is to search for something that seems stupid, ignorant, misguided, etc. and dream up some implausible, ridiculous story that explains it as actually being sophisticated, subtle optimization. The explanation is linked very loosely to the real world by linking selected assumptions and effects to anecdotes or stylized facts, and if someone says it is all nonsense you can retreat to the “as-if” defence. Since you can “explain” it through your convoluted Alice-in-Wonderland model, you can then claim that the phenomena in question is “no longer a challenge to standard rational choice theory.”

Personally, I became aware of this through my PhD work on rational addiction theory. A 1988 note from University of California, Berkeley Professor Jeffrey A. Frankel that was recently linked to in Krugman’s blog points to the same thing in macroeconomics.

Stockman comments on how everything can be made consistent with dynamic stochastic general equilibrium models based on indiviudal optimization (the “equilibrium view”). When economists thought purchasing power parity held, this was interpreted as evidence in favor of the equilibrium view. When they thought it didn’t… well, this was evidence in favor of the equilibrium view. So when a dataset doesn’t allow you to reject the null hypothesis that real exchange rate follows a random walk, this:

[…] is […] interpreted as evidence in favor of the equilibrium theory, even though the latter has no more testable implications for the real exchange rate than does the proposition that 9 is a prime number.

He notes evidence that would seem to favor sticky-price theory, and shows how “equilibrium” theorists work out convoluted, bizarre models to “explain” this within their framework.

Such explanations are clever, and make for good
journal articles that are popular among academic economists. But that doesn't make them true.

Speaking of "agents," spy novels are a good analogy for stories that are clever and make entertaining reading, but have little to do with the truth. Datum: a few minutes ago, I got up from my chair next to Alan Stockman on the stage, and walked over to take my place here at the podium. Hypothesis 1: I am a spy for a foreign power, Alan is a CIA counterspy who was about to assassinate me, and so I got up to move out of range. This hypothesis is "consistent with the facts" in the sense that, if true, it would explain them; but it is convoluted and not very plausible. Hypothesis 2: John Le Carre was in British intelligence before he
began his second career as a novelist. This hypothesis is  interesting to speculate about. I have no idea whether it is true or not. It is also "consistent with the facts" in the weak sense that it does not contradict the datum. But it seems no more relevant than the statement that 9 is a prime number, the proposition that agents dynamically optimize, or the hundreds of other hypotheses that I "fail to reject" every morning in the shower. Hypothesis 3: I came up to the podium for the simple reason that AEI invited me here to comment on Alan's paper. While not as clever as the other propositions, this hypothesis is simple, plausible, and
consistent with the facts in the strong sense that it would explain them while most other hypotheses would
not. (I will leave it to you to decide which hypothesis is the correct one.)

He also has some fun comments on the flip to a state where the goal of macroeconomics became to find nothing (no explanation, no relevant causal variables etc.):

The word "nothing" will play a key role in my comments.
["We know nothing, therefore we should do nothing."] The word does not often appear explicitly in the writings of equilibrium theorists. The popular phrase in the econometric writings is "random walk." [The usual conclusion is stated as "I have found that such-and-such a variable follows a random walk." Or, at
best, "I cannot reject the hypothesis that this variable follows a random walk." You seldom hear someone say, "After studying this variable for 6 months, I have absolutely nothing to say that would help to predict its movements." But the statements mean the same thing.] In Stockman's paper, the phrase is "in the current
state of knowledge:" "In the current state of knowledge...exchange rates and the current account should play little role...[in the conduct of monetary policy]"

Tuesday, August 11, 2009

It’s OK to laugh at Robert Lucas, everyone :-)

One cannot find good, under-forty economists who identify themselves or their work as ‘Keynesian’. Indeed, people even take offense if referred to as ‘Keynesians’. At research seminars, people don’t take Keynesian theorizing seriously anymore; the audience starts to whisper and giggle to one another.

(Robert Lucas)

I loved finding this quote by Lucas. It means I can finally enjoy the Solow quote below without any trace of bad conscience:

Suppose someone […] announces to me that he is Napoleon Bonaparte. The last thing I want to do with him is to get involved in a technical discussion of cavalry tactics at the Battle of Austerlitz. If I do that, I’m getting tacitly drawn into the game that he is Napoleon Bonaparte.

(Bob Solow, talking about Robert Lucas’s rational expectations work in “Conversations with Economists” by Klamer and Colander)

Saturday, August 1, 2009

Economists understand everything – after the fact

The standard defence of absurd theories is usually that they predict. Predictive success is  the only relevant criterion, assumptions be damned. But what should we say when we fail to predict – and fail spectacularly at that, as in the recent financial crisis. The answer: Flip our defence around. We may not predict, but at least we have the understanding and insight to explain what happened after the fact. As Chris Dillows discusses here.

He starts by pointing out that it is possible to find theoretical concepts, small groups or individual researchers that have worked on things in the past that now seem relevant to understanding the financial crisis. He then relates this to Jon Elster’s troubling view of mechanisms as  a form of completely non-predictive understanding.

To put it more crudely than this guy himself does: We have loads and loads of story elements in economic theory that can be combined to join any set of assumptions to any outcome. All we need is to know what outcome actually occured, and we’ll be able to serve up a convincing and sophisticated “explanation” in no time.

[…]we have […] lots of mechanisms, capable of explaining why things happen and the links between them. What we don’t have are laws which generate predictions. In his book, Nuts and Bolts for the Social Sciences, Jon Elster stressed this distinction. The social sciences, he said:

“Can isolate tendencies, propensities and mechanisms and show that they have implications for behaviour that are often surprising and counter-intuitive. What they are more able to do is to state necessary and sufficient conditions under which the various mechanisms are switched on.”

This is precisely the problem economists had in 2007. We knew that there were mechanisms capable of generating disaster. What we didn’t know is whether these were switched on. The upshot is that, although we didn’t predict the crisis, we can more or less explain it after the fact. As Elster wrote:

“Sometimes we can explain without being able to predict, and sometimes predict without being able to explain. True, in many cases one and the same theory will enable us to do both, but I believe that in the social sciences this is the exception rather than rule.”

The interesting question is: will it remain the exception? My hunch is that it will; economists will never be able to produce laws which yield systemically successful forecasts.

What’s more, I am utterly untroubled by this. The desire for such laws is as barmy as the medieval search for the philosopher’s stone. If you need to foresee the future, you are doing something badly wrong.

I especially dig the ending. Friedman in the more extreme statements of his “as-if” article on positive economics made absurd assumptions a virtue, and now this guy is making a lack of predictive success a virtue.

Turning our assumptions on ourselves as economists, we are (of course) always doing the optimal thing. A fact that gives me immense peace of mind: Whatever I’m doing, I can rest assured that I could have done no better.