Stiglitz on Homoeconomicus

|Peter Boettke|

Joe Stiglitz used the opportunity he was given as a keynote speaker at the AEA meetings to argue that the financial crisis provides such overwhelming evidence of irrationality that it is scientifically irresponsible for economists to continue to build models based on rational behavior.

What do you say?

Comments

[kd0imh]

I think it's similar to how he and others argue that you cannot use the Efficient Markets Hypothesis for anything since it's been "proven" not to be valid for macroeconomic analysis. I believe it's called throwing the baby out with the bathwater.

[Charlie Martin]

I'd say there's a difference between what someone might rationally choose to do, and what Joe Stiglitz thinks they ought to do if they were only as rational as he is, and knew then what he knows now.

Consider, eg, the recent financial crisis. It appears that it came about primarily by people rationally maximizing return in the presence of non-market drivers like regulation. This led to "regulatory arbitrage", and so led to situations like AIG's, where one division behind a firewall in Europe was able to expose the entire firm to a prodigious amount of risk that didn't show up on the company's books in the US. But the people in Europe were rationally maximizing their returns in compensation within their rules; the people in the US were rationally obeying the accounting and financial rules as they stood; and AIG was rationally taking advantage of the differences between US and European regulations to generate a return, without grasping the risk to which they were actually exposed.

It would appear that the presence of regulations that made the lack of information transmitted from one corporate entity to another a mechanism for generating real cash returns lef a bunch of smaller actors, each acting rationally, into a situation that appears irrational later.

[Mark Bonica]

Based only on having read the linked blog post, it strikes me that Stiglitz is criticizing the rationality of the financial "system" (which can't be rational because it doesn't have a brain) rather than the rationality of the actors in the financial system. The financial system may indeed fail to "manage risk and allocate capital at low transaction costs" if the incentives for the actors assigned to perform those tasks don't align properly due to moral hazard. Didn't he win a prize for writing about this sort of thing?

[Mark Cancellieri]

I think that people are pretty rational, but obviously not perfectly rational.

The financial crisis certainly does not prove any significant degree of irrationality. Market participants responded in fairly predictable ways. For example, they reacted to artificially low interest rates by borrowing heavily. They responded to regulatory requirements by trying to get around them through regulatory arbitrage. Etc.

I would say that our Keynesian friends are the biggest argument in favor of significant human irrationality.

[Lucas M. Engelhardt]

I think one of the difficulties is that rationality is often vaguely defined - which is very bad, since it usually implicitly assumes a lot. Some of these assumptions are fine (things like ex ante utility maximization - which Mises, for example, suggests is actually necessarily true given the Misesian view of preferences) while others (rational expectations, say) seem quite false. (For example, it seems clear that many people talking out ARMs were probably not functionally aware of the distribution of interest rates across time.)

Personally, I think that the word "rational" or "irrational" is mostly used as an intellectual bludgeon largely for political purposes. Those who want some sort of intervention simply declare the behavior of the market "irrational" (perhaps by benchmarking against an omniscient benevolent social planner), at which point it is declared that the government should act as the benevolent social planner for the market's own good.

[Troy Camplin]

I would say that people rationally interpreted the (false) signals they were given -- the primary false signal being the artificially low interest rates created by the Federal Reserve that told people that there should be more risk taken in the mortgage (and other lending) markets than was actually safe in the market at the time.

And the Fed Chairmen in question were acting rationally based on a bad theory.

So the problem isn't rationality, but bad information and bad theory.

[Greg Ransom]

How many tenure track positions does Stiglitz control?

The weight of his pronouncement rests on that, I'd suggest.

[EF]

Irrational people make rational models popular. Irrational models might become popular when people become more rational -- but this would make the models wrong again, wouldn't it?

[Pietro M.]

With some moral hazard, the apparently irrational behavior of markets in recent times becomes fully rational for every single individual agent. So, at least in principle a DSGE with credit booms engendered by insurance-like countercyclical monetary policies and their sidekicks (regulatory forbearance, bailouts, deposit insurance) could be obtained.

I wouldn't trust a model which assumes such a high level of coordination and hyperrationality, but as a methodological fiction* it would show that agents can rationally behave in a apparently crazy way if they take into account the safety net.

De Grauwe's attempt to revive animal spirits in his recent paper linked on this blog reveals a pitfall in these developments: to kill the walrasian auctioneer to revive animal spirits would be the ultimate fail for free-market theories, surely not a theoretical improvement: BRICE will turn into BRACE.

* This strategy would be similar to the defense of full resource employment in O'Driscoll's "Economics as a coordination problem" and of game-theoretical analyses in Carilli and Dempster's "Expectations in ABCT".

[j3sus]

I agree... but thats the proof of nonsense of the stiglitz-keynesian economics and something, which every austrian knows..

I read "rational" in a human-action meaning, but guess if stiglitz does so.

[www.facebook.com/profile.php?id=36918063]

a) Fellow economists and economically-trained practitioners who relied upon the Austrian concept of rationality seemed to do just fine in anticipating the crisis from a pattern prediction perspective. Some of them (notably the practitioners such as Marc Faber) went on to nail the timing and magnitude of the denouement also.

The last time round we had a Great bust in the West it was the same story. Poor old Irving Fisher was forced to develop his theory of debt deflation after events, in part to explain to his mother-in-law why he had lost a great deal of her money in the market. Whereas Mises and Felix Somary, the great private banker, were warning about the great smash to come well before it happened - they turned out to be right for the right reasons.

On the other hand it is not clear to me that those trained in Stiglitzian information economics or modern behavioural approaches had any idea that a bubble was forming or that its bursting would be so painful.

b) It is hardly surprising that academics feel like their talents should be better recognized (didn't Mises and Hayek write about this phenomenon) and rewarded with prestige, money and power. But when has academic ability (beyond a basic threshold) ever had anything to do with success in business or finance? Viz LTCM and successor organizations.

Stiglitz presumably considers Ben Bernanke and Larry Summers to be people of high innate ability in a certain sense (as would I). Yet being strong in that particular dimension didn't prevent both getting every important aspect of the evolution of markets and the economy in recent years wrong.

As Hayek said, it is a grave mistake to believe that the purpose of compensation in a market economy is to reflect transcendental standards of moral value. If anything, other things being equal, one would expect jobs that allow people to feel good about themselves to be worse compensated than those that have the reverse aspect to them.

[cushion]

very interesting blog. i'm an economics consultant (graduate in economics) and also love to blog about economics. please check out my blog and follow me :)

you can see my articles on economics on:

http://windowsillseat.blogspot.com/search/label/Economics

[Jonatan Krovitsky]

Given
A) Scientists are rational
and
B) Human are irrational
leads to:
1)Scientists are not human
2)Human are not scientists
3)Logic is never an obstacle for a true economist
4)None of the above

[Josh]

Yawn

[TGGP]

Herb Gintis also argues that the economic system as a whole is "irrational" even if all the individual agents are rational:
http://www.amazon.com/review/R225L5O06GDP79

[Stewart Dompe]

Professor Rizzo has an excellent response in this blogpost: http://thinkmarkets.wordpress.com/2009/12/10/a-tiger-of-a-mistake

Specifically, we are making predictions about markets and not individual behavior. The rationality assumption is very similar the the transitivity of preferences assumption. If individuals have intransitive preferences, they are likely to loose all their money and be removed from the market. The emphasis should be on the institutions that create rational outcomes, not on the quality of the individual agents.

[Mario Rizzo]

Thank you, Stewart. After I posted this I corresponded with a couple of mainstream economists. I was very surprised at their disagreement. They each said (essentially) that they are concerned about *real* individuals. One who is very market oriented said that when he did experiments, he experimented on real people, not puppets. (This was not Vernon Smith, btw.)

If this is the widespread view, then one can readily see the opening for behavioral economics and the charge of "irrationality."

Another aspect of this is the argument for system or "ecological rationality" made by Vernon Smith.

[Roger Koppl]

Mario's testimony in this regard is important IMHO. The "puppets" point is basic stuff and it is surprising that it has been, seemingly, completely forgotten by all but a few economists. The old theory was pretty strong when properly interpreted as per Machlup. It is rock stupid if applied as literal truth to named individuals, and yet that's how it is understood. Almost any mediocre theory is an easy match for such a bastardized neoclassicism.

[Jim]

The largest irrationalities I see in the financial crisis amidst a global complex system that is evolving at ever greater speeds are two; first, that a central bank can be smart enough to set interest rates and 'manage' the economy effectively; and second, that a Congress of elected generalists together with quasi-private mortgage firms can legislate full home ownership without huge negative externalities.

The mind boggles and the heart grows weak. And any 'macroeconomic' theory that supports those constructs should now be trashed forever.

[Ryan Vann]

The word rationality seems to be tossed about carelessly, and without ever defining the term. If you were to say markets run based on asymmetric information, and are thus not always in perfect equilibrium, I wouldn't disagree, but I wouldn't agree that means a rationality assumption is invalid.