"behavioral economics"

[Jim Devine <jdevine@xxxxxxxxxxxxxxx>]

Musings du jour:

In BUSINESS WEEK ("Putting a Human Face on Economics," July 31, 2000, pp.
76-7), there's an article about Matthew Rabin and other "behavioral
economists." (Though they don't mention him, the guy in the office next to
mine (James Konow) also does behavecon.) This school points "out all the
ways in which people diverge from the hyperrational behavior that is
assumed by conventional economics. They [people, not behaveconomists]
procrastinate on saving for retirement. They shop for hours to save
pennies, then make snap decisions on big-ticket items. They run up huge
credit-card debts even when they have ample savings to pay them off."

Besides sounding like they're describing my own behavior, I think that this
is a worthwhile break from standard NC economics. Imagine, actually
_studying_ how people behave rather than simply assuming that they are
"rational"!  However, it should be noted that this school seems not only
behavioral but behaviorist (though not of the B.F. Skinner sort). There is
little concern with consciousness or cognition or neurology or the actual
psyche, as far as I can tell. There also seems to be little effort to face
the basic sociological insight that our "tastes" or "preferences" are to a
large extent endogenous to the economy and society. They do note that
"people match their behavior to that of those around them," but I say we
could go much further.

It's important also to note that this school is liberal (in the sense that
both neoliberals of the laissez-faire school or the New Deal/modern liberal
school are). The focus is thus on individual behavior and how it affects
market behavior or (I would guess) the behavior of other human-made
institutions; in sum, if you change the assumption of how individuals
behave, it changes the results of the model. (I doubt, by the way, that the
behaveconomists see markets as  institutions, but that insight might be
joined with theirs.)

For example, Richard Thaler "uses behavioral theories to explain what
economists call the 'equity premium puzzle': the odd fact that the
long-term returns on equities are much higher than those on bonds -- even
more than their higher volatility would seemingly call for. Thaler says the
answer to that puzzle is that people hate losses much more than they enjoy
gains. So investors demand higher returns from stocks to compensate them
for their dread of losses."

This kind of behavior fits my experience. However, I feel dissatisfied with
the theory; my visceral feeling is that it is _ad hoc_. I hate to say that
it makes me feel sympathy for the more orthodox NC viewpoint, which would
try to explain such behavior in terms of the objective conditions faced by
individuals who "play the market" -- though I employ a broader definition
of what these objective conditions are. (Hard-core NC economists see only
prices and incomes as relevant, with the hardest-core leaving out incomes.)

First, I'd bring in the post-Keynesian view that there's fundamental
uncertainty involved in the "market" (as elsewhere). Among other things,
this implies that the "perfect futures markets" that NC economists pine for
(and through their friends at the IMF try to impose on the world) can never
exist. All we _really_ know about the stock market's prices is that they
fluctuate. This also brings in the problems arising from illiquidity. It's
always possible that some emergency will happen and you need to sell your
stocks to get cash. If the stock market is "up" at this point, you can sell
your stocks or use them as collateral to get a loan. On the other hand, if
the market is "down" and you have to sell, it's at a low price, so you're
forgoing possible future gains -- and your stock isn't worth as much as
collateral. There's also an element of irreversibility, since if you sell
when the market is down, when you go back into the market later, it's
likely that prices will be high again, but this time you'll be a buyer.
This irreversibility encourages asymmetric behavior: if the market is "up,"
you try to lock in your gains (if you're risk averse) while if it's "down,"
you struggle to end your bad fortune as quickly as you can. In sum, we
might see the "irrational behavior" that Thaler points to as _rational_,
given the conditions stock-market players face. This in turn encourages the
existence of an equity premium.

In general, I'd guess that the fact that for most people, the dread of
losses exceeds the happiness deriving from an equal level of gains reflects
the fact that the conditions I described for the stock market has also fit
life in general since humanity arose from evolution. If one's
hunter-gatherer band gains a new member, that makes hunting and gathering
easier, but it's relatively easy to reverse that change. On the other hand,
if the band loses a new member, that's bad -- and worse, is very difficult
to reverse. If you have a car accident which totals your wheels (and you
have no car insurance), the value of this fall to you is greater in
magnitude than the benefits of getting a free car of equal price. Even if
you have car insurance, uncertainty can't be solved via insurance and the
company always insists on a deductible (to avoid moral hazard).  The net
monetary loss would usually be more valuable to you than an equal
accidental monetary gain -- because car accidents are scary and might
involve physical harm, because you're used to your own car, there are
bureaucratic hassles dealing with the insurance company, etc., etc. Under
these ubiquitous conditions, the generalization that the behaveconomists
point to makes total sense in terms of traditional economic rationality.

It reminds me of the time I walked in on a conversation among two
economists at my previous job. They were wondering about that
quintessentially Southern California issue: why is it that old folks drive
so slowly?  One was saying that they drive slowly because the opportunity
cost of their time is low (since they don't have waged jobs). The other
said that since they didn't expect to live much longer, they should value
their time _more_ (an application of the life-cycle theory) and thus should
drive more quickly than they do. Said I: "they drive more slowly because
they have slow reaction-time and eyesight." Said economist #2: "oh, these
radicals!" (To his credit, it was partly in a self-mocking tone.) To
physical infirmities, I'd add that old folks also are more likely to
respect the law (including the speed limit) more than younger ones, while
they often drive large heavy vehicles that make them feel safer but also
force them to peer over the dashboard to see; further, they developed
driving habits back when people obeyed the law more. Many or most orthodox
economists seem unconscious of the complexities of the human condition,
instead seeing the world as One Big Market (with a bunch of "imperfections"
thrown in that never seem to interact with, and reinforce, each other).

Since I'm far from being an expert on behavecon, I can't really be
conclusive in my criticism. (I really have to read Thaler's book.) Some of
this work seems worthwhile. For example, James Konow writes about
individuals' sense of justice. He's done experiments that show that people
violate standard NC rationality to live up to these standards of justice.
These standards seem a reflection of the fact that people are basically
social animals rather than the unreal individuals of NC theory. Maybe I can
get him to give me a quick summary of the theory and I'll forward it to you
folks.

Jim Devine jdevine@xxxxxxx &  http://bellarmine.lmu.edu/~jdevine