1584 THE AMERICAN ECONOMIC REVIEW JULY 2016
envelope winners would be paid $2 but if the chip came from the Tails envelope they would win $3. Of course the optimal strategy in this game is to only circle numbers in the Tails row since those have a 50 percent higher expected payoff
(the benefit that you
get from doing something),
but this strategy was not obvious to everyone. About half the subjects (MBA
students at a top university) adopted the correct strategy of circling only
Tails, but the rest used what might called an “inept (someone who is inept does not have
much ability or skill)
mixed strategy,” dividing their choices between Heads and Tails, with the most
common allocation being three Tails and two Heads, matching the ratio of the
payoffs.[3] The
question that Kahneman and I were most interested in, however, was not these
initial choices. This was an experiment about learning. So we had the subjects
repeat the same task nine more times. Each time the subjects got feedback about
the outcome of the coin toss and the number drawn, and the winning guessers
were paid in cash immediately in plain view of the other subjects. Try to guess
the results as a thought experiment. Of the subjects that did not figure out (to be able to understand something) the “all Tails” strategy immediately, how many
learned to use that strategy over the course of the nine additional trials? The
answer is one. One subject switched at some point to an all Tails strategy, but
that subject was offset (to balance the effect of
something, with the result that
there is no advantage or disadvantage) by another subject who had
circled only Tails on the first trial, but then switched to the inept mixed
strategy at some point during the “learning” phase. It is instructive to
consider why there was essentially no learning in this experiment. We know from
psychology that learning takes place when
there is useful, immediate feedback. When learning to drive we quickly see
how much pressure to use on the accelerator and brake pedals in order to start
and stop smoothly. In the experiment, however, the subjects were first told the
outcome of the coin flip, then the number drawn. Obviously, about half the time
the coin came up Heads, and those who were including Heads in their portfolio
were pleased to be still in the game (if only for another few seconds). Furthermore,
every time that someone won some money from a Heads outcome, there was some
reinforcement for continuing to include some of that “strategy” in the
portfolio. The general point is that learning can be difficult even in a very
simple environment. Those who teach an introductory course in economics know
that many of the first principles that are basic to rational choice models
(such as the notion of opportunity costs
(Opportunity
cost is the profit lost when one alternative is selected over another. The
concept is useful simply as a reminder to examine all reasonable alternatives
before making a decision, https://www.youtube.com/watch?v=Ho6WuSAuECs )
are by no means intuitively obvious to the students. But our models assume they
can understand much more difficult concepts such as backward induction (backward induction is the process of reasoning
backwards in time, from the end of a problem or situation, to determine a
sequence of optimal actions, https://www.youtube.com/watch?v=pyLKkN5HpDY ).
As for the argument that people will do better in experimental tasks if the
stakes are raised, there is little or no evidence to support this hypothesis.
The first empirical test (empirical
testing is a research method that employs direct and indirect observation
and experience)
of this idea was conducted by
David Grether and Charles Plott (1979) in the context of an investigation of
the “preference reversal phenomenon,” discovered by psychologists Sarah
Lichtenstein and Paul Slovic (1971).
Lichtenstein and Slovic
presented subjects with two gambles, one a near sure thing they called the
p-bet (for high probability) such as a 35/36 chance to win $10, the other more
risky called the
$-bet, such as an 11/36
chance to win $30, a higher potential payoff. Subjects were
VOL. 106 NO. 7 THALER: BEHAVIORAL ECONOMICS: PAST,
PRESENT AND FUTURE 1585
asked to value each bet by
naming the lowest price at which they would sell it if they owned it, and also
to choose which of the bets they would rather have. The term “preference reversal (preference reversals
refer to the observation that there are systematic changes in people’s
preference order between options. Preference order refers to an abstract
relation between two options)”
emerged from the fact that of
those who preferred the p-bet, a majority reported a higher selling price for
the $-bet, implying that they valued it more than the p-bet.
Grether and Plott (1979) were
perplexed (confused because
you cannot understand something,
https://www.youtube.com/watch?v=ZM1TpzQUsIs )
by this finding and set out to determine which mistake the psychologists must
have made to obtain such an obviously wrong result. Since the original study
was based on hypothetical questions, one of the hypotheses Grether and Plott
investigated was whether the preference reversals would disappear if the bets
were played for real money. (They favored this hypothesis in spite of the fact
that Lichtenstein and Slovic (1973) had already replicated their findings for
real money on the floor of a Las Vegas casino)
What Grether and Plott found
surprised them. Raising the stakes did have the intended effect of inducing (to cause something, especially a mental or physical change, https://www.youtube.com/watch?v=bcRCYD1DZ7Q )
the subjects to pay more attention to their choices (so noise was reduced) but
preference reversals did not thereby vanish; rather, their frequency went up!
In the nearly 40 years since Grether and Plott’s seminal paper (seminal works, sometimes called pivotal or landmark
studies, are articles that initially presented an idea of great importance or
influence within a particular discipline), I do not know of any
findings of “cognitive errors” that were discovered and replicated with
hypothetical questions but then vanished as soon as significant stakes were
introduced.
C. The Invisible Handwave
There is a variation on the
“if there is enough money at stake people will behave like Econs” story that is
a bit more complicated. In this version markets replace the enlightening role
of money. The idea is that when agents interact in a market environment, any
tendencies to misbehave will be vanquished
(defeat
thoroughly, https://www.youtube.com/watch?v=5akHhbqv_o4 ).
I call this argument the “invisible handwave” because there is a vague allusion (an expression designed to call something to
mind without mentioning it explicitly; an indirect or passing reference, https://www.youtube.com/watch?v=si8dQKrDIBg )
to Adam Smith embedded in there somewhere, and I claim that it is impossible to
complete the argument with both hands remaining still. Suppose, for example,
that Homer falls prey (to be influenced by
someone or something) to the “sunk
cost fallacy (the idea that
a company or organization is more likely to continue with
a project if they
have already invested a lot
of money, time, or effort in it
even when continuing not the best thing to
do, https://www.youtube.com/watch?v=vpnxd31y0Fo )”
and always finishes whatever is put on his plate for dinner, since he doesn’t
like to waste money. An invisible handwaver might say, fine, he can do that at
home, but when Homer engages in markets, such misbehaving will be eliminated.
Which raises the question: how exactly does this occur? If Homer goes to a
restaurant and finishes a rich dessert “because he paid for it” all that
happens to him is that he gets a bit chubbier
(slightly fat, in the way
a healthy baby or young child is).
Competition does not solve the problem because there is no market for
restaurants that whisk (take or move
(someone or something) somewhere suddenly and quickly, a kitchen tool )
the food away from customers
as soon as they have eaten more than X calories. Indeed, thinking that markets
will eradicate aberrant behavior (is something
that does not follow the correct or expected course or something that is not
typical or normal)
shows a failure to understand
how markets work. Let’s consider two possible strategies firms might adopt in
the face of consumers making errors. Firms could try to teach them about the
costs of their errors or could devise a strategy to exploit the error to make higher profits. The latter strategy will
almost always be more profitable. As a
rule it is easier to cater to (to provide
people with something they want or need, especially something unusual or special)
biases (to influence someone’s opinions, decisions etc. so
that they behave or think in
an unfair way)
than to eradicate them. DellaVigna and
Malmendier (2006)
provide an instructive
example in their article “Paying Not to Go to the Gym.” The authors study the
usage of customers of three gyms that offer members the choice of paying $70 a
month for unlimited usage, or a package of 10 entry tickets for $100. They find
that the members paying the monthly fee go to the gym an average of 4.3 times
per month, implying an average cost of over $17 per visit.
[3] Likewise, when Heads and Tails aren’t equally likely, people tend to engage in
“probability matching” behavior instead of just picking the more likely outcome
every time. See Vulkan (2000) for a survey aimed at economists
Комментариев нет:
Отправить комментарий