Jus ved NTNU? (var: Koking etc.)

[Trond Andresen]

Unders=F8kelser av spesielle kategorier studenter for om mulig =E5 kartle=
gge trekk som
er framherskende i den aktuelle gruppe, er interessante greier. Jeg har a=
llerede
henvist til en unders=F8klese om jus-studenter. Her f=F8lger en sak om =F8=
konomistudenter
i USA.

(Teknisk merknad: Artikkelen er optisk innlest fra en papirkopi. Derfor k=
an den
inneholde sm=E5 feil p.g.a. det.)

Trond Andresen

________________________

(the following is from THE ECONOMIST, May 29, 1993)

HOW DO YOU MEAN, "FAIR"?

SOMEBODY, presumably Groucho Marx, once offered the following advice: "Th=
e secret of success is honesty and fair dealing. If you can fake those, y=
ou've got it made."

If you aren't smiling, you may be cut out for economics. Students of the =
subject are trained to regard self-interest as the force that decides eco=
nomic choices. It is easy to imagine cases where cheating is advantageous=
=2E The economist's view is: others will see that the logic of the situat=
ion calls for cheating, so you had better cheat, too. This idea pervades =
the literature. But here's a disturbing thing: it may be having some effe=
ct. Nothing personal, but economists are an unpleasant lot.

The evidence is in a new paper by a team of one economist and two psychol=
ogists from Cornell University *. It reviews several behavioural studies.=
 In one, firstyear graduate students were asked to take part in an experi=
ment. They were given some money, and told to divide it into two accounts=
, one "private", one "public". Money in the private account was given to =
the student at the end of the ex- periment. Money in the public account w=
as pooled, multiplied by a factor of more than one (the exact figure vari=
ed), and then divided equally among all the students.

For society as a whole, as it were, the best thing is for the stu- dents =
to put all their money into the public account. That creates the biggest =
pie, which is then shared equally. But for each individual student, the b=
est thing is to put everything into the private account . That way, you g=
et back all your own stake, plus a full share of the pool provided by the=
 suckers. The study found that economics students contributed, on average=
, 20% of their stakes to the public account. Students of other subjects c=
ontributed 50%. The researchers then asked the students to explain their =
actions: had they worried about whether their decision had been fair? Nea=
rly all the non-economists said yes, they had worried. The response of th=
e economists was different:

More than one-third of Ihe economists either refused to answer the questi=
on regarding what is fair, or gave very complex, uncodable responses. It =
seems that the meaning of "fairness" in this context was somewhat alien f=
or this group. Those who did respond were much more likely to say that li=
ttle or no contribution was fair.

Another study involved a game played by an "allocator" and a "receiver". =
The allocator was given $10 and asked to divide the cash between himself =
and the receiver. The receiver could either accept the division (in which=
 case, both parties kept the sums proposed by the allocator) or refuse it=
 (in which case, both got nothing). Fairness calls for an equal split. Bu=
t what does self-interest tell the allocator to do? Only a non-economist =
could ask. The answer is: keep $9.99, and give the receiver one cent. The=
 receiver will not refuse because one cent is better than nothing (and se=
lf-interest does not understand spite). Note also that the game was playe=
d just once for each pair, so there was no reason for the receiver to ref=
use in the hope of prompting a better offer next time. As before, the stu=
dy found that economics students "performed significantly more in accord =
with the self-interest model" than non-economists.

Other studies have found the same. A survey asked 1,245 randomly selected=
 college professors how much they gave to charity each year. About 9% of =
the economics professors gave nothing; the proportion of professors in ot=
her disciplines giving nothing ranged between 1.1% and 4.2% (despite gene=
rally lower incomes than the economists). The median gift of economists t=
o big charities such as the United Way and viewer-supported public televi=
sion was substantially smaller than the median gift of non-economists.

The prisoner's dilemma a game where two players have to decide whether to=
 co-operate with each other or cheat has long been of great interest to e=
conomists. The key feature is that for each player, "defecting" secures t=
he best outcome regardless of what the other does. But if both players ac=
cept this logic and defect, they end up worse off than if they had co-ope=
rated. The Cornell team conducted an experiment involving 267 prisoner's =
dilemma games. Economics stu- dents defected 60% of the time; noneconomis=
ts defected 39% of the time.

Does training in economics make you mean or is it just that mean people a=
re somehow attracted to economics? To find out, the Cornell team did a fu=
rther experiment, to see whether students became more or less "honest" in=
 a hypothetical situation, after doing some economics. They compared thre=
e sets of students: the first took a course in mainstream microeconomics,=
 taught by an instructor with an interest in industrial organisation and =
game theory; the second took a similar course, but taught by a specialist=
 on development in Maoist China; the third took a placebo (astronomy). Ac=
ross a range of questions, the pattern was consistent: the first set cont=
ained the largest proponion of students who became less honest; next came=
 the second set; honourably in the rear were the astronomers, with the sm=
allest proportion of students who became less honest.

Perhaps, then, there is a public interest in curbing the study of economi=
cs. Or alternatively a conclusion that this column would prefer to endors=
e economics needs to take psychology more seriously. The fact is that peo=
ple do co-operate more than the self-interest model (useful though it is)=
 seems to predict. As the Cornell team points out, recent research sheds =
light on one reason for this.

Imagine a world in which people move from one prisoner's dilemma to the n=
ext (ie, the real world). If people can choose their "partners" freely, a=
nd if honest types can spot each other in advance, co-operators will be a=
ble to interact selectively with each other - and will therefore do bette=
r than cheats. Experiments have shown that people are surprisingly good a=
t telling co-operators and cheats apart, even on the basis of what seems =
to be limited information.

So there you have it: narrowly self-interested behaviour is ultimately se=
lf-defeating. Economics practised with that in mind could become the upli=
fting science. If economists can only incorporate a bit of psychology, th=
ey've got it made.

----

* Does Studying Economics Inhibit Co-operation?" By Robert Frank, Thomas =
Gilovich and Dennis Regan. Journal of Economic Perspecives, Spring 1993.