craschworks

http://www.economist.com/science/displayStory.cfm?story_id=1045223

Human behaviour
Deviations from the mean
Mar 21st 2002
From The Economist print edition
AP

Biologists are helping economists to explain why humans are not always selfish

AT SOME point during their education, biology students are told about a
conversation in a pub that took place over 50 years ago. J.B.S. Haldane, a
British geneticist, was asked whether he would lay down his life for his
country. After doing a quick calculation on the back of a napkin, he said
he would do so for two brothers or eight cousins. In other words, he would
die to protect the equivalent of his genetic contribution to the next
generation.

The theory of kin selection-the idea that animals can pass on their genes
by helping their close relatives-is biology's explanation for seemingly
altruistic acts. An individual carrying genes that promote altruism might
be expected to die younger than one with “selfish” genes, and thus to have
a reduced contribution to the next generation's genetic pool. But if the
same individual acts altruistically to protect its relatives, genes for
altruistic behaviour might nevertheless propagate.

Acts of apparent altruism to non-relatives can also be explained away, in
what has become a cottage industry within biology. An animal might care for
the offspring of another that it is unrelated to because it hopes to obtain
the same benefits for itself later on (a phenomenon known as reciprocal
altruism). The hunter who generously shares his spoils with others may be
doing so in order to signal his superior status to females, and ultimately
boost his breeding success. These apparently selfless acts are therefore
disguised acts of self-interest.

All of these examples fit economists' arguments that Homo sapiens is also
Homo economicus-maximising something that economists call utility, and
biologists fitness. But there is a residuum of human activity that defies
such explanations: people contribute to charities for the homeless, return
lost wallets, do voluntary work and tip waiters in restaurants to which
they do not plan to return. Both economic rationalism and natural selection
offer few explanations for such random acts of kindness. Nor can they
easily explain the opposite: spiteful behaviour, when someone harms his own
interest in order to damage that of another. But people are now trying to
find answers.
Good for you, good for me

The favoured tools for exploring seemingly odd behaviours are called
public-goods experiments and ultimatum games. In a public-goods experiment,
each member of a group of subjects is given a financial stake and asked to
use this money to contribute to a common pool. This pool is then expanded
by the experimenters and redistributed to individuals in the group.

One recent public-goods experiment was conducted by Ernst Fehr from the
University of Zurich and Simon Gächter at the University of St Gallen in
Switzerland. Sets of four students were given around $20 to participate in
a group project. They were told that they could keep any money they did not
invest. The rules of the game were that every $1 invested would yield
$1.60, but that these proceeds would be distributed to all group members-in
other words everyone would get 40 cents. Thus if everyone invested, each
would walk away with $32, and the public good would have been served. If,
however, only one person invested, that sucker would take home a mere $8.

The experiment was run first as a series of six rounds in which the
participants were anonymous, investment decisions were taken
simultaneously, and no two participants interacted more than once. At the
end of each round, each player was told how other members of the group had
invested and what the individual pay-offs would be. In this version of the
game, people quickly learned not to invest. By the sixth round,
contributions to the pool had dwindled to nothing.

The second version of the experiment, though, gave the individuals the
opportunity to punish each other financially at the end of each investment
round. At a cost of $1, a group member could anonymously fine another $3.
Of the 240 participants in the experiment, 84% punished at least once and
around 9% punished more than ten times. Most punishment was imposed by
above-average contributors (deemed co-operators) on below-average
contributors (deemed defectors or free-riders). Punishment was related to
notions of fairness; the amount of punishment given was directly related to
how far a free-rider's contributions deviated from the group's average. And
punishment had an effect. It substantially increased the amount that was
invested in the public good: more than 90% of the participants contributed
more money when punishment was a possibility.

These punishments might look spiteful. In fact, they were altruistic. This
is because, at the end of each investment round, the participants were
swapped between groups so they would never invest with the same people
again. Participants were therefore paying to punish someone with whom they
would not interact again. Altruism, in other words, emerged in the second
form of the game, as punishment, but not in the first.

The idea that people sometimes value “fairness” over personal gain was also
suggested by an ultimatum game run last year by Joseph Henrich, an
anthropologist at the University of Michigan's business school, and his
colleagues. In an ultimatum game, a participant is given what is for him
the equivalent of a day's wages, and asked to contribute a portion of this
to a second, anonymous person whom he will never meet again. The recipient
can accept or reject this offer, but if he rejects it, neither party
receives any money.

The logic behind an ultimatum game suggests that, if people wish to
maximise their resources, as standard economics assumes, a recipient will
accept any offer made by a donor. Conversely, the donor will always offer
as little as possible. But that is not how things turn out. Offers are
guided by notions of what is fair, as are rejections. The researchers found
that offers ranged from 22% to 58%, and offers higher than 50% were
sometimes rejected.

This evidence and the findings of public-goods experiments led the
anthropologists to conclude that acceptance and rejection were strongly
linked to feelings of fairness and reciprocity in addition to material
benefits. People reward those who act in a co-operative manner and punish
those who do not-even if such behaviour costs them something personally.

That is, of course, only a scientific confirmation of a human commonplace.
And when a new phenomenon is recognised by science, a name always helps. In
a forthcoming paper in Human Nature, Dr Fehr and his colleagues argue for a
behavioural propensity they call “strong reciprocity”. This name is
intended to distinguish it from reciprocal altruism. According to Dr Fehr,
a person is a strong reciprocator if he is willing to sacrifice resources
to be kind to those who are being kind, and to punish those who are being
unkind. Significantly, strong reciprocators will behave this way even if
doing so provides no prospect of material rewards in the future.
Turn and face the change

That does not, of course, explain how strong reciprocity evolved in the
first place. The modern theory of natural selection sees the process
working on individuals: if you are “selected out” (ie, killed), your genes
cannot contribute to the next generation. Most selective pressures
(predators, sexual competitors and so on) work at the individual level.
Those that might wipe out whole groups at a time, such as infectious
diseases, are not susceptible to collective action, so might as well be
individual pressures. However, if what kills individuals frequently kills
entire groups, but is something that collective action might successfully
combat, the unit of selection might become the group. In the case of human
beings, who (thanks to language) can collaborate in detail, a wide range of
group-threatening risks-wars, famine, environmental catastrophes and so
on-might be susceptible to such collective action. Groups containing strong
reciprocators might be better adapted to survive, particularly since their
behaviour coerces even the selfish into action that favours the common
good. Genes for strong reciprocity would thus spread.

The idea of strong reciprocity explains both previously inexplicable
altruistic acts and the existence of spite. That makes it theoretically
attractive. Its boosters also think that it might have practical
applications. A pair of economists, Samuel Bowles at the Santa Fe Institute
in New Mexico and Herbert Gintis of the University of Massachusetts in
Amherst, has suggested that if policymakers want to achieve certain public
goods, such as the sharing of common resources, it might be useful to
provide opportunities for the public-spirited to punish the free-riders in
society. This kind of thing works in small fishing communities, where
free-riders are punished by social exclusion. Extending these sorts of
penalties to society at large could pit the better side of human nature
against its other half, and make things better for everyone.