Katherine Mangu-Ward | October 9, 2007
Economists have long been bedeviled by the fact that
people consistently refuse to behave as economic models predict.
But now the Max Planck Institute of Evolutionary Anthropology in
Leipzig, Germany, has moved on to more compliantly rational
subjects: chimps.
They find that "unlike humans, chimpanzees conform to traditional economic models." When presented with a simplified version of the ultimatum game, in which one player decides how to divide a windfall between himself and another player, people frequently reject "unfair" offers (say, 8 for me, 2 for you). When an offer is rejected, both parties get nothing. Because if this human tendency, people tend to wind up with something close to a 50/50 split. But rejected small offers is not rational, and it infuriates economists when people consistently turn down small gains just to punish the other player for being unfair.
Unlike humans faced with these games, chimpanzee responders accepted any nonzero offer, whether it was unfair or not. The only offer that was reliably rejected was the 10/0 option (responder gets nothing). The researchers conclude that chimpanzees do not show a willingness to make fair offers and reject unfair ones. In this way, they behave like selfish economists rather than as social reciprocators.
In celebration of rationality, send a Monk-e-Mail to someone you love today.
Help Reason celebrate its next 40 years. Donate Now!
Try Reason's award-winning print edition today! Your first issue is FREE if you are not completely satisfied.
When presented with a simplified version of the ultimatum
game, in which one player decides how to divide a windfall between
himself and another player, people frequently reject "unfair"
offers (say, 8 for me, 2 for you). When an offer is rejected, both
parties get nothing. Because if this human tendency, people tend to
wind up with something close to a 50/50 split. But rejected small
offers is not rational, and it infuriates economists when people
consistently turn down small gains just to punish the other player
for being unfair.
In our defense, it could just be that humans understand that it's
more rational in the long run to demand fairness than to accept a
small, unfair amount.
In celebration of rationality, send a Monk-e-Mail to someone
you love today.
I despise linked web sites that I can't easily exit. That's
all.
Pride and ego is more important than your minor windfall. Include the costs of a bruised ego into your models and you will see rational behavior.
I think there may be a branch of Economists that assume that economic actors will not act rationally. Ludwig von Mises, Murray Rothbard, etc.
The role of schadenfreude in economics has long been overlooked. I see a PhD thesis and, perhaps, a Nobel Prize for the first economist to address this issue.
In this way, they behave like selfish economists rather than
as social reciprocators.
I wonder if we can get one appointed Fed Chairman? I wonder if we
already did?
You know what that chimp reminds me of?
I remember back in the 1960s. Me and the boys used to chase any
nigger drivin through our town. We didn't want to hang em without
do warning so we'd ring bells and use a megaphone to let them know
they weren't wanted. They got the message.
Today, it ain't so easy and many of the youngsters in the village
don't look too kindly on that. Ah well, we still sometimes will
ring a nigger's doorbell and just stare at'em or tell him about our
proud confederate past.
Humans are more capable of long term strategy. Most of us are more used to being on the recieveing end of the ultimatum game than the offering end. As such, we want to condition the other side to making fair offers, especially since there is little to lose in rejecting "unfair" offers.
But if human tendency to reject unfair offers leads to more fair
offers, isn't that more rational behavior?
KMW's article tells us that the chimps who make the choice often
get very small amounts of money, while the human who makes the
choice usually gets close to 50%. So who is more rational
again?
KMW's article tells us that the chimps who make the choice
often get very small amounts of money, while the human who makes
the choice usually gets close to 50%. So who is more rational
again?
Dan makes a good point here (gag, cough), except I don't think he
RTFA because the chimps used raisins.
Whatever the motivation (punishment, offense, reputation, etc.),
most times the offerer will give a "fair" offer because they are
afraid an unfair one will be refused. So the responder gets a good
deal. Therefore refusal of "unfair" offers creates motivation for
"fair" offers by the offerer and produces, overall, better results
for the responder than the "rational" chimp approach.
Chimps are OK by me even if they don't understand
"fairness".
Now those Hippie Dyke Bonobos ought to all be turned into
bushmeat.
Dan T.,
"The ultimatum game is an experimental economics game in which two
parties interact anonymously and only once, so reciprocation is not
an issue."
Can we PLEASE get the 3:06pm poster removed? I'm asking nice, but I don't need pathetic loser rants polluting the often reasoonable, insightful, and humorous discussion I find here. PRETTY PLEASE! WITH A CHERRY ON TOP!
Can we PLEASE get the 3:06pm poster removed?
Isn't that just a joke? If not, ignore it.
The perplexing irrationality of humans isn't their behaviour in
repeated games (where the rational strategy is definitely to reject
low-ball offers to benefit from higher offers over the
long-term).
It's their behaviour in single (one-time) games, where they still
reject small offers, even though there's no reason to punish an
anonymous counterparty with whom they won't interact again.
The only reason to turn down the offer is so they can walk away
confident that, even though they lost, they didn't let the other
guy win.
"That's pride fucking with you. Fuck pride. Pride only hurts,
it never helps."
isildur,
I know you're kidding and everything, but it turns out that your
little joke is how the Planet of the Apes got started.
The 3:06 poster is looking for attention, if you remove it or
punish him it will only excite him. Let his words hang around like
a stale fart.
I think this study shows more importantly that monkeys have no
shame, so they can be used for whatever purpose we desire.
Isn't that really what bothers capitalist pigs, that human beings
aren't tools of the establishment like the noble monkey?
(Channelling Dan T)
But if human tendency to reject unfair offers leads to more
fair offers, isn't that more rational behavior?
KMW's article tells us that the chimps who make the choice often
get very small amounts of money, while the human who makes the
choice usually gets close to 50%. So who is more rational
again?
Dan, congratulations on some twenty-percenters (I'm bumping you up
from 10% based on your comments here).
In a one-time, no-repeat deal where your only objective is to get
as much money as possible, the "chimp" response is the "correct"
one. But, if the game is played for several rounds, then insisting
on fairness becomes the optimal strategy.
I think that the human results in the one-shot deal can be
explained by the small amounts of money involved. We value fairness
and integrity -- it makes us feel good to think of ourselves that
way -- and so in the experiment at hand, a couple of bucks is
outweighed by other, nonmonetary considerations of more value. And,
the person handing out the money knows we are likely to reject
insultingly low offers, and thus rationally proposes something
close to a 50/50 split. Play it with the chimp doing the dividing,
and the human doing the accepting, and I think the humans would be
more likely to accept low offers.
Raise the stakes so a million dollars is being divided, and I'll
bet no one turns down an 80/20 or even a 95/5 split -- we value
fleeting feelings of integrity and self-worth, but we don't price
it at $200,000 or even $50,000. Quite a few liberals might turn
down a 99/1 split, while I suspect conservatives and libertarians
would grudgingly accept the $10,000.
But Russ R, it's harder for humans to disengage themselves from
the long-term "justice" concept and adopt the one-time strategy
because the longer term strategy is engrained. Pride is the
motivator, but it has a long-term benefit.
Humans also have more long-term planning potential, courtesy of
more developed abstract thinking ability. Chimps have to think
smaller, trapped as they are in little social bands where they have
no recourse but to cooperate with dominant troop members who may
well have killed their children or otherwise abused them.
Y'know? It'd be interesting to do Ultimatum Game experiments with
humans who have had to live under brutal despotic conditions most
or all of their lives, vs. people who have had more personal
experience of what's commonly considered "justice".
The perplexing irrationality of humans isn't their behaviour
in repeated games (where the rational strategy is definitely to
reject low-ball offers to benefit from higher offers over the
long-term).
It's their behaviour in single (one-time) games, where they still
reject small offers, even though there's no reason to punish an
anonymous counterparty with whom they won't interact again.
The only reason to turn down the offer is so they can walk away
confident that, even though they lost, they didn't let the other
guy win.
Basically, the experiment shows that people consider fairness to be
worth the cost of not accepting unfairness. You might even say that
it shows people are willing to sacrifice as individuals to help the
collective. Which might explain why we have civilization and chimps
do not.
You listen here, boy. Best to keep your mouth shut or I'll treat
you like I did them other dark skinned folk in the 60s.
I'm a proud man standin up for my race and my land. I don't need
self-hating whites like yourself tryin to push your way on to
me.
Raise the stakes so a million dollars is being divided, and
I'll bet no one turns down an 80/20 or even a 95/5 split -- we
value fleeting feelings of integrity and self-worth, but we don't
price it at $200,000 or even $50,000. Quite a few liberals might
turn down a 99/1 split, while I suspect conservatives and
libertarians would grudgingly accept the $10,000.
The beauty of it, though, is that only a crazy person is going to
offer you anything close to a 99/1 split with $1 million being
divided.
I'd bet that the more money at stake, the more likely the money is
to be divided equally.
In this way, they behave like selfish economists rather than
as social reciprocators.
Of course, the souless beasts.
Basically, the experiment shows that people consider
fairness to be worth the cost of not accepting unfairness. You
might even say that it shows people are willing to sacrifice as
individuals to help the collective. Which might explain why we have
civilization and chimps do not.
Um, no. I thought I agreed with you, Dan (that their refusals were
rational as it seemed to promote "fair" offers since offerers
realized that refusals would happen), but you are now tying this
into some grand altruism and collectivist scheme.
I totally disagree with that. What I think it is, fundamentally, is
jealousy and envy, two particularly powerful human emotions. Many
people, if offered a 90/10 split of a cool million, might turn down
the 100 grand because it just pisses them off so much that the
offerer is going to get 9 times as much for no rational reason. I
can't tell you how many people I've known that hated anybody with
more money then them, even if they were doing well themselves.
Totally irrational and very powerful.
This foundation of envy refusal causes the offerer to realize that
very "unfair" offers may be refused based on human nature and
motivates them to make much more "fair" offers as this is the best
strategy to get something rather than nothing.
Apply this logic to a presidential election. Assume that a
curious series of plane crashes (being investigated by the FBI,
natch) leaves you with a choice in the Republican primary of Ron
Paul or Mitt Romney, with Hillary Clinton having the Democratic
nomination locked up. Further assume you consider Hillary evil
incarnate, and Mitt slightly less so. Should a libertarian then
vote for Ron Paul, or for Mitt Romney as the electable lesser of
two evils for the general election?
Explain.
I've always wondered about this game and how they controlled for
it in humans. Most of these studies are done on college freshman -
stupid young men and women who are deluded by their parents and
educational institutions that the Universe is FAIR. It makes sense
they will walk away from a couple of bucks out of spite and expect
the same if they offered that. Now grad students are hungrier and
they might not be so quick to turndown a few extra bucks and might
assume the same from others. Maybe up the stakes to a $1000 pot and
see if the behavior changes.
The chimps were working with sweet, succulent raisin. Maybe they
were happy to get just one or two.
Humans seem to have a built in sense of fairness or justice --
either learned at a very young age or part of human nature.
This is rational in a social context. By punishing those who act
unfairly, you encourage society as a whole to be as fair and just
as possible. That's why people will reject excessively unfair
options on principle.
It's because we can envision future interactions that we are
willing to demand fairness even in a one time interaction. It's
better to get nothing now if it means that the people you deal with
will learn a lesson and act fairly in the future.
This is also why some people feel outrage about financial
inequality. It doesn't hurt them that CEOs get paid 100x more than
a worker. At most they'd get a small pay increase if things were
evened out. But it strikes at that sense of fairness.
Raise the stakes so a million dollars is being divided, and
I'll bet no one turns down an 80/20 or even a 95/5 split -- we
value fleeting feelings of integrity and self-worth, but we don't
price it at $200,000 or even $50,000. Quite a few liberals might
turn down a 99/1 split, while I suspect conservatives and
libertarians would grudgingly accept the $10,000.
The beauty of it, though, is that only a crazy person is going to
offer you anything close to a 99/1 split with $1 million being
divided.
I'd bet that the more money at stake, the more likely the money is
to be divided equally.
Good comment, Dan.
Assume the being doing the dividing is a chimp, or an autistic
person who doesn't understand human motivations, and the split HAS
ALREADY BEEN MADE. Do you take the $200,000? The $50,000? The
$10,000? Would you take a split of 99.9 / 0.1, since that is still
$1,000?
Re: your earlier comment, a few people value collective goods, but
most simply have it hard-wired that it usually is in their selfish
personal interest to insist on fairness -- and then cloak their
choice in the mantle of altruism so they can attract chicks (or
d00ds).
Um, no. I thought I agreed with you, Dan (that their
refusals were rational as it seemed to promote "fair" offers since
offerers realized that refusals would happen), but you are now
tying this into some grand altruism and collectivist
scheme.
I didn't say anything about a "scheme", but there must be something
about humans that makes us work collectively together, even if it
means sometimes making "irrational" decisions. For example, we
generally agree not to steal things, even though theft is very
rational as a one-time event from an individual point of
view.
I totally disagree with that. What I think it is, fundamentally, is
jealousy and envy, two particularly powerful human emotions. Many
people, if offered a 90/10 split of a cool million, might turn down
the 100 grand because it just pisses them off so much that the
offerer is going to get 9 times as much for no rational reason. I
can't tell you how many people I've known that hated anybody with
more money then them, even if they were doing well themselves.
Totally irrational and very powerful.
Call it envy if you like, but there must be a reason we've evolved
such an emotion, which is really just a reaction to unfairness. The
people who couldn't distinguish between a fair situation and an
unfair one seem to have died off...
PL,
That comment was meant to be read in the voice of Hudson from
Aliens. Note to self: don't use the less-than and greater-than
symbols in HTML-parsed comments for added hilarity.
Call it envy if you like, but there must be a reason we've
evolved such an emotion, which is really just a reaction to
unfairness.
If we evolved this reaction, wouldn't it be likely that the chimps
would show it as well? But they don't. It is most likely a cultural
thing. If I don't get a substantial piece of the pie, nobody gets
nothin'.
Remember, Dan, that this isn't about unfairness, it is about
inequality of reward. It is neither fair or unfair for one person
to get more of a windfall (i.e. unexpected, unearned money) than
another. Refusal means nothing for anyone. Is that more
"fair"?
For example, what if people who got 5 out of 6 lottery numbers
could make it so that anyone who got 6 out of 6 didn't get the
prize, because getting 5 out of 6 was so close that it "wasn't
fair". What would you think of that? And how would you feel if you
were somebody who rolled that one in a billion 6 out of 6 and then
had it denied by a 5 out of 6?
Chimps do not have markets and do not engage in trade.
Humans do....maybe markets and trade exist because we, unlike our
chimp cousins, have a sense of fair play.
Give a chimp a nut and he will take it...give a man a nut he will look for a lower price from a competitor.
Actually, it is rational to refuse smaller offers as long as you
can convince the other player you will do so before they make the
offer. This is a relaxation of the formulation anonymous coward
gives, but the only factor it relaxes is anonymity.
Being bound by one's nature to reject the low ball offers is an
effective way of persuading others that you'll reject the offer and
is more credible than a "rational" actor who will profess that
they'll reject the offer but when the question is put to them will
accept it. Thus, the irrational actor will get the better outcome -
it becomes rational to make sure you'll be irrational during the
game.
If you relax the rules futher to allow binding contracts, the
optimal strategy would be to enter into a binding contract that
you'll pay a 3rd party twice what you recieve if you accept less
than some amount very close to all of the goods being divided and
make sure the other player knows it. The person dividing the goods
would give you the amount specified in the contract since they know
they won't get anything otherwise since you'll reject to avoid
paying the 3rd party.
In single round ultimatum game variations, the first party to lock
their behavior in such a way as the other party knows it and is
convinced gets to pick how the goods are divided. In the anonymous
variation, if the offer-maker assumes that the other player is
bound to reject an unfair offer, they're the first one to have
locked in, so they've effectively dictated the sum; if the
offer-maker does not believe the other person is locked in, they
get to lock in first and make a slightly greater than zero offer.
The communications barrier is what allows the non-pareto efficient
outcome to occur by allowing the responder to think they've locked
in when the offer-maker doesn't or creating confusion about the
amount at which a player has locked in. With open communication,
the refusal outcome can be avoided entirely.
I fucking hate this as an example of "irrational" people.
People don't reject offers because they're irrational. They do so
because they value the schadenfreude more than the money they've
been offered. That is not per se irrational.
Why people value the schadenfreude at non-zero amounts delves into
psychology and culture at least as much as economics. So when this
game is used as such an example, it mostly reminds me how dismal
the science behind economics can be.
Ah, Dr. K rises to the challenge. Your Nobel prize and 10 million kroner are on the way.
Here I thought economics specifically and science in general
were supposed to describe behavior, not label it as to whether it
agrees with the theories.
If observed behavior doesn't correspond with the theory, then the
theory needs revising.
"Fairness" is a social construct and varies tremendously across
human history and cultures.
In this example, I believe that self-interest is the ultimate
driver of both parties, with each calculating, in essence, what
they can "get away with." Of course, it's also quite
artificial.
Thank you, Sir Liberate. I will gladly donate the money back if the Foundation if they can stop journalists from using this as an example of irrational economic behavior in people.
I think that all of human behavior can be traced to schadenfreude. The Germans are so wise.
My biggest long term worry about libertarianism is that our
brain is simply not built for it.
They should run this study on infants to see if this
counterintuitive economics in ingrained in the human brain, or
simply the result of cultures that emphasize fairness.
I know there are studies that showed babies are naturally
altruistic. They need to see if babies have schadenfreude.
To me this does not suggest economics is a science of
counterintuitive truths, but that economics may be a fundamentally
flawed method to understand human behavior. Give me ammunition next
time someone brings up Freakonomics or Tyler Cowen.
Oh, and shouldn't this go into the category of "spite"?
"Schadenfreude" does not necessarily carry the connotation that one
is losing something to hurt others, merely taking pleasure at
other's misfortune.
To me this does not suggest economics is a science of counterintuitive truths, but that economics may be a fundamentally flawed method to understand human behavior. Give me ammunition next time someone brings up Freakonomics or Tyler Cowen.
"Rational" (the economic sense, which normally is concerned with
outcomes of choices but doesn't care about process, is different
from the popular sense which includes process) people are to
economics as ideal gases are to thermodynamics. Despite being
absent in reality, the mechanisms they describe are indeed present,
they're used as the framework on which more complicated models are
built, and depending on the situation, you can often get reasonably
accurate results using them as models without modification,
especially when you just need to determine the sign of a change and
not the magnitude.
My biggest long term worry about libertarianism is that our
brain is simply not built for it.
How is a sense of fairness, which is necessary for functioning
markets by the way, anti-libertarian.
Lets take this example and put it in a real world scenario. Lets
say a property owner offers a developer some land for a price. The
property owner values his land at its potential value after it is
developed. If he then takes the whole profit from the developer the
developer will walk and no one gains. the property owner has to
sell his property at a price that will leave a margin for that
developer to make money or he will not sell his property.
Without a sense of fairness, empathizing with the other party, any
transaction of this sort is an impossibility. Two monkeys will
quickly either go broke or brake themselves leaving no resources to
negotiated over.
MattXIV:
You're right, but a lot of people treat economics as faith (DEMAND
KURV) and not the useful tool (with constraints) it actually is.
Those that base economic theories on their deeply held views on how
individuals behave ("communism fails because it goes against human
nature"), they overstate their case.
It just reminds me how people were certain their instruments were
faulty as they picked up data that reflected the discrete nature of
mass rather than the continuous nature they expected. Maybe the
micro-est of microeconomics (nanoeconomics?)needs to ceded over to
neuroscience alone.
Joshua Corning:
I don't think our brains are incompatible with libertarianism
considering current research, but just as there is some
inconclusive research that shows our brains may be hard-wired for
faith, I worry that some scientist will eventually prove that the
reason that there were very few long-term successful societies that
we can agree upon as libertarian is because there is something
about how we are wired that is incompatible with
libertarianism.
Its easy to ignore anthropologists, historians and economists,
because you can challenge some assumption or some methodology as
incomplete.
When you get to the harder sciences, its hard to contest the
results. Neuroscientists seems to be blending increasingly with
economics and I don't know how it will turn out.
I already sometimes feel like the bearded guy spouting Marx when I
talk about libertarianism, I don't want to become the Churchy guy
preaching creationism.
Luckily, we are nowhere near that state, and there are a lot of
good reasons why it may never happen, but it does concern me a
little bit in the back of my mind.
Way too long-winded, sorry about that.
Empathy may be the key here. If someone offers a 95/5 split of $100, many people believe it says that the offerer thinks he's 20 times better than they are. We tend to take things personally. Most people don't want to accept crumbs because of their own self image. Chimps have no self image and so act more concretely and in the here and now.
A couple comments.
This is almost a tragedy of the commons sort of problem, since the
"reward" did not belong to either parties at the beginning, so the
first person didn't really earn anything.
I wouldn't necessarily call it irrational to reject something even
in light of it being a one time thing. If the two parties come from
similar cultures, the first one can imagine what they would do if
offered a much smaller amount and decide that a more equitable cut
would be more likely to be accepted.
I think there is an assumption that the amounts offered don't make
a significant impact to the people's lifestyle. Otherwise, I would
agree that a large amount would find more tolerance for an unequal
split.
To put it another way, negligible amounts would probably need to be
50/50 for a human to agree to it. They don't gain much by taking a
little, so it might as well be fair. On the other hand, if whatever
the amount given is, for example, more than the person makes in a
year (or more), it would be difficult to turn down even if percent
was small.
It's similar to leaving a waiter a 25 cent tip. It's going to make him mad, because he perceives it to be the insult that it is probably intended to be.
anomdebus-
I think you're right. If I turn down $1 because I felt I should
have gotten $5 (assume $10 is being divided) then I haven't lost
much. My irrational decision was a cheap one.
OTOH, I'd have a hard time turning down $100 because somebody was
dividing $1,000. In that case, irrationality would be
expensive.
So maybe some economic principles do apply here: People will
indulge in lots of irrationality if it's cheap (or if the costs are
hidden), and they will engage in less irrationality when it's
expensive. See, the demand curve still works!
I just thought of a really cool and insightful way to look at
this. But I'm not going to waste it on you guys.
Neeener neeener.
I'll bet you'd be hard pressed to find a chimp who would send money to the Ron Paul campaign.
Edward, all the rich chimps are behind Giuliani. Everyone knows
that.
Anyway -
OTTO: Look, we can split the money from the car 60/40.
BUD: Who gets the 60, kid?
OTTO: Well, I thought since I found the car...(sound of gun
cocking)... that you do.
If observed behavior doesn't correspond with the theory,
then the theory needs revising.
Huh. Ive always gone with "Theory is always right. Sometimes
reality is wrong."
most modern (and wrong) economic theory is based on the concept
of the "economic man" that acts rationally.
any trader (that's how i make my $$$) knows that economic man
doesn't exist (well, maybe warren buffet, but other than
that).
man is irrational, emotional, herdlike, and does not act in ways
that would be optimal in game theory. not even CLOSE. Dr. brett
steenbarger has done a lot of work in studying how people actually
act in economic situations (while mostly concentrating on trading,
the ideas are valid in general in terms of economic actions)
that's why "efficient market theory" and other such hogwash was (or
is ) SO sexy to academics, and so ridiculously absurdly provably
ridiculous in the real world.
Yup I think you are all missing the point here.
There is no market for getting free stuff with one person choosing
how to split the free stuff and the other party choosing to veto
that split or not.
What we have is a an owner and a purchaser. The owner is the one
with the choice of the split and purchaser is the veto.
It seems fairly obvious to me is that Chimp is a terrible shopper
who will take what ever deal he can get at the same time a terrible
salesmen who will over price all his goods and destroy his consumer
market.
A human is both able to price to sell and be a discriminating
shopper both abilities are essential for a functioning market. One
should note that Chimps do not have markets and we humans have a
multi-trillion dollar global economy that has raised our life spans
and reduced our child mortality.
To claim that humans are irrational for behaving in a manner that
allows markets to exist at all is a miss representation of ration
and the bases of markets.
Can a wild chimp (one not taught by humans) even count to 2? Also what is used as the currency for chimps? Food? Well I think it's pretty obvious if you put any amount of food in front of an animal they will accept it. The thing I find most interesting about the ultimate game is that the giver wouldn't automatically offer a 50/50 split knowing that that offer will most likely to be accepted therefore gaining some money for themselves.
Site comments/questions:
Media Inquiries and Reprint Permissions:
(310) 367-6109
Editorial & Production Offices:
3415 S. Sepulveda Blvd.
Suite 400
Los Angeles, CA 90034
(310) 391-2245