Bryan Caplan  

Gintis on the Evolution of Private Property

Selgin on Gorton... On Bastiat and the Edifice Com...
Herb Gintis' "The Evolution of Private Property" (Journal of Economic Behavior and Organization, 2007) is one of the most fascinating articles I've read in years. (ungated draft)  I'm slightly jealous because I've planned to write a similar piece for a decade, but Gintis did a much better job than I would have done. 

Key ideas:

First, the endowment effect - our tendency to value our stuff extra just because it's ours - is widespread throughout the animal kingdom.  Even butterflies have it!
Among the many animal behaviorists who put this theory to the test, perhaps none is more elegant and unambiguous than Davies, who studied the speckled wood (Pararge aegeria), a butterfly found in the Wytham Woods, near Oxford, England. Territories for this butterfly are shafts of sunlight breaking through the tree canopy. Males occupying these spots enjoyed heightened mating success, and on average only 60% of males occupied the sunlit spots at any one time. A vacant spot was generally occupied within seconds, but an intruder on an already occupied spot was invariably driven away, even if the incumbent had occupied the spot only for a few seconds. When Davies "tricked" two butterflies into thinking each had occupied the sunny patch first, the contest between the two lasted, on average, ten times as long as the brief flurry that occurs when an incumbent chases off an intruder.
And non-human primates, of course:
In general, the taking of an object held by another individual is a rare event in primate societies (Torii, 1974). A reasonable test of the respect for property in primates with a strong dominance hierarchy is the likelihood of a dominant individual refraining from taking an attractive object from a lower-ranking individual. In a study of hamadryas baboons (Papio hamadryas), for instance, Sigg and Falett (1985) hand a food-can to a subordinate who was allowed to manipulate and eat from it for 5 min before a dominant individual who had been watching from an adjacent cage was allowed to enter the subordinate's cage. A "takeover" was defined as the rival taking possession of the can before 30 min had elapsed. They found that (a) males never took the food-can from other males; (b) dominant males took the can from subordinate females 2/3 of the time; (c) dominant females took the can from subordinate females 1/2 of the time. With females, closer inspection showed that when the difference in rank was one or two, females showed respect for the property of other females, but when the rank difference was three or greater, takeovers tended to occur.
Second, this ubiquitous self-enforcing territory/property cannot arise without an endowment effect:
Consider, for instance, the sparrows that built a nest in a vine in my garden. The location is choice, and the couple spent days preparing the structure. The nest is quite as valuable to another sparrow couple. Why does another couple not try to evict the first? If they are equally strong, and both value the territory equally, each has a 50% chance of winning the territorial battle.Why bother investing if one can simply steal (Hirshleifer, 1988)? Of course, if stealing were profitable, then there would be no nest building, and hence no sparrows, but that heightens rather than resolves the puzzle.
Third, given various technical conditions, evolution affirmatively selects for an endowment effect.  Read the paper for details.

My favorite part is Gintis' discussion of (harmless) psychological experiments on human toddlers and preschoolers:
Long before they become acquainted with money, markets, bargaining and trade, children exhibit possessive behavior and recognize the property rights of others on the basis of incumbency. In one study (Bakeman and Brownlee, 1982), participant observers studied a group of 11 toddlers (12-24 months old) and a group of 13 preschoolers (40-48 months old) at a day care center. The observers found that each group was organized into a fairly consistent linear dominance hierarchy. They then cataloged possession episodes, defined as a situation in which a holder touched or held an object and a taker touched the object and attempted to remove it from the holder's possession. Possession episodes averaged 11.7/h in the toddler group, and 5.4/h in the preschool group.

For each possession episode, the observers noted (a) whether the taker had been playing with the object within the previous 60 s (prior possession), (b) whether the holder resisted the take attempt (resistance), and (c) whether the take was successful (success). They found that success was strongly and about equally associated with both dominance and prior possession. They also found that resistance was associated mainly with dominance in the toddlers, and with prior possession in the preschoolers. They suggest that toddlers recognize possession as a basis for asserting control rights, but do not respect the same rights in others. The preschoolers, more than twice the age of the toddlers, use physical proximity both to justify their own claims and to respect the claims of others.
Fortunately for me, Gintis doesn't discuss the application that most interests me: black markets.  On a typical Hobbesian view, exemplified by Holmes and Sunstein's The Cost of Rights, markets where government not only refuses to enforce property rights but actively forbids property rights should not exist.  Plainly they do.  How are black markets possible?  Because of the endowment effect - the very psychological trait that explains almost all the property and territory that's ever existed.

HT: GMU Ph.D. student Eric Hammer, who's co-authoring the black market piece with me.

COMMENTS (12 to date)
HH writes:

Never understood why the endowment effect is such a puzzle. It makes perfect sense to me if you consider informational asymmetry.

If I own something, I've (usually) had an opportunity to examine it carefully and be familiar with its features and its value. Anything I don't own (such as items offered to me in a trade) are potential lemons. It makes sense that I'd reject a seemingly even trade: I KNOW what my property is like, while what you're offering might be a lemon in disguise.

You can then see why evolution would select for people who have an endowment effect: they're less likely to be cheated.

Keith writes:


Yes, an object you don't have could be a lemon. But it could also be something awesome. By itself, lack of information shouldn't change the willingness to make a seemingly even trade.

Of course, humans are also risk-averse. Maybe if you threw that it, it and lack of information would together explain the endowment effect.

Gintis is interesting, among the most interesting Marxists I know;

My colleagues and I found dramatic evidence of this positive relationship between markets and morality in our study of fairness in simple societies—hunter-gatherers, horticulturalists, nomadic herders, and small-scale sedentary farmers—in Africa, Latin America, and Asia. Twelve professional anthropologists and economists visited these societies and played standard ultimatum, public goods, and trust games with the locals. As in advanced industrial societies, members of all of these societies exhibited a considerable degree of moral motivation and a willingness to sacrifice monetary gain to achieve fairness and reciprocity, even in anonymous one-shot situations. More interesting for our purposes, we measured the degree of market exposure and cooperation in production for each society, and we found that the ones that regularly engage in market exchange with larger surrounding groups have more pronounced fairness motivations. The notion that the market economy makes people greedy, selfish, and amoral is simply fallacious.

Finch writes:

Can anyone point to research on endowment effects in small children? I notice young children seem very reluctant to trade (not least because they also seem very eager to cheat). They need to be taught to consider that trade might be a good idea.

My guess is that the possibility of a trade being reasonably fair is a very recent phenomenon for humans, so the endowment effect made a lot of sense for most of human evolutionary history.

Hazel Meade writes:

I think HH is right here.
And the only information about it might not be "is it a lemon?", but also knowing quirks about it's features that you would have to re-learn on an unfamiliar object. The particular hammer that you posess may be balanced in a way that you've learned to control, while a new one would feel wierd and you'd waste time getting back to the same proficiency level.

Daniel Klein writes:

A related important piece by David Friedman, perhaps already familiar to all, but just in case, is here.

Tom West writes:

As pointed out by the article, it's quite possible that *not* having an endowment effect might result in species extinction.

Through survivor bias, we only observe species who not only possess the endowment effect characteristic, but have a low enough 'defector' rate that the odd mutant who doesn't possess it is quickly extinguished rather than taking over.

Eric Hammer writes:

I find the informational asymmetry reason for the endowment effect somewhat uninspiring myself. I think the source of endowment effects might be closer to the effort humans must put into producing things, or the probability of finding them.

The sparrow example covers the former pretty well, especially when you consider that there is a pretty short window in which one can lay eggs and have them hatch and grow fully before the season is gone. If a sparrow builds a nest only to just give it up when a competitor wanders by, it might not have time to build another in the relevant window. As a result, he loses not just the nest, but the opportunity to raise chicks that year at all. The stealing sparrow apparently didn't value that enough to go to the effort of building in the first place, and so is not likely to be willing to fight as hard.

As to the latter, if one can't produce something but has to just randomly find it, and there is a low probability of finding it, there is also a greater incentive to hold onto it once you have it, to the extent that it is likely you will never come by it again. In this context the information asymmetry is probably more relevant, in the sense that we often don't realize how much we liked something till we lose it. The feeling of loss, missing something we can't easily replace, is felt pretty early in life, and is something we want to avoid. We don't miss what we don't have, and so once we have it, the knowledge of what the loss will feel like makes us value having it more.
Of course, we often want things more when we don't have them than when we do, so that's something to figure out too. At the very least, that makes a lot of sense from an evolutionary perspective too. Why specific things matter more when we have them while perhaps a class of things matters more when we don't is pretty interesting.

HH writes:

"I think the source of endowment effects might be closer to the effort humans must put into producing things, or the probability of finding them."

I have to disagree, for two reasons:
1. The endowment effect has been demonstrated to be just as strong for items that took no effort to obtain (gifts, etc). (see wasp note below)

2. While there is evidence that humans & animals succumb to the sunk cost fallacy (which is what this is) it couldn't have evolved that way. Natural selection would have weeded out those who expend significant energy defending property that isn't worth very much but took some effort to obtain.

Richard Dawkins tells a story of wasps that defend lairs they prepared with an intensity proportional to the amount of time they've spent preparing the lair. That would seem to agree with Eric Hammer's statement above, but Dawkins explains that it's not an actual sunk cost fallacy: the effort put into the lair is proportional to the value of the lair, and thus the effort put into fighting over it should be proportional to the effort put into preparing it. If this were the case for humans, the endowment effect would be significantly weaker for labor-intensive items. That's not been demonstrated.

Eric Hammer writes:


I think on your point one, gifts would be covered under my second option, that of found/randomly occurring objects. If it might be a long time before some valuable found object is likely to be found again, it still holds. In effect, effort/low likelihood of rediscovery are equivalent in this case.
Further, I don't think that evolution necessarily would work on the same margin that we are looking at with more modern humans. While the sunk cost fallacy is a thing, just as often it really is a question of "Is it easier to fight to keep what I have, or make a new one?" If that is often the case, and the critter in question doesn't have advanced abstract thinking abilities, it comes down to a basic rule of thumb. An animal that fights harder in general for things it already has might well come out ahead, even if some of those fights are not really necessary from the stand point of the sunk cost fallacy, relative to the animal that just lets everything go.

As an example of that in action, I submit the entire mustelid family. Most other animals don't bother fighting with them, because they are crazy and always fight (a little hyperbole, but only a little). The fact they are relatively primitive suggests that actually is a pretty good strategy if you are willing to really dedicate yourself to it, and don't mind not living in groups. Being the craziest guy in prison isn't a bad strategy, though being the second craziest is a very bad one.
Basically, what I mean to say is that many strategies can work, along many different margins. The sunk cost fallacy causing inefficient behavior is not enough to say that it should be driven out by evolution, you have to show that there are many alternatives that are lower cost. If the only way to really optimize away from fighting over sunk costs is a large and super expensive brain, sunk costs seem like a pretty acceptable cost.

Eric Hammer writes:

Another quick thought I had after I walked away:

One huge difference between modern humans and animals is that we live in relative plenty. Most things living in a state of nature are closer to subsistence, and thus face a very big discontinuity in their utility curves, that of dropping below what is required to live at all. That might lead some animals to be very willing to fight over certain things and largely indifferent over others, based on relative need or scarcity. So say a herd of horses might be quite willing to fight a different herd for water rights in very arid areas, but tolerate their presence in areas with a regularly flowing river, for example.

That raises an interesting quirk for a tool using species like humans. While food, water and shelter are important to us just like any animal, we don't necessarily have the tools to acquire them built into our bodies. Rather, we often have to acquire or build strange and unnatural objects as an intermediate step. This creates a new category for us to care about: Random Useful Object. Throw this need for RUOs into the evolutionary strategy of "fight extra hard for scarce things you have already" and you can see why 60,000 years later we become very attached to a mug that we have held for a few seconds. Our feeling brains don't differentiate well between the types of objects we want as tools, only between categories like "tools, food, water, etc." As a result we rely on our more modern, abstract processing centers to steer us away from such emotion bases mistakes.

Glen S. McGhee writes:

Gintis has Marxist sympathies, so I wonder where the underlying dialectic is -- the internal relation that simultaneously gives rise to both division of labor AND private property.

Comments for this entry have been closed
Return to top