The Dark Side of Cooperation

Cooperation is a popular topic these days.  For example, see this Science review article and this Nature book review of de Waal’s The Age of Empathy:

A repeated foil throughout is Gordon Gekko, from the 1987 movie Wall Street, who reiterates in various forms the basic credo that “greed is good”. … [de Waal's] main political message is that we should not continue to harp on about evolution justifying only the selfish side of human nature, although of course that exists. He urges that we must also capitalize on the empathetic and cooperative attitudes that evolution has equipped us with, writing: “A society that ignores these tendencies can’t be optimal”.

Here is a New Scientist book review:

Given all that we know about empathy in animals, why do so many persist in seeing ours as a dog-eat-dog world? De Waal chalks it up to what he calls “macho origin myths”, which insist that “our species has been waging war for as long as it has been around”. But humans have shown empathy for as long as we’ve been around too.

Many stories discuss recent research on how cooperation can be sustained by norms to punish non-cooperators, and punish those who don’t punish non-cooperators, etc.  For example, New Scientist:

The temptation to freeload – reap the rewards without contributing anything – often leads to rapidly disintegrating cooperation.  Previous research found that cooperation is promoted by allowing players to punish freeloaders: cooperative players would pay a small cost that enables them to inflict a loss on the offender.

The unstated moral behind most media stories on our biological instincts to cooperate seems to be that we would do better to empower and emphasize these instincts.  Such as, oh, taxing carbon, and shaming those who don’t tax carbon.

But such stories mostly ignore the dark side of cooperationpro-cooperation instincts rely on dangerous conformity. Yes groups can be better off if individuals can see who do things that hurt the group overall, and punish those folks, and punish those who don’t punish them, etc.  But our evolved instincts about which are the individual actions that actually hurt others might be quite out of whack.

For example, in big disasters like hurricanes, certain goods like gas, wood, water, or food become especially valuable.  While natural selfish reactions lead to higher prices for these key items, humans clearly evolved to see this behavior as uncooperative; we resist such price rises, and want to punish those who allow them.

Perhaps this made sense for our distant ancestors, but today it is counter-productive.  If these goods are not allocated by price, they will instead be allocated by standing in lines, personal connections, etc., processes that are consistently worse at giving goods to those who value them the most, and do worse at creating incentives to prepare for such scenarios.

But even when some of us realize that disaster price rises are actually cooperative behavior, pro-”cooperation” instincts get in the way of acting on this insight.  If others mistakenly intuit, that we are suggesting acts they consider uncooperative, they will punish us for such suggestions.  They will similarly punish us if their usual conformity rumor mill, not exactly designed for subtle analysis, tells them our suggestions are uncooperative.  And even if others agree that we are actually suggesting cooperation, they may still have to punish us to avoid being punished themselves by others for failing to punish non-cooperators.

The problem is that evolved cooperation instincts reward supporting behavior that most people feel is cooperative, and not what is actually cooperative.  In novel situations, where our ancient instincts and simple rumor mills are poor guides, ordinary folks can be quite mistaken about which actions help vs. hurt everyone.  In this case our “cooperative” instincts can make it much harder to share info about what actually helps or hurts.  In contrast, if it is accepted that we will each act selfishly, cooperating selfishly via exchange and contract, we can more easily rethink and relearn what actions are actually helpful in our new changing world.

GD Star Rating
loading...
Tagged as: , , ,
Trackback URL:
  • Peter Twieg

    I agree that when allocation by markets are a feasible alternative to allocation by empathy (or some similar emotional mechanism), allocation by markets should be preferred… but it strikes me as unreasonably strong to argue that this makes these mechanisms a net minus in the modern day. There are so many opportunities for purely selfish individuals to engage in moderate-return negative-sum activities in daily life (crimes, etc.) that it’s hard to imagine that the benefits of unselfish abstinence outweigh the cost of the very few situations we encounter where misguided cooperation leads to worse outcomes than what would be achieved by purely selfish actors.

    Why not simply argue for empathy guided by an understanding of markets and various cognitive biases?

    • Peter Twieg

      That should read, rather, “it’s hard to imagine that the benefits of unselfish abstinence don’t outweigh the cost…”, my bad.

    • http://hanson.gmu.edu Robin Hanson

      With ordinary crime, such as robbery, there is a direct victim that usually makes it clear the loss outweighs the gain. The problem arises from relying on weaker intuitions or rumors about less clear less direct harm to a wider society.

    • anon

      “moderate-return negative-sum activities in daily life (crimes, etc.)”

      Actually, rent-seeking behavior is much more relevant than ordinary crime here. While it’s true that rent-seeking lacks a “direct victim”, I’m not sure why this makes its harm to society any less clear.

      Wasteful arms-races of all kinds belong in the same category: there’s no “direct victim” either, but the harm is nonetheless clear.

      I also find it surprising that Robin didn’t mention conflict with out-groups as an even darker side of the human cooperative instinct; modern society seems to have been fairly successful in avoiding the harmful effects of this evolved tendency.

      • http://hanson.gmu.edu Robin Hanson

        Yes rent-seeking is harmful behavior which our “cooperation” intuitions often admire, and cooperating to hurt out-groups is another important dark side of cooperation.

  • http://t-a-w.blogspot.com/ Tomasz Wegrzanowski

    I’m curious if there are any empirical studies showing to what extent crisis-time price controls are harmful, by comparing crises with and without such price controls. I’m not entirely convinced that the damage is high – standard economic analysis of such cases applies only to long term equilibria, and what you have in crisis is a bunch of panicking misinformed people, and no time for market to reach any sort of balance.

    So what do the facts say?

    • Michael Turner

      It’s a good question. Working from admittedly-vague memory: Lenin instituted market reforms (his New Economic Program) partly in response to post-civil war famine conditions. And it worked. E.F. Schumacher et al. later cited this same market success in strong-arming Konrad Adenauer to end post-WW II food rationing in Germany, with similar effects.

      Rationing can create huge incentives for hoarders, who are willing to take the risk of censure, criminal charges, even summary execution (the last was a real risk when facing Bolsheviks.) The resulting black market for necessities must pass along their social/criminal risk-premia costs to their consumers. However, the resulting high prices are taken as a (sometimes deceptive?) signal of high profit, drawing ever more players into the market to speculate on the value of the rationed necessities. This bubble-like vicious cycle that can suck the necessities of life away into stagnant inventories faster than rationing organizations can hand out goods.

      Many years ago, in George Debreu’s mathematical econ course at U.C. Berkeley, I noticed he liked to use the example of necessities of life — goods that *must* be consumed in order for consumers to survive and continue being consumers — to show distinct market failure in circumstances when the supply is reduced below sustenance level. Hoarders and black markets can have that effect on supply — the very chaos that makes rationing an apparent humanitarian imperative also creates supply-chain conditions that are very difficult to police, and claims on the attention of even the most incorruptible ration administrators that are very hard to verify, but also very hard to refuse.

      So it may well be that the most humanitarian thing one can do for a region with radically disrupted logistics for the necessities is to maintain a high supply of goods available at region’s borders, and let trade do its thing. And you don’t have to be a market fundamentalist to think so — I’m not, Schumacher sure wasn’t, and Adenauer was quite happy to take credit later for a marketization decision he acceded to only grudgingly, at best.

      • Michael Turner

        “George Debreu”? I meant Gérard. What’s happening with my brain these days? It’s not good. :-(

  • Brian Hewes

    I know Samuel Bowles and Herbert Gintis write about this in their “Cooperative Homo economics” paper. If I understand their paper correctly they think the current model of the self-regarding individuals is inadequate and that “human cooperation must be part of an adequate explanation”. It seems they are interested in explaining the system not so much in changing it.

    Robin are you saying that we should ignore the evolutionary trait of cooperation and accept “that we will each act selfishly, cooperating selfishly via exchange and contract, we can more easily rethink and relearn what actions are actually helpful in our new changing world”? That seems like a mighty tall order.

    • http://hanson.gmu.edu Robin Hanson

      I said I was questioning the claim “that we would do better to [more] empower and emphasize these instincts.”

      • Brian Hewes

        Ok that makes more sense, maybe I should read more carefully next time.
        I worked and lived in New Orleans after Katrina and many times I saw the negative effects of empowering and emphasizing the cooperation instinct. More often then not caused nothing to get repaired or sold.

  • Doug S.

    The person who pays the highest price is not necessarily the person with the most valued use.

    I’m going to quote one of my favorite bloggers, Alonzo Fyfe, on this.

    From his blog post Economics and Morality:

    Now, I would like to turn to another example – an example that I used when discussing “price gouging” in the wake of hurricane Katrina. It concerns a case in which an agent has a bottle of water to sell and two customers to sell it to. One is a wealthy person who wants the water so she can shampoo her poodle. The other is a poor person with twenty dollars who wants the water to give to her sick child.

    In order to get the bottle of water to shampoo her dog, the wealthy person offers $20.01, and she gets the water.

    According to standard economic claims, the water went to its most highly valued use.

    Only, this is not true. If the second person had as much money as the first, we may assume that she would have bid the price far above twenty dollars. In fact, it is not unreasonable to assume that she would have clearly outbid the woman who wants to shampoo her dog. The second woman values the welfare of her child. However, she is powerless to put express that value in the marketplace.

    This is an extreme situation. However, it is a type of situation that works its way into the supply and demand curves of economics. One of the reasons that demand tends to go down as price goes up is because lower-income purchasers have to drop out. When they do, it is a mistake to say that the goods actually went to those who valued them the most. In many cases, those with money get to bid resources away from those who value them more, because the rich can afford to take them instead and use them for trivial purposes.

    The rich do not have to bid the price above what the poor are capable of paying for this to be a problem. The fact is, a gallon of gas at $3.50 costs a poor person more (in terms of the value of alternatives for that $3.50) than a gallon of gas to a rich person at that same price. As the price of gasoline goes up, rich people can continue to buy their SUV’s and use them for trivial purposes. As they do so, they keep gasoline scarce, and they keep the price up. This forces the poor people to do without gasoline – even when the poor person would have valued the use of $3.50 in gasoline more than the rich person.

    [RH - I cut off the orignally 765 word comment here, to keep within the 500 word comment limit.]

  • http://bizop.ca michael webster

    Your point about scarcity in time of need could probably be better made if you used the distinction between coordination and cooperation.

    Traffic signs allow us to coordinate our movement. We are not cooperating because I don’t need to know what you want, only that you will follow conventions.

    While if I let you merge into my traffic lane, I am cooperating with you because I have to know what you want.

    You could write something like this:

    “In times of scarcity, then, needed goods can be distributed through a number of coordination mechanisms, including a market price without a ceiling. For many of us, however, this strikes us as minimally uncooperative behavior given the group’s needs.”

  • Sebastian Franck

    Isn’t this part of the reason why firms exist? View the firm as a cooperative community in which greater efficiencies arise than through pure market transactions. Probably (but maybe not solely) through greatly reduced transaction costs (viz. Coase).

    There is plenty of cooperation in market economies. Mainly within firms. De Waal’s observations may be relevant more for management issues than political issues.

    • Michael Turner

      I think there’s been some more recent work on the theory of the firm. Reduction of transaction costs was never, by itself, a completely satisfactory story.

      The more recent idea is that firms do better than pure free-agency in part because they reduce the natural human tendency to procrastinate. This makes sense to me.

      Rules like “You have to be here at 8:30am” and “You must work 8 hours, with only the breaks scheduled”, etc. are kind of dumb, when you think about it. But people are kind of dumb. They are more motivated to cooperate when they could lose big-time — i.e., lose their regular income because they get fired for showing up late and/or leaving early — than they are by winning big-time — i.e., by being independent, working more, working harder, marketing themselves better, etc.

      Working for bosses is dreary and often demeaning, but most people choose that existence over a more entrepreneurial one. Why? Maybe because it’s easier to cooperate under a system of simple (if rather arbitrary) rules. After all which is harder? Being steadily milked for your labor value as you chew the social cud from the pastures of fellow-employee acquaintances that the corporation conveniently supplies in the cubicles around you and across from you on the assembly line? Or busting a move on your own?

      This might be considered another “dark side of cooperation”, in a way, though it’s closely related to the general idea Hanson has here, of cooperation fostering somewhat-mindless conformity. I’m not sure there’s much improvement to be had. I sometimes imagine a more futarchical corporate existence, one that would be a less drearily conformist, insofar as it offered ways to legitimate deviant employee opinion by objectively recognizing as winners — through monetary reward — many of those who can observe more dispassionately, think more clearly, and be right more often. If that’s possible, maybe it wouldn’t be a dramatic improvement. But it would still be something.

      • Sebastian Franck

        Hmm, to me rules are not the same as cooperation. In fact, the workplaces you describe here are the exact opposite of cooperative environments, IMO. Rules are needed only where people have no other incentives to cooperate.

        I’m convinced there are other returns to “being in it together”, e.g. social identiy, a sense of belonging and being needed, but also efficiencies that pertain to the entire organization, e.g. better quality controls through higher levels of trust, more efficient communication (transaction cost reduction), implicit knowledge embedded in the culture of the corporation. These are significant.

        I believe we are moving towards a lot more hybrid forms where the efficiencies of the firm co-exist with the freedoms, bonuses, risks and responsibilities of free agency.

  • Will

    I think saying “dark side of cooperation” is a bit misleading here – instead, I would propose the “dark side of social instincts.” As you point out, we can have cooperative behavior with selfish motives through market processes, but this form of cooperation does not have the dark side you allude to. On the other hand, harming out-groups is not bad because of the cooperation the in-group uses to do so. You should be speaking out against the true culprit.

  • magfrump

    Taking a second to look at what value the conformity-induction has:

    -People have different ideas about what is right.
    -In the ancestral environment, if more people do something it’s better.
    -Getting people to conform is in their interest, unless they have a better idea.
    -If they have a better idea, other people want it.
    -Using conformity punishment will either quash this idea (presumably if it is bad) or spread it very rapidly (if it is good).

    The result here in my mind, is that what makes cooperation and conformity-destruction bad, is that people are not negotiating their ideas in good faith. Before punishing someone for nonconformity (in the carbon tax case, by making economic sanctions) have a mass discussion about it.

    The discussion about carbon emissions, whether or not people like Robin continue to discuss it (which is good!) has been resolved. On the other hand, nations and politicians are very poorly positioned to negotiate in good faith, which results in inefficiencies based on our institutions rather than our instincts.

  • Pingback: The Institute

  • Pingback: Price gouging and the “dark side of cooperation” « Knowledge Problem

  • http://bizop.ca michael webster
  • http://brokensymmetry.typepad.com Michael F. Martin

    If these goods are not allocated by price, they will instead be allocated by standing in lines, personal connections, etc., processes that are consistently worse at giving goods to those who value them the most, and do worse at creating incentives to prepare for such scenarios.

    This treats as discrete what is a continuous phenomenon. If allocations by price are made using a common currency, then the reciprocal ties between party and counter-party are minimized. But as we move to less and less common currencies — from bartering to personal connections, or (finally) familial obligations — the reciprocal ties are enhanced. Allocation of fungible resources may be most efficient through a common currency. But allocation of scarcer resources, or resources more subject to idiosyncratic valuations, may be more efficient through a currency that incorporates more interconnections between party and counterparty.

  • Pingback: Weekend Links - Maggie's Farm

  • Pingback: 20090920 - Adam Crowe

  • Pingback: Linkpile