247 | Samuel Bowles on Economics, Cooperation, and Inequality
What this covers
Sean Carroll interviews Samuel Bowles, an economist at the Santa Fe Institute, about the foundations of economic theory and the evidence for human cooperation. Bowles argues that mainstream economics rests on faulty axioms about human nature and the structure of economic life. The conversation ranges across behavioral experiments, evolutionary biology, and the history of economic thought, building toward a case that power, social norms, and cooperation must become central to economics rather than peripheral concerns. The starting point is empirical: incomplete contracts — the fact that labor and credit agreements cannot specify every contingency — combined with evidence that humans are neither purely selfish nor rational, which together invalidate the standard supply-and-demand model and the invisible hand.
Bowles traces how Hayek's insight about local and limited information explains why central planning fails, but the same logic generates a very different conclusion when applied to contract theory: markets need not clear, unemployment can persist involuntarily, and employers hold genuine power over workers. He draws on game theory and cross-cultural behavioral experiments — conducted among herders, farmers, and truck drivers, not just university students — showing that altruism, reciprocity, and altruistic punishment are widespread, contradicting the portrait of self-interested homo economicus. On evolutionary grounds, he argues that group selection combined with gene-culture coevolution can explain how humans evolved a capacity for genetic altruism. The discussion moves into policy implications: why the 2008 crisis resulted from misapplying market models, how crowding out of intrinsic motivation undermines incentive schemes, and why community as third governance arena — neither state nor market — becomes crucial for modern work that cannot be easily contracted or measured.
Bowles argues that economics is being fundamentally rewritten because contracts are inherently incomplete and humans are neither perfectly selfish nor perfectly rational, which means power, social norms, and cooperation must become central to economic theory and policy.
- The incompleteness of labor and credit contracts (rooted in Hayek's insight about local, limited information) breaks the supply-and-demand and invisible-hand models
- Behavioral experiments worldwide show genuine altruism, reciprocity, and fairness are widespread, so homo economicus is empirically false
- Group selection plus cultural transmission can explain how an altruistic species evolved, giving cooperation a scientific foundation
Standard supply-and-demand and deregulated-market models fail for labor, financial markets, and unmeasurable modern work.
- Only about one in seven US workers is in manufacturing, mining, and 'smokestack' industries; the vast majority do work like care of the old and young, teaching, services, and cultural production that is hard to measure, which is why contracts work poorly for most modern labor.
“In the United States, one in seven works in manufacturing, mining, and things which have smokestacks... So most of the things we're doing are really hard to measure.”
- The 2008 financial crisis resulted from treating financial markets like an ordinary market for goods where the standard model works, then deregulating them in a way almost certain to produce speculative bubbles and instability.
“treating it like it was an ordinary market for shirts... and then de-regulating it in such a way that you're almost certain to have a speculative bubbles and instability.”
- The supply-and-demand model works for goods like shirts but fails for labor because there is persistently a substantial fraction of people unable to find work, a failure dramatized by the Great Depression and recognized by Keynes and Samuelson, who insisted labor markets need a different theory.
“it might work for shirts but it certainly doesn't work for labor, that is, for people searching to find work and people searching to find employees, because we know we typically have a substantial fraction of the population that is unable to find work.”
Incomplete-contract theory built on Hayek's insight reveals persistent employer power and unemployment, requiring economics to recognize power and social norms.
- Hayek's central idea that information is local and limited correctly explains why central planning fails, but the same insight is the foundation of incomplete-contract theory, which shows supply need not equal demand and the invisible hand can fail — so information economics built on Hayek's premise produces conclusions opposite to the neoliberal school he inspired.
“Hayek's ideas about information are also the key idea behind why contracts are incomplete. The reason why the contract between you and me is incomplete is because you can't know what I'm doing all the time.”
- Because an employer can purchase a worker's time but cannot enforce a worker's effort (an incomplete contract), employers must pay more than the worker could get elsewhere to make the threat of firing meaningful; this implies involuntary unemployment must exist, gives employers genuine power over employees, and forces economics to take social norms and power seriously.
“My agreement to show up at your place, that's also enforceable. My agreement to work is not enforceable.”
- Any paradigm that replaces neoliberalism as a policy framework must, like neoliberalism, combine a good predictive model of the economy with a coherent set of moral principles — in particular one that can recognize and interrogate the private exercise of power that incomplete-contract economics shows is unavoidable.
“anything which is going to replace the neoliberal paradigm as a policy paradigm, has to do something like the same thing. That is, it has to have a model of how the economy works, that is a good model... but at the same time, is one which recognizes the importance of things that people who are opposed to the neoliberal prescriptions would think.”
Humans across cultures are predominantly altruistic, reciprocal, and fair-minded, punishing free riders and cooperating when local information and moral judgment guide behavior.
- Punishment and pressure to cooperate work better when applied by peers in the same situation as an apparently altruistic act than when imposed from the top; for example, mandating COVID vaccines provoked hostility, whereas neighbors expressing personal concern that someone get vaccinated was more effective.
“mandating vaccines incurred a very hostile response, whereas your neighbors saying, 'You know, I really feel a lot better if you go and get vaccinated,' was more effective.”
- The view of humans as the smart, selfish homo economicus that dominated economics 30-40 years ago is no longer believed even by economists, and biologists now study conditions under which a species might become genetically predisposed to care for others.
“we were still back in the dark old days of homo economicus, economic man, the one who was really smart and selfish... And I don't think anyone believes that anymore. Not even economists don't believe it.”
- In repeated public-goods experiments, contributions collapse toward zero, but when players can pay a cost to reduce free riders' payoffs, they punish heavily, contributions rise to around 80%, and punishment becomes unnecessary by the end — and people will pay to punish free riders even when the target won't learn of it until after the game, showing a deep predisposition to punish unfairness.
“in the second set of 10 rounds, the amount contributed which had gone down almost to zero, goes up much, much higher up to 80% or something. And by the end of the game, almost everyone is contributing. Nobody's punishing.”
- Beyond the state (regulations enforced by obedience or fear) and the market (buying and selling motivated by self-interest), there is a third neglected arena — community or civil society — where high local information and mutual moral judgment can control free riding and enable effective cooperation; this matters more today because most modern work (care, teaching, services, ideas) cannot be measured or contracted, so neither contracts nor government fiat suffices.
“imagine a line running from state over on the left, market over on the right... when I say there's something missing, I guess the expression I would use is community or civil society.”
- Across behavioral economics experiments conducted worldwide with real money payoffs — among hunter-gatherers, herders, farmers, and truck drivers, not just students — the consistently selfish person is a minority; most people are altruists, reciprocators, or fair-minded individuals who dislike even being unfairly advantaged.
“the entirely selfish, consistently selfish person, was really a minority. Very few people acted selfish in the way that economists had assumed.”
Human strategic interaction and intentionality make physical-science analogies inappropriate for economics.
- Physical-science analogies fail in economics because human actors have intentions and goals, and economic interactions are strategic — each party's outcome depends on what the other does and both parties recognize this mutual interdependence, which non-thinking particles or genes cannot capture.
“people, when they act, are trying to accomplish something. They've got an objective. Their intentionality really matters. And very often we find ourselves running up against the intentionality of somebody else.”
- Economic interactions are better understood through the metaphor of conversation or language than through physical metaphors of transportation — because each party anticipates the other's response and acts with intention in an endless regress, which is closer to how society actually works.
“some economists, myself included, have been more attracted to the idea that economic interactions are more like a conversation. So we should really be thinking about linguistics.”
Group selection and cultural evolution can sustain genetic altruism and larger solidarity through shared norms controlling free riding.
- Cultural transmission and natural selection share the same abstract replicator-dynamic math (strategies above average fitness grow), but cultural selection can override biological fitness — as when cities, which had higher death rates than the countryside until two centuries ago, kept attracting migrants, making them a cultural success but a biological disaster.
“cities were a biological disaster and a cultural success, and the death rates in cities were higher than in the countryside. And then people kept flooding to the cities, and the way we got cities was the cultural selection process was swamping the biological one.”
- Although within any group altruists are disadvantaged because they bear fitness costs, a group composed mostly of altruists outcompetes groups without them; if past human groups were genetically different enough and competed for resources (e.g., during environmental crises or conflict), group selection could overcome within-group erosion of altruism, making humans possibly uniquely capable of evolving genetic altruism.
“a group that was composed mostly of altruists would do very well as a group, for example, in competition with another group.”
- Hunter-gatherer societies that depended on sharing large, unpredictable game evolved norms to control free riders — such as eating in public so everyone sees who takes from the pot, and ridiculing, joking about, and eventually physically punishing those who take without contributing.
“You eat in public so everybody knows who's getting the stuff from the pot. And the other thing is... the free riders are not treated well. They're joked about, they're ridiculed. And eventually, if they insist, they're actually physically punished.”
- Nationalism, despite its harms, demonstrates that humans as cultural animals can overcome antiquated 'us versus them' divisions to construct a much larger 'us' — as Italy unified previously parochial regions and America forged solidarity among people of utterly different origins — which gives grounds for optimism that we could extend that sense of solidarity globally, though extending it to distant future generations is harder.
“that kind of 'us and them' sentiment is the basis of many wars and many horrors... but it also means that as a cultural animal, human beings are capable of overcoming the past differences, the old antiquated 'uses and thems' for a larger 'us.'”
- Group selection works as an individual-based evolutionary process: it requires only that members of a group be more genetically related to each other than to a random member of the population, so group selection is essentially kin selection extended to larger, more genetically distant groups.
“if they're different enough genetically, it means I am actually more closely related to the people in my group than I am to a randomly selected member of the population... this is kin selection but large.”
Economics should become modest about its predictive power and recognize its tools work only in some cases, not universally.
- Paying children for activities they already enjoy (like painting) causes them to stop doing it later, because the payment 'overjustifies' the behavior and undermines their intrinsic sense of self-determination — demonstrating that incentives can crowd out motivation.
“Even two weeks later, the ones who were paid for their pictures have stopped painting. They're not painting anymore.”
- Classical economics, exemplified by Adam Smith's invisible hand, was about designing ways of organizing society that would produce good outcomes even if everyone were entirely self-interested — Smith did not believe people were actually selfish but used selfishness as a worst-case design assumption.
“let's think of a way of organizing our society which would work even if everyone was basically a jerk.”
- Because realistic models of intentional, strategically interacting humans produce intractable equations with many equilibria that resist analytic intuition and require simulation, economics should become more modest — recognizing it has a box of tools that help in some cases and are powerless in others, rather than claiming a single tool for understanding society.
“We have a box of tools that help in some cases, and in some cases, really quite amazingly. And in other cases, are quite powerless to give us the kind of analysis that we really would need.”
- Adam Smith's case for self-interest and markets, when read with the very next sentence ('Nobody but a beggar would rely on the benevolence of others'), was not really about efficiency or the invisible hand but about markets as a way of organizing society consistent with equal human dignity, in contrast to feudal hierarchy.
“Nobody but a beggar would rely on the benevolence of others for their dinner. So Smith's argument about self-interest and the market... It had to do with the market as being a way of organizing society that's consistent with people being dignified and having equal dignity.”
- Darwin derived his idea of natural selection from the economist Thomas Malthus, explicitly applying Malthus's ideas to all species — an example of fruitful interplay where biology borrowed from economics.
“Darwin got his idea of natural selection from Thomas Malthus, the economist. And in fact, he said, 'I'm just applying Malthus' ideas to all species.'”