Berkeley Talks transcript: Economist Samuel Bowles on why good incentives are no substitute for good citizens

[Music: “Contrarian” by Blue Dot Sessions]

Anne Brice: This is Berkeley Talks, a Berkeley News podcast that features lectures and conversations at UC Berkeley. You can find more talks at And you can subscribe on Apple Podcasts or wherever you listen.

Speaker 1: This podcast is a presentation to, University of California Television. Like what you learn? Help others discover UCTV podcasts by leaving a comment or rating in iTunes.

Rui De Figueiredo: Good afternoon everyone, and thanks, Ellen, for great organization. My name is Rui De Figueiredo. I’m an emeritus professor at the Haas School of Business and the department of political science here at Berkeley. I’m also the chair of the Weinestock Lecture Committee. Along with the graduate division and the graduate council of the academic senate, I’m very pleased to welcome today’s speaker, professor Samuel Bowles, this year’s speaker in the Barbara Weinestock Lectures on the morals of trade. I’ll introduce Dr. Bowles in a minute, but first, let me take a moment to tell you how the endowment supporting the lectures came to the University of California, Berkeley.

In 1902, Harris Weinestock, a well-known business man based in Sacramento provided the university with a fund in honor of his wife Barbara, whom the lectures are named after, to support an annual public lecture on the morals of trade. Weinestock explained his motivations in an article written after the first lecture was delivered in 1904. Thus, he said, “Hope is in the air, and there’s a better and clearer day in store for all destined to spend their lives in commercial pursuits.” The thing to do at this hour is to accelerate the movement and to bring this hope for a day as near to our own as possible. The California University lectureship on the morals of trade is a small effort in that direction. Others who have delivered the lecture include consumer advocate Ralph Nadar, member of the British Parliament Neil Kinnock, Nobel Laureate Amartya Sen, former US Secretary of Labor Robert Reich, former Secretary of the U.S. Department of Health and Human Services Kathleen Sebelius and most recently, nutritionist and author Marion Nestle, and then last year Governor Jennifer Granholm who was a two-term governor of the State of Michigan.

Now a few words about our distinguished speaker today. A very noted economist, professor Samuel Bowles is the research professor and director of the behavioral sciences program at the Santa Fe Institute in Santa Fe, New Mexico. He is published prolifically in many journals including Science, Nature, the New Scientist, American Economic Review, the Journal of Theoretical Biology, Antiquity, the Harvard Business Review and the Journal of Political Economy, amongst others, a very wide spread of disciplines. He’s also published a number of books including A Cooperative Species: Human Reciprocity and its Evolution (2013), Microeconomics: Behavior Institutions and Evolution (2006), and his most recent book, The Moral Economy: Why Good Laws are No Substitute for Good Citizens, which was published in 2017. He’s currently working on another manuscript called Equality’s Moment: The Origins and Future of Economic Disparity and Political Hierarchy.

With a global team of researchers and teachers, he’s also developed a new introduction to economics that demonstrates the power of modern economics to eliminate problems such as growing inequality, climate change, innovation, wealth creation and instability. A year-long highly interactive e-text called The Economy is open access and freely available to anyone in the world with internet connectivity on their phone or other device at Professor Bowles received his bachelor’s degree from Yale University, his PhD in economics from Harvard University. He’s advised South African President Nelson Mandela, Dr. Martin Luther King Jr., U.S. presidential candidates including Robert F. Kennedy and Jessie Jackson, and in 2006 he was awarded the Leontief Prize for his outstanding contributions to economic theory. Today he’ll be speaking about the moral economy, why good incentives are no substitute for good citizens. With that, please welcome professor Samuel Bowles. He will give a lecture, and then we will have a discussion.

Samuel Bowles: I got to say morals and trade and a moral economy sound like an oxymoron, doesn’t it? There’s something a little jarring about somebody who gives a grant to study morals and trade, at least in economics, where we try to separate these issues as much as possible, but I’ll try to convince you that’s a mistake. The picture in front of you is a fresco in the town hall in Siena where I often teach. It’s a representation in the 14th century of good government. Notice it’s a secular image. If you look carefully, you can see it’s an image about commerce and peaceful intercourse among people. The constitution of the Town of Siena at the same time made it quite clear and said, “The city must be governed by men who love peace and justice.” The marriage of morality and good governance was the basic idea of that period, an idea which came to be displaced particularly in economics in the years later, a story which I’ll tell and then try to reverse before we’re finished.
Samuel Bowles: It’s a long story, so let me give you the basic idea first, just a few summary points. This is what I would like to persuade you is worth considering and is perhaps true. First, is something which I think you’re aware of. It’s widely held that in designing public policies and legal systems and so on, we should assume, at least on grounds of prudence, that citizens are entirely self-interested and amoral. That’s the so-called homo economicus, this assumption which is the foundation of the public policy sciences and economics, what’s called mechanism design and so on. Of course, it affects legal practices as well, but, and this is the main point here, it’s anything but prudent to assume that homo economicus is a good representation of our citizens and the people who make up the body politic, which we’re governing. There are two reasons for this.

The first is, and I’ll try to convince you this is really true, policies that follow from that paradigm are more likely to make its assumptions true, that is are more likely to convert people to acting more like homo economicus than they would in the presence of other policies, so the policies are self-defeating and backfire. The reason why this takes place is not per se is something wrong with incentive. This is a more subtle point. I’m not going to argue here that incentives are the problem. I’m going to argue that incentives are used for purposes that people reject because they’re either attempting to implement an unfair outcome or attempting to implement a type of domination of the incentive giver over the target of the incentives. We should be thinking not about incentives as intrinsically a problem, but rather the social relationships that are behind the incentives as the problem. Again, I’ll try to give you some reasons for believing that.

Finally, in economics we have a field called mechanism design in which we think of very clever ways of defining new property rights, new incentives, new constraints, and so on whereby we can implement good outcomes no matter what, no matter what the people are like. I’ll return to this theme later, but the conclusion that I think I will convince you of is no matter how clever we are in designing incentives, and property rights, and so on, that alone will never be sufficient for good governance in facing the kinds of problems that humanity now faces such as climate change, mounting inequality, economic instability, political violence, and so on. That’s what I’m going to talk about.

The ideas, many of them here, are not new. It begins in the social sciences with Richard Titmuss’ book called The Gift Relationship in which he said the commercialization of blood and donor relationships represses the expression of altruism and erodes a sense of community. The idea was that public spirited motives are crowded out by the explicit application of incentives and therefore, we should use these incentives less because they have that crowding out feature. Now at the time that he wrote, early 70s, there was a lot of research being done. A lot of it happened to be in nearby Stanford University by Nisbett, Greene, Lepper, and others, in particular who I’ll come back to. In psychology, it’s called the overjustification literature. What it shows, for example, is that you can crowd out children’s desire to paint by paying them for their paintings, things of that nature. It received a warm welcome in psychology, but a rather chilly one in economics.

It was surprising. Arrow, and also Solow, reviewed the book. I suppose if you’re sociologist, that’s the most you can hope for, that an economist would actually read your book and review it. Two Nobel Prize winners is not bad, but what they said is they didn’t believe that there was evidence for this crowding out phenomena. Here’s Arrow: “Why should it be that the creation of a market for blood would decrease the altruism embodied in giving blood? I did not find any clear answer in Titmuss. I see no real evidence that the presence of a commercial blood supply decreases the amount of altruism.” I had the same reaction. I read the book with great interest, and I didn’t think that he made his case.

Now, of course, since then a lot has changed. The skepticism announced by these two great economists Arrow and Solow was on the basis of the fact that at the time there was very little evidence that the kind of moral incentives that Titmuss was talking about were important. There was even less evidence that they would be crowded out by the use of incentives. Arrow and Solow were on good grounds, and Titmuss was essentially making a good argument that turned out to be probably correct based on later evidence, in fact, an unpublished dissertation the same year that Titmuss wrote made a pretty good case for the fact that Titmuss had been right, but Titmuss had been unaware of this dissertation.

Now since then in economics, a lot has changed. Here I’m going to refer to things going on within economics. Bear with me because I’ll come back to them in the course of the talk. If you’re not familiar with the themes, they’ll come around again. The first is obvious. You all know about this. There’s now a lot of experimental evidence in economics that these moral incentives or ethical principles regarding preferences, generically called social preferences, these are important as a basis for behavior. It doesn’t mean they’re important instead of self-interest. Self-interest is also important, but there’s no question on the basis of the experimental evidence that our introspection and our natural observation around us, which says to us, yes, other motives matter, that is strongly confirmed in the experimental evidence.

Equally important, but a little less obvious, economics has now taken up the idea as the norm that when I contract with you for something, the contract is probably incomplete. If I hire you, for example, or you hire me, the contract says I’m supposed to show up on time and obey the supervisor. It doesn’t say what work I’m supposed to do. Now why is that important? How does that get norms into play? It gets the norms into play directly because if I hire you, I care about your work ethic. If I loan money to you, I care about whether you’re telling me the truth about what you’re going to do with the money because the contract of a lender to a borrower is also incomplete. Because most contracts are incomplete including very important ones for credit and labor, for example, it now becomes obvious even to economists who have resisted interdisciplinarity. It becomes obvious that the exercise of power and their salience of social norms is a part of very conventional economic problems like what’s the wage? What’s the interest rate?

Now finally, perhaps even more esoteric to those of you who’ve never heard of the Lucas critique. The Lucas critique is a very important intervention in economics. It said, “When you design an economic policy, you have to take account of the effect that that policy will have on the beliefs that individuals in the society have about what’s going to happen next and how the government will behave.” In other words, when you design a policy don’t take the beliefs of people as exogenous. They’ll be affected by the policy. All I’m doing here is I’m taking the Lucas critique and instead of applying it to beliefs, I’m applying it to norms and values and saying, when you implement a policy don’t think that the preferences which existed before you had the intervention are going to be around later. I’ll come back to that.
Samuel Bowles: Now in response to these three things, some economists have rediscovered Aristotle, at least that’s what they said. I suspect that some of them just discovered Aristotle, but if they did, probably the thing they would like to read most is the following. This is from The Ethics: “Law givers make the citizen good by inculcating habits in them, and this is the aim of every law giver. If he does not succeed in doing that, his legislation is a failure. It is in this that a good constitution differs from a bad one.” Now notice, it’s not this is the characteristic of a good constitution. This is the characteristic of a good constitution. This is the sine qua non of a good constitution, teaching the public to be good citizens.

Now this gives us a dilemma then. If Titmuss is right that incentives compromise the kind of values that we would hope to have in the citizenry, then what is Aristotle’s sophisticated legislator supposed to do because by now Aristotle’s legislator knows that incentives are essential. You’re not going to run a society without incentives. You’re not going to abolish markets. We set up a little tension here between Titmuss’ insight and Aristotle’s view of politics. Now there’s one way out of this. Maybe we don’t have to worry about this.

Now economists are really of the view that we can do without Aristotle’s values. That’s the way we do our public economics. I started to trace where this idea came from, and it really comes not from the meaning of an important body of work, but rather by a supposition that has been adopted by great thinkers as a thought exercise. Think about this. This is Machiavelli. I’ll come back to the rest of the quote in just a minute, but the important part in red is, “All men are wicked. Hunger and poverty make them industrious, and laws make them good.” Does that sound familiar? That’s the idea that we can organize a constitution on the basis of the incentives, based on hunger, in this case, to induce people to do things, which will add up to the social good.

Now it’s not that, of course, Machiavelli was not saying that hunger makes people good. He says, “Hunger makes people act as if they were good,” and that’s good enough. Now there are many others who had this idea, a contemporary Machiavellian, Marsilio, but also if you think about the way Hobbs posed the problem of order and limited freedom in what was then modern society, Mandeville, Hume, Smith, Bentham. Bernard Mandeville, famous in economics for a wonderful book that he wrote, The Fable of the Bees, this is what he had to say about the relationship between vice and governance: “Thus every part was full of vice, yet the whole mass a paradise.” He’s describing a hive of bees were nasty animals: “Such were the blessings of that state. Their crimes conspired to make them great. The worst of all the multitude did something for the common good.” This is a really big idea, that we should distinguish between motives and consequences. This is Mandeville, and it’s all of economics after that.

He wrote it about bees because it would have been too scandalous at the beginning of the 18th century had he written it about humans. In fact, he got into a lot of trouble as it was. What happened in the hive was that the hive was a scandalous hive, and everything worked out pretty well, but then they got religion, and they all acted like good little bees, and the hive collapsed, and so on. That was the story of the bees. Now the subtitle of one of the editions of his book was Human Frailties May Be Turned to the Advantage of Civil Society and Made to Supply the Place of Moral Virtue. Now that is a very, very interesting idea, that is the ordinary motives could substitute for virtues. He didn’t explain how that could be done, but it would very soon be provided by the classical economist. The invisible hand is, of course, just a statement of that view, that ordinary motives when organized properly by the proper social institutions can induce people to act in the public good.

Here we have one of the most famous quotes even rivaling one where he mentions the invisible hand: “It’s not from the benevolence of the butcher, the brewer or the baker that we expect our dinner, but from their regard to their self-interest.” Now this view was taken up by people at the time. Hume, of course, preceded Adam Smith by about half a generation. What Hume said about policy making and constitution making really epitomizes the way modern economists have thought. Political writers have established it as a maxim in that contriving any system of government every man ought to be supposed to be a knave and to have no other end in all of his actions than his private interest. By this interest, we must govern him, and by means of it, make him not withstanding his insatiable avarice and ambition cooperate to the public good.

Now by the way, doesn’t it sound like he must have been reading Machiavelli? Even the structure of the sentence. anybody who wants to form a republic should suppose that. That’s the translation of Machiavelli’s Italian. Hume is saying the same thing, and we get Bentham who wrote the first book in what we now call public economics, how to design public policies. He announced what he called his duty on interest principle. He was talking about compensation of public officials, and he said, “We should make it every man’s interest to observe that conduct which it is duty to observe.” Now in economics we just call that aligning incentives or an incentive compatible policy, that is it’s a policy that induces people as a result of their self-interest to do the thing that the society as a whole would deem profitable or good for the society. Once again, these two statements here are about not making people care about their morality or the benefits they give to others, but rather giving in private incentives to do by the clever design of public policy.

What happened during this period from Machiavelli through to Bentham was that good laws replaced good citizens as the sine qua non of good government. We could displace Aristotle’s concern by the design of good principles, and that was, by the way, a very key idea in Machiavelli’s Discourses, not The Prince, but in The Discourses, that’s what he’s really saying. The thing that we have that we can do is we can design rules, which will induce people, as he put it, of ordinary humors. We could induce people of ordinary humors to conduct themselves in a way that contributes to the public good. This is really a big idea right up to the present.

I think it’s familiar to you, but this is a wonderful editorial from 1988 from The New York Times of previous financial crisis. Program trading, index arbitrage, and so on. I could go on about this, but the key line is this: “Perhaps the most important idea here is that to distinguish between motive and consequence. Derivative securities attract the greedy the way raw meat attracts piranhas, but so what? Private greed can lead to public good. The sensible goal for securities regulation is to channel selfish behavior, not to thwart it.” I’m imagining that that was written by some, that line was put in there by some Yale graduate from economics who was assisting the editorial board, had read Mandeville, and so on, and thought that was a clever thing to put in. It’s just straight Mandeville.

Now think about it. It may be repugnant to you to think that way, but should we worry, really? Why not just harness greed? Economists say that a market economy or a capitalist economy has the benefit of not relying on morality to function well. Of course, a very famous expositor of this view is Hayek. The liberal market economy is a system under which bad men can do least harm. It does not require our finding good men for running it or that on all men becoming better than they are now. That’s just Machiavelli. Ordinary humerus, that’s all we need. A little more explicitly, Charles Schultze of the Brookings Institution: “Market like arrangements reduce the need for compassion, patriotism, brotherly love and cultural solidarity.”

Finally, I think the most astute one is Buchanan, Nobel Laureate in economics. He’s shopping near his home in Virginia: “I do not know the fruit salesman personally. I have no particular interest in his well-being. He reciprocates this attitude. I do not know and have no need to know whether he is in the direst of poverty, extremely wealthy or somewhere in between, yet the two of us are able to transact exchanges efficiently because both parties agree on the property rights relevant to them.” Now that is a very bright statement. I don’t endorse his indifference towards the fruit salesman, but he understands what it is about the economics, which I’m happy to say we can now call former conventional economics, which is no longer taught in grad school, thank you very much because we agree on the property rights relevant to them. That’s just another statement for saying, “The contract is complete.” Everything that we care about has been contracted for, so I don’t care about his morality. If he gives me a bag of bananas, and it’s not a bag of bananas, I get my money back. That’s what he’s saying. Complete contract is what makes his indifference an appropriate stance.

Economists have had, you may wonder why I said it was x. Did you wonder? It is x, I’m coming to that, why it is that we once thought that, but we no longer think it. Why did economists think that we shouldn’t be concerned with preferences, we could put them aside? We once had a theory that showed that desirable social outcomes are possible with unrestricted preferences. Let me run through this argument. Unrestricted preferences mean, essentially, bring them on. We don’t care what the preferences are, we can handle them. Here I’ll lay out in a single slide what the argument is, and then I’ll tell you why it’s wrong and why economists have now rejected it.

First, the market is obviously decentralized. It’s privacy preserving. When guided possibly by a good, but not omniscient planner or social designer, policy maker, it’s capable of implementing outcomes by means of procedures. The italicized outcomes are the three things that matter as far as a liberal commitment like that of, for example, Hayek, that is it avoids coercion. It avoids paternalism, and it’s free of waste, those three things. I’ll come back to them. It avoids waste. What we mean in economics by that is just the outcome is Pareto efficient. It’s not possible in the outcome for all parties to be made better off by some change without anyone being made worse off. That’s the rather minimalist view of efficiency, which is part of this liberal commitment.

It avoids paternalism. It doesn’t say that people have to be of one type of preference or the other. It completely rejects Aristotle’s view that the constitution should take a position on what the citizens should be like. By unrestricted preferences, it means that you can have any preferences you want. This is now called preference neutrality among liberal philosophers. By the way, it’s a very new term, it’s not an ancient term. It just means if you’re a liberal, you don’t take a position about the nature of the good or what people consider to be the good life, and so on. Finally, whatever the outcome is, it should avoid coercion. If you participate in an exchange, it has to be voluntary.

Now think about it. That’s not all you could ask of a philosophy as the basis for economics, but you must agree that’s pretty basic. Those are the things that liberal people, I count myself among them, liberal people are committed to. Now some people think that it’s been done. A great professor from this great university certainly claimed that it had been done. The superiority of liberalism has been mathematically demonstrated by Gerard Debreu. This is in Figueroa. Actually, Figueroa misquoted him in the headline. What he said, actually, was virtually identical. What he said is, “The superiority of the liberal economy is incontestable and can be demonstrated mathematically.” He was talking about the famous theorem, so-called invisible hand theorem that he and kind of Arrow proved 30 years earlier, which showed the conditions under which a competitive economy would support a Pareto efficient outcome.

Let’s think of this liberal trinity. Has it been demonstrated? Those are the three desiderata that I mentioned before, Pareto efficiency, preference neutrality, voluntary participation. This is what economists thought as of the late 80s, or the early 80s, that we had actually shown. Let me say why we thought that, that is we had some theories under the heading of mechanism design, which appeared to show that we could implement Pareto efficient outcomes with unrestricted preferences and with voluntary participation, but that hope turned out to be false. Four decades of mechanism design and a good number of Nobel Prizes for these results showed, and I’m summarizing here what the Swedish Nobel Committee said in awarding the prizes the first time, the really big mechanism designers got prizes, the implementation of Pareto efficient outcomes in the absence of complete contracts is generally impossible for unrestricted preferences. If participation is voluntary, and there’s a limit on the extent of penalties that may be opposed excluding physical punishment, imprisonment for debt, and so on. That’s the result of this body of work.

Now there are many parts of it. Getting people to truthfully tell what their preferences are or their capacity turns out to be not incentive compatible, that is you can’t design a way that does that if there are more than two options. That was shown in ’75. Private voluntary exchange even when you have complete contracts does not implement efficient outcomes. That was shown my Myerson, and Satterthwaite, and Chatterjee, and Samuelson also in the late 80s. Now as a result, the mechanism design literature has for the most part abandoned the Pareto criterion entirely. What it does, it talks about something called incentive efficiency, which means the best you can do given the preferences, which obviously means that then preferences are important. They’ve abandoned the idea, the strong idea of Pareto efficiency. These are the guys here, which Myerson and Maskin won the prize for their contribution to this.

Now a number of people have worked on this problem, and I want to say that this is a trilemma, not a trinity. The idea is you can have any two vertices, but not all three. I’ll show some results on the edges about why that’s true and what you lose if you do. For example, you can have Pareto efficiency and voluntary participation, but you can’t have preference neutrality. If you have that one and that one, that’s Wang there. What Wang and I showed is you can have these two, but only by violating preference neutrality. By the way, not true that both of the agents have to be honest or altruistic, just one, but they have to be sufficiently so. You have to violate the idea that bring them on when it comes to preferences. You could have voluntary participation and preference neutrality, but this is what Chatterjee and Samuelson showed. Then there will be unrealized gains from trade. You can have Pareto efficiency and preference neutrality, but only by a very clever mechanism, which requires people to commit to making a trade before they find out what the deal is in which some of them, once they find out what the deal is they won’t want to be part of it, so that violates voluntary participation.

I conclude that that attempt failed, and it failed a long time ago. Interestingly, the paper from Figueroa in which Debreu was quoted was exactly the time that these papers were coming out. He, Debreu, was extremely on top of that literature to his wonderful credit. Why he said that to Figueroa, I can’t imagine because the house of cards was collapsing right in those years. The date on Figueroa was ’84, and these papers were all in the early 80s. Virtue, to use the old style expression, it is still essential. This is especially the case where relevant information is not verifiable. The nonverifiability of information is what gives rise to incomplete contracts. If there’s some kind of information that either is lacking, so-called asymmetric information, or even if present can’t be used in a court of law to enforce a contract, then you will have an incomplete contract, and then all of these problems will arise and norms and other things will become necessary.

Now it’s also the case that imagine a state that had the capability of implementing a system whereby contracts would be complete. That would be an extremely invasive state capable of collecting information from you, which is now it cannot be collected. In other words, trying to make all of the information about your life verifiable in a court is exactly what most liberal people on privacy grounds would resist. Making the liberal economy, that is the neoclassical model, which was supposed to be the economic basis of the liberal philosophy, making it work is a dystopic vision because it would require contracts to be complete, which would require a state that no liberal would ever accept.

Now other regarding preferences, ethical commitments become important. Here was have Arrow again, once again getting it right, 1971 he wrote this: “The process of exchanges requires or at least is greatly facilitated by the presence of several virtues, not only truth but also trust, loyalty and justice in future dealings.” Again, still Arrow: “In the absence of trust, opportunities for mutually beneficial cooperation would have to be foregone. Norms of social behavior including ethical and moral codes may be reactions of society to compensate for market failures.” Once again, he’s got it right except suggesting that we had market failures before we had morals is a little bit ahistorical because clearly, human beings long had morals long before they had markets, and therefore, they couldn’t have had market failures, but I’ll set it aside for the great and sadly departed economist Kenneth Arrow.

Now let’s think about where economics went wrong. The classical writers assumed something that nobody has really put their finger on. They did not assume that people were selfish, far from it. They didn’t assume that preferences were somehow outside of economics or exogenous. Think about it. Hume’s moral conventions. Hume who said we should act as if people were knaves, the next sentences said, “It may seem odd to you, reader, that something which is a true axiom for politics should be false, in fact.” That’s what he said. We’re going to assume this thing, we know it’s not true. Think about Adam Smith, Theory of Moral Sentiments, very aware of that.

Let’s go back to Machiavelli: “It’s necessary to anyone who would organize a republic and its laws in the republic to suppose that people are wicked.” He didn’t say, “People are wicked.” He said, “You should adopt this assumption if you’re going to design a constitution.” As Hume put it, we want a knave-proof constitution. We want a wicked-proof constitution. In fact, Machiavelli went on and on about how society, if we’re well-governed, where they have high levels of morals. People understand what Machiavelli was all about. Read the Discourses, not The Prince. They were very aware of the importance of norms, but they did make the following assumption.

They assumed what Smith called moral sentiments and material interests could be pursued independently, that is they assumed that policies which were designed to shape behavior by appealing to the material interests, monetary incentives, for example, wouldn’t interfere with the moral sentiments, which might also assist in governing a society and similarly, that appeals to the moral incentives would not interfere with the exercise of the material interests. The key idea here is that incentives affect actions only by altering the costs and benefits of the targeted activities. There’s no affect on the preferences.

Now these propositions, as you will recognize, now come into play in policies to get kids to stay in school, to not get pregnant, sentencing guidelines, not to mention more standard aspects of economics. Let me lay out what I just said somewhat more analytically. We have here something called values, which is here, sorry, social preferences. Suppose there’s an incentive maker, a designer, social planner, you call them, and she wants to affect the action, the pro-social action on the right. She has some incentive that she can impose, tax, or subsidy, or something. That would affect the material payoffs. Then that will have a positive affect on the right-hand term pro-social action.

The problem is that people have social preferences, which might have led them to take the pro-social action anyway in the first place. That’s not a problem. The problem is the arrow going from downwards from incentive down to social preferences. If that arrow is negative, so that the incentive has a negative impact on social preferences, then you have a problem. The total effect, the direct effect, that’s the top route going through the benefits and costs. The indirect effect is the effect going from the incentive to the social preferences and to the pro-social action. Typically, economics has ignored the lower root and in recent years, we’ve found out that the lower root really matters. What I mean by separability is that the total is just the direct effect. There isn’t any indirect effect.

Now empirically, there’s some bad news. The firemen in Boston in 2001 just before Christmas got a message from the fire commissioner who said that he noticed that people calling in sick pretended to get sick on Mondays and Fridays, and he found that irritating. Until then, they had had unlimited sick days. In fact, very few people abused the sick days, or at least very few people abused them very much, so then went up to 15. The commissioner then said, “As of now, if you call in sick more than 15 times a year your pay will be docked,” so it was a penalty. What happened on Christmas and New Year’s was the call ins doubled. What happened in the next year was over the whole year the number of people calling in sick doubled. The firemen who were interviewed said, “Sorry, I go to work when I’m injured. I go to work when I’m sick. I do that because I’m a public servant,” and so on, and so on. They felt insulted, so they just retaliated. That’s what’s called strong crowding out. That’s when the incentive not only blunts the effect, but actually reverses it.

Now Sandra Polania Reyes is, and that’s Sandra, by the way in Bogotá. She’s from Colombia. She’s doing an experiment in the streets among market people in Bogotá. She’s a fantastic experimental economist. We did a bunch of papers. In most of the experiments of which we collected a large number, there was crowding out. The incentive in economics would be called a substitute for the morals. A substitute in economics is the following. If an incentive and a moral preference is a substitute, then the higher the incentive, the less is the effect of social preferences. The larger one variable is, the less is the derivative of the outcome effect of the other one. That’s what a substitute is. Now that’s what we found out. We did find out that in some cases there was crowding in. Obviously, crowding in is gold. That’s what these mechanism design or social planner should be thinking of.

Now just to get a few words on the table, I want to talk about the nature of crowding out. It can be categorical. If there’s an incentive for something, you’ll say it has an effect. He did it for the money. How do you know that? There’s some tax incentive, I don’t know how big it is. That’s a categorical effect. I wouldn’t be as likely to, for example, give to the university if there was some tax advantage because the people might discount my generosity. Now the other is that there’s, sorry, there’s marginal crowding out. It just means that it depends on how big the incentive is. Obviously, you can imagine, psychologically, both could be the case, and I’ll tell you in two slides that we observe in experiments they both indeed are the case.

An entirely different dimension is the causes, of which I think there are two. One has to do with framing, that is the incentive makes it a different kind of situation. It doesn’t change your preferences, it just tells you, this is like shopping and not like dealing with my family. That’s framing, it’s called situation dependence in social psychology. Economists would call it state dependence. The states differ because in one case there’s an incentive or an incentive of a particular size, and in the other case there is no incentive. Those two states differ, and then we have state-dependent preferences.

Finally, a consequence of this crowding out phenomenon is that the idea in economics is the way you implement an outcome, you have some Nash equilibrium, which is somehow not okay, and you want to change it. What do you do? You change the situation, so that there’s a new Nash equilibrium, which has the desirable features. That’s called a mechanism design implementation by Nash equilibrium. The idea of the game is to shift the equilibrium, and it’s a very good idea because if you want to shift an outcome you better make sure that the desirable outcome you’re trying to get is actually a Nash equilibrium. Otherwise, it will be undone by private actions and the policy will fail. That’s not a bad idea, but we have to make sure that new Nash equilibrium that we’re desiring actually reproduces the preferences of which support the equilibrium. We have a rather unusual concept, remember where you heard it first, equilibrium preferences. What it means is what has to be stationary for a policy outcome to be the expectation that you get by implementing a policy is that given the preferences, which will result from these incentives, the outcome that you want will actually be implemented by private action. This is the analogy to the so-called Lucas critique.

Now let’s look at some data. This is an experiment. This is a state-dependent preferences subject both to categorical and marginal just using the terms I just introduced in the last slide. It’s a wonderful experiment with the public goods game. I won’t describe the details of the game, but the public goods game is an end person prisoner’s dilemma in which the dominant strategy is not to contribute, self-interested person would not contribute. Everybody else would contribute. Sorry, in this game the strategy of a purely selfish person, the person’s own benefit from the public good would motivate them, contributing 25. Here, there’s no incentive. Here, there’s an incentive of 12, and that’s 60. This is just you get an incentive if you’re a high contributor. This is all played anonymously with real money.

Let’s look at the red dots here. That’s what a selfish person would contribute with no incentive. If you add the incentive, then they would contribute more, if they were totally selfish. Then if you have a big incentive, they contribute a lot more. That’s what a selfish person would do. Now let’s look at what people actually did in the experiment. With no incentive, they contributed a lot more than 25. They contributed 37. This instance here is a measure of their social preferences or their ethical desires to contribute to the public good. When you add the incentive, sure enough, it goes up. It goes up from 37 to 38. If you just looked at that, you’d say the incentive worked.

Now look at this. If it goes up to 60, what do you get? There’s no effect at all in the social preferences. Those two numbers out there, obviously, statistically and significantly different. These individuals with no incentives made a very major commitment to the social good, which they didn’t do at all when they’re incentives. The differences, as you can see, were quite large. That’s crowding out. Now is it marginal crowding out or categorical crowding out? To do that, what I did as well, I said, “Suppose that there was an incentive of epsilon, small number and that would be somewhere right here right near the vertical axis, how much would they have contributed?” I just projected this line from that purple dot. That’s what they did with that incentive to that and then back to here and said, “That’s roughly what they would have considered if you had a vanishingly small incentive, but it was there.” That gives you 34.

Now notice this says that this is the categorical crowding out. The fact that this line is flatter means that varying the incentive has less effect. That’s marginal crowding out, so both are going on and both were quite substantial. When Sandra and I did this in lots of different experiments, it turns out that almost always you get both effects. It’s quite surprising. Okay, now let’s turn. This is the kind of information. People say, “Okay, how do you use this to design policy?” I think a combination of looking at these experiments and also looking at what social psychologists already know about why crowding out occurs will help us.

This is for Mark Leper whose work I admire very much. That’s the social planner. It suggests how complicated his, I guess his, in that case, task is. Leper says this: “The multiple social meanings of the use of tangible rewards are reflected in our every day distinction between bribes and bonuses, incentives and salaries. They carry different connotations concerning, for example, the likely conditions under which the reward was offered, the presumed motives of the person administering the reward and the relationship between the agent and the recipient of the award.” Yes, that sounds right. That sounds like something economists should pay attention to.

Let’s look at some actual policies. We have rural Colombians, and they’re playing a public goods game. These are real rural people, not students. The game was designed to look like how much stuff are you taking out of the forest because they were deforesting the place. This was an art experiment by Juan Camilo Cardenas who intervened there in Colombia before he became a grad student. He only found out later that he could publish papers on this. He was trying to change whether people could cooperate and not deforest the place. When he came to grad school he got a couple of publications from stuff that he was doing primarily because he was a community organizer, which is very nice.

If you look here, or there, this is the game played. What the vertical axis is is how far did they deviate from what a selfish person would do? Now the ideal thing for all of the society would be deviated by six. An entirely selfish person would have deviated not at all, so they were kind of in between. This says round after round, they’re halfway between being in the center and the same. Then he stops the game here, and he says, “Okay. The group that’s black are going to be subjected to a fine if they take out too much. They’ll be monitored.” If they’re discovered to have taken out too much, they’ll be fined. The group that’s red are going to be allowed to communicate with each other verbally, and then after that the game is played anonymously.

You see what happened here. At first, the fine worked exactly. They exactly implemented the social optimum, which is to deviate six from what a selfish person would do. As time went on, they went right down to zero. The fine wasn’t big enough. Essentially, these people who had acted halfway between being selfish and being completely generous ended up after eight rounds being entirely selfish. Sounds a little like the previous experiment. The ones who communicated, they did a little better. These red dots here are a little better than those. They did surprisingly not well to me. It was a class divided society, and these groups were of different classes, and they didn’t communicate very well. There’s some great stories I can tell you about that. I think the best way of explaining this is once the fine was announced, people engaged in what psychologists call moral disengagement. They just thought, “It’s not a moral problem.” They’re fining us, so not playing the game in a publicly responsible way is a commodity which I can buy despite paying the price.”

Here’s another experiment. This is a trust game. In the trust game, there are two parties. The first party gets a sum of money, which he or she then can allocate to the second party. Before it gets to the second party, it’s tripled in amount. The person who gets the tripled amount then can send some back to the first person. It’s like an investment. You invest in the second person, and they give you some stuff back, but there’s no way of enforcing it. You don’t have to be a genius in game theory to figure out that if you thought the other person was entirely selfish you should send nothing because they won’t send it back anyway. That’s the way the game should be played. That’s not the way the game is played hardly ever. People send a lot, and they get a lot reciprocated. If they send more, they get a lot more back.

If we look at just the gray bars, the gray bar says that if you send between two and four, you can get an average four back. If you send between five and seven, you get a little over six back. Essentially, the more you send, the more they send back though they don’t have to send back anything. People are reciprocal. They respond to that signal. Then the experimenters did a very clever thing. They gave the sender, the first person, the option to impose a fine. I can send a message to you saying, “I’m sending you five, and if you don’t send me back at least 10,” remember the thing is tripled. If you don’t send me back at least 10, then I’m going to fine you.

When the fine was imposed here, the black bars, that’s what happens. Now look at the black bars. What do they tell you? Much less reciprocity. They’re not reciprocating. Imposing the fine should improve the situation because you’re improving the incentives. That’s what economists are always trying to do to get the thing closer to a complete contract, but the closer you got the farther people came from the thing which would maximize the well-being of the two summed. Then, an even cleverer move with the experiments gave the option to say, “You can impose a fine, but you can publicly say, ‘I’m not imposing it.'” Here’s five. I can impose a fine, but I’m not going to impose a fine. Then what happens? Look at the white bars. That’s what happens.

Now why did they change anything? What changed? The fact that the fine was possible, but you didn’t impose it allowed you, the first person, to signal to the other person, I trust you. It said, “I’m not trying to get the most out of you.” This tells us something about designing institutions. A little change like that allowed a piece of information to be transmitted in a truthful way. I’m going to go past this because I want to get to … This is just a summary from a paper that Sandra and I wrote about how everything works.

After I published this paper in Science, I guess it was in 2008, Thomas Shelling, unfortunately, also departed wrote me this note. Dear Sam, “Your article in Science, blah, blah. I worked in Washington in The White House in the executive office of the President from November, blah, blah. These were exciting and stimulating times. People worked long hours and felt compensated by the sense of accomplishment, and I believe a sense of personal importance. Regularly, a Friday afternoon meeting would go on until 8:00 or 9:00 when the chairman would suggest resuming Saturday morning. Nobody demurred. We all knew it was important, we were important. Sometime in ’52, I believe it was, President Truman issued an order that anyone who worked on Saturday would receive overtime pay. What happened? Saturday meetings virtually disappeared.” That’s crowding out. President Truman was just introduced to the crowding out phenomenon, or the chairman of the committee.

Now other people who have real responsibilities face this problem. Consider the tax collector somewhere near Hyderabad. A number of people were not paying taxes. What he did was he hired a bunch of drummers to go outside the houses of the people and just drum outside the houses. He sent along a couple clerks with computers in case the people decided to pay up their taxes. It turns out it was very effective. Now why was that effective? It was effective because actually people know they’re doing something wrong when not paying their taxes, and they don’t want to be seen in that light by the neighbors. This drumming taxes out of the recalcitrant taxpayers, this, by the way, I’m sure the guy invented it himself.

Look at this. This is a picture from the 14th century, early 14th century of what’s called a Shari Vari. A Shari Vari in Europe was a method by which communities enforced moral standards on people by gathering outside the homes of the miscreants, the baker who charged high for bread when there was a bread shortage, the person engaging in sexual misconduct, and so on. It was very often women. They almost always had musical instruments or some kind of sound making equipment. You still find Shari Varis in Europe mostly on New Year’s Day now, just at celebrations. This thing about shaming people into conforming with social procedures is an old idea.

Now it’s been carried out also. I love the town of Bogotá. This is the mayor of Bogotá, by the way, a mathematician and philosopher, former rector of the university. When he became mayor he had a very serious problem, which is tremendous traffic fatalities on these so-called zebras, these striped things. Bogotá was really notorious for this, and it was a very serious problem. What did he do? He hired over 100 clowns to go and ridicule drivers, and there they are. You can see there. They were very aggressive and extremely funny, and then everybody talked about it. He appointed what he called the Order of the Zebra, which was a bunch of taxi drivers. They were the worst offenders.

There were 100 taxi drivers that got compliments from the pedestrians, and so on, then they had a little zebra they could carry in the car. There was a Lady of the Zebra. One of the taxi drivers was female, and so on. What was going on? He did a lot of things like this in terms of water shortage, and so on. Interestingly, he did not reduce the enforcement by ordinary policy. In fact, he upped it a little bit. This was a combination of making this a moral issue, how you drive, and then also, making sure they didn’t get away with it. By the way, which is another part of that great fresco in Siena. Go and have a look. It’s not about the virtues only, it’s also about getting the cops and the virtues to work together.

I’m going to close with what I think we may have learned about some of these cases. This is an iconic experiment. I didn’t put it first because you’d all say, “We all know that.” This is an experiment. It wasn’t actually an experiment done in Haifa, Israel. People coming late to pick up their kids at the day care. I think any economist here has already heard about this, but if there’s anybody in the room who hasn’t, I’ll tell you a few of the details. A sign was posted saying, “As of tomorrow, anybody coming more than 10 minutes late will be fined.” Actually, the fine happened to be 10 Israeli shekels. They counted before the lateness, and then they counted afterwards. The groups with the fine are the dark line, and the groups without the fine are the open squares. As you can see, what happened is once they imposed the fine, people doubled the amount of coming late. Not only that, they aborted the experiment here in week 17, not surprisingly, because it was inducing a lot of lateness. What happened was lateness continued.

Now, there are lots of ways of interpreting this, but I think the simplest way, the most parsimonious way and the way I think is true, the title of the paper is A Fine is a Price. “A fine is a price” means a price is something if you pay for it, that’s it. Nobody is going to arrest you on your way out of the store if you paid for the good. A fine is a price. What fining people for coming late is, you turn lateness into a commodity. Step right up. If you want to pay 10 Israeli shekels, fine. Now if you want to ask me if it had been 100 or 1,000, would they have come late? Of course, they wouldn’t have come late. That’s not the question. Of course, there’s some incentive which will get people to come on time, but that particular incentive wasn’t enough to do the job. That experiment is talked about a lot.

Think about this. Think about another fine. It looks identical. Ireland imposes a small tax on plastic grocery bags. It was preceded by a huge media campaign about don’t trash the Emerald Isle and so on, and so on. Within two weeks, no one was using bags. It was an extraordinary elimination of plastic bags. Plastic bags were like wearing a fur coat in Berkeley, you just didn’t do it. It’s the same, impose a fine. The fine was so small it’d be hard to think that it got rubbed out just because of the fine. You can’t exclude that possibility, but it’s probably not just the price. It probably crowded in social preferences somehow.

Now there are lots of differences we might draw upon, but I want to come back to Bentham, who I chided before, because this is in the morals and legislation. A very wise thing that he says: “A punishment should be a moral lesson when by reason of the ignominy it stamps upon the offense. It’s calculated to inspire in the public the sentiments of aversion towards those pernicious habits and dispositions with which the offense appears to be connected, and thereby to inculcate the opposite beneficial habits and dispositions.” A fine is a lesson. It’s not a payment. It’s teaching something. A punishment is supposed to be a teacher. Who does that sound like? That’s Aristotle. That’s Aristotle saying, “In doing this, we’re trying to sustain or create a population of citizens capable of good governance.” And that’s what the Shari Vari is all about too.

Now I don’t want you to go away from this thinking that fines are wrong. The Irish case suggest that that would be a mistake. Here’s the case. This is a public goods game in which here’s the standard game. You can contribute or not. You can contribute 20. People start by contributing two, and then it goes, basically, down to zero. Basically, people lower their contribution because other people are not contributing, and the only way they have to punish them is by not contributing themselves. Then here at period 10 you change the rules, and you say, “After each round you’ll see on your screen, not the names, of course, but the numbers of all the people in your group and how much they contributed.” You then have the option of taking some money out of your own kitty, out of your own pocket and reducing the finding of somebody else. We never use the word punishment, of course. We just say, “You can pay to reduce the fining of somebody else,” so they start off a little higher, and it goes up.

Now what happens here, those places there, is there are a lot of people not contributing much and people are punishing. By the end of the game, everyone is contributing a lot and nobody is punishing. Now notice, that’s a very small change in the rules that had a huge impact. Here, there were a few selfish people who were driving this downwards because the ones who would have liked to have contributed were getting angry because the other people weren’t contributing. Over here, possibly a small number of people who are civic minded and willing to actually punish the others were driving the equilibrium. Now there’s a wonderful puzzle here, by the way.

This game here is a public goods game in which the dominant strategy is to not contribute. That means whatever anyone else does, what you should do if you’re maximizing your self-interest, is don’t give. This game over here, the punishment game, it’s the same game. A punishment is a public good. The dominant strategy about punishing is you shouldn’t punish. Let somebody else punish. In the second round, people should have been as selfish as the first. It’s a very interesting characteristic, and I think there are good evolutionary reasons why we have this, that people here were acting selfish, but when it came to the ability to punish people who were breaking rules, people are very happy to participate. It would take us somewhat far afield to explain why that would be true evolutionarily.

If you compare these two games with which you’re familiar, the top one I’ve just described, and this was the one in which the incentives drove out reciprocity. That’s a contrast. In the one case, fining people over here, fining people was leading people to contributing more. They didn’t contribute less when they got fined, they contributed more. In the lower one, the prospect of a fine made people less reciprocal. In one case, it seemed to be building up social preferences, the top, and the other case not. I was able to look into why this experiment worked out the way it was. Remember, I can give you five, and then I can say, “I want seven and a half back.” Then we’d split the total, or I could say, “I want $10 back.” Then I’d get most of the total. The ones who were greedy who were the ones who were being negatively responded to. The negative response was not a response to incentives. It was a response to greed. The ones who were saying, “I’ll find you if you don’t split it 50/50 with me were not negatively affected.” There’s a lesson there, that it’s not the incentive, per se. I’m going to run through these.

I want to go back to Aristotle. By the way, this was during Aristotle’s lifetime, but I do not know that he was involved, but the date is right. There’s a mission to set up a colony in the Adriatic because of the Etruscan pirates. The ships had to be recruited from private lives. The ships were owned, and the crews, it was a huge operation. There was 200 ships, and horses, and soldiers, and sailors, and so on. The people who were wealthy were appointed to deliver a fully equipped ship to the port, Piraeus. Fairness was assured. This is amazing, but what the council said is okay, these things were called liturgies. I’m not sure why. Hundreds of them were given out to wealth people and said, “You have to produce this ship, you have to do that.”

Now here’s how fairness was ensured. Brilliant. The liturgy was allocated in the following way. Suppose I got a liturgy, but I happened to know that Michael who didn’t get a liturgy was wealthier than me. I go to Michael, and I say, “Michael, take my liturgy.” If he refuses, then his alternative is we exchange our wealth. Got it? I’m a citizen. The citizen’s council doesn’t know how wealthy he is. They just thought I was wealthier, but I know, I’m his neighbor. I don’t know how many cows he has or whatever. I go, and I say, “You take my liturgy, or I take your cows.” That’s mechanism design at its best because it’s using private information, the fact that I know he’s wealthier than me to essentially allocate the burdens of this thing. This was a set of incentives, which were fair.

The incentives were given as prizes, they were not set as incentives. For example, the herald of the council of 500 is to announce the crowns of the contest of the festival in order that the competitive zeal of the triarchs would be evidence, and so on. A clear public purpose was stated. Substantial fines for those who didn’t do it were imposed. Now think about if the Athenians instead of two economists had designed the Haifa day care experiment. It was two economists who did it. They do great work, it’s a wonderful experiment. The Athenians in their time machine, they got to Haifa first. How would they have done it?

This is what on the door of the day care centers in Haifa. As you know, official closing time, blah, blah. As of next Sunday, a fine of NIS 10 will be da, da, offered. That’s what the economists did. The Haifa people, they didn’t make it in time, but here’s what they had planned to put on the door. They would have said, “The council affairs wishes to thank you for arriving on time to pick up your children since this reduces the anxiety that the children sometimes feel and allows our staff to leave in a timely manner to be with their own families. We recognize that all parents who have a perfect record of unblemished by lateness for the next three months with an award of 500 NIS to be given at the annual parents and staff party with the option to contribute your award to the school’s teacher of the year celebration. Now do you think if that’s what they had had on the door they would have people coming later? I don’t think so. We haven’t run that experiment, but if the Athenians had just gotten there first, it would have all worked out a lot better.

Now the economy and society into which we’re now entering couldn’t be farther from the world of James Buchanan dealing with his fruit cellar. We’re not living in a world in which we agree on the property rights being transacted. We’re not living in a world in which everything that matters that goes on between is covered by a costs-enforceable contract whether it’s secondhand smoke, or climate change, or epidemic spread, or hiring a worker, or giving a loan, or taking a loan. Think of the 18th, 19th century. The sinews that connected the world was trade, and grain, and steel. That’s not what connects us today. We do that, but it’s also sneezes and burning carbon, and all kind of other things that will never be subject to contract.

The economist’s job until now is extremely valuable still, perfecting incentives, which will work even in the case of self-interest, but these will fail if they compromise moral commitments that are necessary to govern a modern society in which incomplete contracts is just a very dangerous kind of utopia. Now securing a sustainable and just future will require that we take seriously as Aristotle did, the kinds of moral development that we are engaging in in our society. I don’t underestimate how troubling this is to any liberal. I’ll say twice now, and I count myself among them who sees this as also a problem in society, a society which could be dominated by the Commissar of culture, and so on. Let me end with a hopeful thought.

I had the word CORE next to my name. Curriculum Open Access Resources for Economics. This is a new introduction to economics based on modern economics, not on what’s in the textbooks today, which is mostly something that came from the 1980s. We take, for example, incomplete contracts as the normal case, market failures as the standard case, and so on. It’s available on your telephone. You can get it right now if you want. I will not count it as being impolite. There it is up there. It’s around the world. It’s already accepted as the standard course at the Toulouse School of Economics thanks to Jean Tirole and at the UCL, the top economics department in the UK and one of the top in Europe, Humboldt University in Germany, and so on. It’s also available in a book form. Crazily enough, a publisher agreed to publish it even though it’s available on everybody’s telephone. It’s a very new way of teaching and learning economics. It’s highly interactive. We’re digital first. We didn’t do a book. We just did a platform.

We also have a course called Economy, Society and Public Policy for non-majors, also being used at graduate level of public policy courses. We have a new course also called doing economics which is a fairly sophisticated teaching of how to handle data on economics using a lot of the economic arguments and cases in these other books, but it’s basically a data handling, and it’s very popular among students. That’s it. Thanks to all of my collaborators, and thank you for coming.

Rui De Figueiredo: Thank you very much for a extremely stimulating lecture. Couldn’t think of one that fit the Weinstock Lectures better. We have time now for questions and discussion. The only thing that we would ask. I’ll let Professor Bowles call on people himself. Ellen will provide a microphone. Just we would appreciate if you could keep your comments or questions brief, so we can hear from Professor Bowles as well. Thank you. Just put your hand up.

Samuel Bowles: Could I ask the first question to come from a student? You can’t lie if you’re a student. I know you’re not a student.

Audience 1: Hello. My name is Kelsey Jones. I’m a student here in the Energy and Resources Group. I worked a lot with indigenous communities who were facing development projects implemented by large transactional companies. This is fascinating to me, and I’m wondering if any of the experiments that have been run include scenarios in which there’s a very large power imbalance? It sounds like most of these games include people who are generally operating maybe with the same resources or with the same incentives. Could you just speak more about these games might be different, or whether they’re experiments that have been done under other scenarios of those types of large power imbalances?

Samuel Bowles: There are two kinds of power imbalances in the experiments. There are experimentally induced power imbalances, that is you give a principal a lot of power. I can set bounds on what you do. That’s an experimentally given difference, but then there are real world differences in power, that is the people actually … The experimental subjects come from different locations. I’ll give you two examples. I mentioned the great work that Juan Camilo Cardenas in Colombia does and how the communications didn’t actually improve things very much. I said the problem was that the groups were heterogeneous. In one case, in one of the cases I represented, these were groups of five. There was one group in the public good game, there were three women and two men. The older man who’s name was Don Ricardo and his son, they were the two richest guys in town. The two rich guys in the group, obviously, a huge imbalance. They, the two rich guys were trying to persuade everybody in the group, look, don’t take anything out of the forest. We’ll all do better, so do the optimal thing, six, deviating six.

Then Juan Camilo noticed the group hadn’t done well. He asked one of the women, “What was the problem?” She said, “Don Ricardo was saying we should all not take anything from the forest.” Juan Camilo said, “Why did you take so much?” She said, “I don’t trust him. I could never look in his eyes. I didn’t really believe him.” I asked Juan Camilo what had Don Ricardo done? He had taken the maximum, and his son had taken the maximum. The women in the group had taken some middle thing. That was a general result, that is groups of mixed wealth in his experiments did extremely poorly. Also, homogeneous wealthy groups did poorly. The groups that did best were homogeneous poor people, and I don’t know if there’s a lesson there.

The other thing is that in some places there are a lot of these fantastic experimental economics community based in University of Cape Town. Justine Burns is a name to follow. They do a lot of experiments across racial and class boundaries. You see how do white and black people, and how do people of Indian origin, and so on relate to each other in these games. Those are two things, but I would ask Juan Camilo Cardenas and Justine Burns, and follow on. I hope you find something, or maybe think of a good experiment to do yourself.

Audience 2: Hi. We’re a group of Ph.D. students. How do you think we could kind of apply these ideas or new framework to nonexperimental kind of empirical work?

Samuel Bowles: I think if by this framework you mean the crowding out by incentives and so on, what you’re always looking for is a natural experiment of some kind like changing the rules for giving blood and so on. Those are quasi-experimental, but if you mean the more traditional econometric work, I don’t think it’s been done. The problem is when you have radically different incentives, so that you can actually see a difference, too much else is also changing. I’ve just been trying for a long time to say, okay, a society which lives primarily monetary incentives is a different kind of society culturally than the one that basically deals with other kind of ways of enforcing norms.

Then it’s very hard to find clean changes. National Boundaries, as you probably know, in Switzerland, there’s some very nice cases about two sides of the German French boundary, German-speaking, French-speaking boundaries. It’s very hard to do. It would be very good if you could. What’s been done is there’s been a lot of studies of the validity of the experiments, that is seeing whether or not fishermen who contribute to the public goods game are also those who engage in environmentally friendly fishing practices. The answer to that is yes. That’s showing there is a real world analog to it, but it’s not the kind of thing you’re asking for.

I wanted to ask, is it necessary that the actions are visible to others? I’m thinking you can be immoral in an invisible way, like a company that is polluting or something. If they’re not accountable to anyone, they can still pollute. Another thing is we also maybe need communities to which you are accountable, is another question I have. Do we need that, those two things?

Yes, I agree. You probably know at least as much about that as I do because of your research. The role of community is suggested by the fact that when you have a peer punishing in a public good game it has a big effect. By the way, that effect takes place even if we introduce what’s called the perfect stranger treatment, which means every round the groups are shuffled. If we’re in a group together now, and I don’t contribute, and you punish me, you couldn’t possibly be doing that in order that I would shape up to your benefit because you know I’ll be somewhere else next. The other thing which we notice is being punished by a peer who apparently has no selfish reason for doing that has a big impact, that is it’s not the punishment. For example, in these games a verbal admonishment is almost as effective as a financial fine.

What people care about is they care about their own standing in the community. When you said, “How important is it that it’s observed?” distinguish, please, with me between the two following concepts. One is guilt and the other is shame. Guilt is something that you feel privately even if nobody knows about it. Shame is what you feel as if you’ve done something about which you feel guilty, and it’s publicly recognized. Now I think it’s useful to have those distinctions because both are at work. When I do something good, I do something generous, most economists would say, “You probably did that because you’re signaling to others your generosity, and that will improve the kind of trades that you can engage in in the future,” which of course could be true.

There’s a concept in psychology called the looking glass self. You’re looking in the looking glass. A lot of our signaling we do to ourselves for ourselves, and here’s why. We’re never sure about our goodness, our bravery, our kindness, or whatever it is. We occasionally engage in activities to reassure ourselves, to daily reproduce ourself, reconstitute ourselves. That kind of self-signaling is something which goes on even in the absence of others observing you. Now how important is one, and how important is the other? It surely depends on the context, and so on. When people do things which are kind, brave, and so on entirely with nobody looking it could have a part of that looking glass self aspect of it.

I think the last question is, your last question about the role of a community, it’s absolutely essential. The difference between Ireland and Haifa was the explanation at Haifa really deprived people of any opportunity to say, “This is something we should be doing together” whereas for accidental reasons the deal by which the law got passed in Ireland had as part of it a massive publicity campaign, which said, “Look, we Irish people are all in this thing together. Garbage bags are really anti-Irish,” and so on. It was a notion of an us that was implementing that.

What you said was very interesting, but I would like to ask one of the tenets in Buddhism is right action independent of results. Do you have any comments on how all you’ve been saying tonight might work in a Buddhist community?

I have thought about that. I spent a lot of time in India when I was a youth. I was very attracted to Buddhism. Unfortunately, I think one of the leading experts in the world on Buddhist economics was just sitting behind you. Unfortunately, she left. We should ask her. The idea of doing good because inconsequential, that it’s not for consequential reasons, but because it’s the right thing to do, I think that’s a lot of our morality. In economics, the term other regarding preferences is used. It means I’m doing this action because I have concern for you. The way that’s done in mathematics is I have some benefits of my action, and I care about the benefits of yours, so I put some little extra next to your benefits. That’s a lot of what generosity and being a good citizen is, but a lot of it is not, that I’m doing something like I’m transferring something to you. I’m doing something because it’s the right thing to do deontologically, that is not consequentially. That’s a fundamental aspect of Buddhism.

For example, a lot of the self-signaling is actually reassuring ourselves that we are good quite apart from who’s looking. I think there’s a missing set of terms in the way, at least in economics, we think about these things because we’re always … If we get away from homo economicus, then what the individual is is putting a weight on someone else’s well-being instead of being a moral being in its own right. Both are important, obviously.

Kind of maybe a little bit of a more bizarre question, but I’m thinking of, you mentioned Buchanan. I wasn’t clear if you were sort of trying to create sort of this comparison between a past time when economics was simpler, tapped into simpler relationships, and now, particularly as someone who likes to do comparative studies that incorporates the past. Couldn’t, say, Buchanan’s relationship with the fruit vendor be problematic, say, the fruit vendor spit on the fruit or something like that, this idea that Buchanan tapped into this particular economic phenomenon that is, as you mentioned, related to the property and the belief that, anyways. Could you explore that a little bit more, and could you actually tell me if that’s what you mean, or am I misinterpreting you?

No, I can say something about that. For most of the time, when biologically modern humans have existed we did not live in Buchanan’s world. We lived in worlds in which we interacted with a relatively small number of people, not as small as most people think. It could be dozens or even hundreds, but we interacted with a group of people in which there were no contracts. There was no state to enforce the contracts. If there were, there was expectations, promises, and so on. None of that was remotely like the perfect contracting rule, nor were there rules of private property applied to virtually all things of value. A few things were.

I’m talking about hunter-gatherer life up until 11,000 years ago, and even the first 3,000 or 4,000 of the neolithic. There was very little of what matters with private property. For example, almost certainly not land, but probably stores. As a species, we evolved to do very well in economy in which there was a limited role for private property and contracts. I think that the world that Buchanan was describing was really a very short period of time in which a lot of the stuff that mattered, you could actually weight and measure, and therefore transact it in a contract. Maybe half a millennium or something like, some period of time. I think from what I said at the end, I think we’re now moving into an economy in which that will increasingly not be the case. The most important stuff that’s going to go on in a modern economy you cannot weigh. You can’t measure it. You can’t contract for it. That’s knowledge-based, that’s affect. That’s caring for people.

The knowledge and caring-based economy is a lot more like the hunter-gatherer economy than it was like the economy of Buchanan and his fruit stand. We may be able to learn a lesson from that. Obviously, given the modern communication abilities, given also modern states, the modern means of coercion, and so on, it’s an entirely new world. I’m not suggesting we can model our society on a sort of modern day Flinstone economy, but we do have something to learn. Our capacities to cooperate are really extraordinary. If you look at human institutions that have been a big success, the cooperative and egalitarian hunter-gatherer society is far and away the most successful enduring for 100,000 years. Capitalism is just a few centuries old.

I think we may be moving into an economic situation in which, if I can quote Kennith Arrow again, it wasn’t on the slide. Arrow said, “Information, knowledge is a fugitive resource. It runs away.” Just like the animals ran away from the hunters and gatherers, you cannot own it. He said, “We’re just beginning to understand the contradictions between fundamental economics and property rights.” He was talking, and this was three decades ago. Arrow was talking about a fundamental problem about moving into an economy in which it’s not possible to own stuff, and when you try to own it with intellectual property rights, it’s actually dysfunctional for society. We’re moving into a society in which those rules don’t work. Arrow was saying three decades ago, “We’re going to have fundamental institutional changes because the stuff the economy is now made of doesn’t work anymore for this private property and complete contracting regime.”

Now I don’t think it means we’re going to do away with private property. I certainly don’t think we’re going to do away with contracts and markets. It means it’ll play a much more limited role, and we better find out some way to deal with the other stuff whether it’s intrinsic motivation, trust, morality and other things. Arrow didn’t say what they might be.

Audience 3: Hi. Do you have a stance on whether we should use public policy to foster social norms? For one point, it’s clear that some norms are say, trustworthiness, and being generous, and so on, but saying forcing egalitarian norms would be much more controversial.

Samuel Bowles: I think this is a very difficult question. Having listened to my lecture, it’s a question which absolutely has to be asked. If you take the Aristotelian position, you must also then say, “What would it mean to be a good citizen?” Now there’s a easy way to answer the question, which is states today are engaged in inculcating norms. They do it in schools, they do it through patriotism. They do it through political leadership, and so on. It goes on all the time. I could say at a minimum, what I would like us to do is have a public discussion about which norms we would like to see brought to the fore.

Now if you have a reliably democratic society in which sincere communication and debate following Habermas, for example, as a philosophical principle, you could see that might be a framework in which we could then have some confidence in the kinds of norms that would be inculcated. I don’t think we have any reason to think that that’s going to solve the problem. I think there’s a lot to fear from it. For example, I think human beings are generous and also very given to us and them distinctions, both. We’re generous towards the us, and we’re often indifferent towards the them. I think we evolved that way genetically. I worked on that for a long time in biology. I think there’s a very good case that that’s who we are. If that’s true, that’s a problem because we’d like to inculcate the altruistic and common norms. We’d like in most cases, not all, but in many cases to obliterate those repugnant us and them distinctions, which work to the disadvantage of the less well off and people of other groups.

I don’t know how that can be done. I don’t think it can be avoided. I think the idea that this is going on anyway without an open discussion about the people we would like to be, and I mean globally, that’s certainly a problem. Whether or not bringing that discussion to the fore would address the problem and address all of the possibilities of norms, which would come to the fore in that discussion might be racist or homophobic, or other repugnant things. That, of course, we can’t know, but thank you for your question. Are you from economics?

Audience 3: Yeah.

Samuel Bowles: Don’t write your dissertation on that one unless you have a really good idea. By the way, a brilliant dissertation was written in Harvard in the mid-60s by Herbert Gintis on endogenous preferences. You might go and find it. Marchesin was on his committee, but have a look. Some people have gotten away with writing dissertations about really big questions like the one you asked me, but as I say …

Rui De Figueiredo: We’ll have one final question from one of the committee members.

Audience 4: Thank you very much. I’d like to ask a question not so much about the economy part, but when you were speaking about Aristotle, and being a good citizen, and also the title of your book, I started thinking about hundreds of years later Michel Foucault. I’m wondering what you would consider the distinction is between, I think, a fundamentally liberal notion, little l liberal notion of the good citizen and maybe Foucault’s notion of the governable citizen or the governable subject, which has always struck me as somehow at its core not very liberal.

Samuel Bowles: This is a time in which people are reticent to criticize liberalism and for good reasons because liberalism is in serious trouble. One of the reasons why liberalism is in serious trouble is the very limited nature of the idea of the individual and the role of society, recognizing families and states, basically. That seems to me to be a major flaw, if you look at the history of liberalism. I started working on a paper. It was if John Stuart Mill had been a Florentine, if he’d come from Florence, Italy, because there the idea of liberta was a both individual and a community idea. That was a fundamental, philosophical commitment of those places. Machiavelli of the discourses was a very good representative of that. Now I think it’s the lack of the emphasis on community between family and nation that’s the flaw. I don’t know how that relates. Why don’t you say? What do you think the relationship between that of the governable individual of Foucault is?

Audience 4: I’m sure there are other people who know much more about Foucault than I do, but I always interpreted Foucault as saying that laws, and norms, and schooling, which you referred to is to make people sort of docile. Live harmoniously according to the convenience of a stably functioning society, and that doesn’t carry the connotations of individual autonomy, agency and liberty that the liberal vision of the good citizen typically carries all the way from Athenian times, which I thought you were more harkening to when you spoke. That’s all.

Samuel Bowles: Again, this is a long story. If you think of early liberalism, it was a body of thought that defended the weak against the strong and advocated for tolerance. For example, even the early advocacy of property rights, remember, it was advocating private property rights against states that were run by the elite and run by the rich. The idea of private property rights was a defense against a state, which was an autocratic state of the rich. If you think about a lot of early liberalism, it had a definitely pro, less powerful outsider. Then at a certain point, obviously, political liberalism married economic liberalism laissez faire. That was the beginning of the current troubles of modern liberalism. That turns out to be a very unstable marriage because it went through, as you surely know, liberalism was not democratic. It became democratic early in the last century when the working class and women got the vote.

Now we have a situation of a politically empowered working class, which is living under an economic model, which is destroying their communities and essentially serving them very ill. That’s a dangerous, dangerous situation. I recently wrote a paper called The End of Liberalism. By the way, it was not a prediction. The double entendre was intended. The end of liberalism used to be defense of the weak against the strong. I meant people to think maybe it’s over, but maybe we should to rewed liberalism to its original ends. Having done that, I wouldn’t say that is all that we could expect from a normative position for the good society, but we better defend it now while we need to.
Rui: With that, thank you very, very much. It’s been a pleasure.

Outtro: You’ve been listening to a podcast by University of California Television. For more information about this program, or UCTV, visit us online at