 What are the welfare implications of satisfying people's true preferences if they don't recognize that they are their true preferences? Does it make them happier? In fact, it might make them worse off. If somebody doesn't realize that his or her true preferences are not to eat junk food, but thinks that their true preferences are to eat junk food, then putting a tax on junk food or somehow inhibiting them from eating junk food doesn't make them happier. It makes them worse off. I'm Mario Rizzo. I'm a professor of economics at New York University and the co-director of the Classical Liberal Institute at the New York University School of Law. I thought that behavioral economics in some ways was a refreshing development in economics because a lot of things had been taken for granted in the standard new classical approach, which I thought should be questioned and should not be taken for granted. But they connected it very closely with a kind of policy agenda. So what I wanted to do was to see whether the policy agenda really followed from the behavioral foundations that they were laying and just sorted all out in terms of the theory and in terms of policy agenda. I would say the behavioral economics, the fundamental concept of behavioral economics is something called decision-making failure. In standard economics we have a notion of market failure when markets don't work well to satisfy the preferences of consumers and there are many cases, monopoly, externalities, whatever. But this is a case of decision-making failure that is to say when individuals do not make decisions that are consistent with their own best interests. And so I think that's the core behavioral economics is decision-making failure. What follows from that is the behavioral policy analysis. In fact, there's a journal called Behavioral Public Policy which is devoted to the policy implications of the idea of decision-making failure. Behavioral biases that have been catalogued by this Wikipedia page called Cognitive Biases amount to about 175. Now in all fairness the biases listed are not all separate from each other but it gives you a sense of the behavioral literature, not only behavioral economics but the cognitive science literature which has developed this idea of biases, how many they have come up with. So there are many biases. I think probably the most important of these biases is something called present bias. And present bias is the idea that people when they're making decisions for the farther future tend to make quote-unquote rational decisions but when these decisions are upfront when they refer to say the next period people tend to be myopic and they tend to make decisions which are only in their short-run interest. So it's this conflict between kind of short-run interest and long-run interest with the bias being the short-run interest that's taken the focus of behavioral economics I think. There are many others but I think I would say that's the central bias. What behavioral economists used to call new paternalism they're now calling behavioral paternalism and behavioral paternalism differs from the classic form of paternalism insofar as the classic form of paternalism was not about a person's own true preferences. It was about some objective criterion for what is good for a person and sometimes it was a moral paternalism you know you should go to church on Sunday or sometimes it was a health paternalism but it had to do with some objective criteria. Behavioral paternalism is a little different. It purports to be about not objective criteria but the true preferences of individuals of what they really want to do not what somebody else thinks they should do. So it's still rooted in a kind of traditional economics framework but has to do with preferences that people would have if they were free of biases. So it's this sort of bias-free preference. Now the issue of course is how do you discover bias-free preferences and that's where a lot of the debate is today how easy it is to discover these and whether behavioral economists have been successful. Now I've argued among other things that it is quite difficult to in fact determine bias-free preferences probably in part because there are so many biases so one have to systematically investigate a situation to determine whether various biases are present so you go through essentially a checklist and it would be a lot of things to sort of check off. Unfortunately in most of the analysis people seem to be content with analyzing one or two biases and therefore coming to rather simple conclusions about what bias-free preferences would be. So there are really important epistemic problems here that I think need to be further explored. In the junk food area there would be the so-called fat tax or soda tax. It depends on what kind of junk food you're talking about. There are some jurisdictions which have a tax on the sugar content of soft drinks or non-alcoholic drinks. So what those are supposed to do is to dissuade people from drinking sugar sweetened drinks with the idea that that will help the obesity problem. Unfortunately enough there have been studies to try to investigate whether these taxes or other nudges away from junk foods do any good and there's no real evidence that it does. Partly it's because for example in the soda case people switch to artificially sweetened sodas. Artificially sweetened sodas if you hold everything else constant will cause you to lose weight but in fact they don't satisfy your appetite as much. People tend to eat more. They feel it's a license to eat more and so everything is not constant and in fact people who consume the sodas are not more likely to control their obesity problem than anybody else. There are other examples. One example which is a little bit strange because it doesn't really involve government is setting up a default of automatic enrollment in retirement savings programs. Now the background is this. Many large employers have employer sponsored retirement programs. It used to be that you had to opt in to get to be part of the program so that if you did nothing you were not in the program. In recent years more firms have developed this idea of automatic enrollment so if you do nothing you're enrolled and you're enrolled at a certain rate in a certain retirement vehicle. Now the way it comes into behavioral economics is not in the as I say in terms of government policy but in the idea that this in fact is the reason that there has been greater enrollment in these programs over recent years. Automatic enrollment because it would seem that what's the difference. If you could sign in or sign out it's all very simple. But behavioral economists have claimed it's due to some behavioral nudge that people are greater enrolled. We have argued it's due to more information and a recommendation on the part of employers that is manifest in the automatic enrollment. So people are doing it for weakling purely rational reasons and not for behavioral nudge reasons. In economics there are certain criteria that decision makers have to fulfill in order to be considered rational. And one of the criteria is that they have to have a complete preference ordering. In other words they have to have an opinion, a clear cut opinion about every option in front of them. Other criteria are that they have to be consistent in their choices. And you go on and on with a number of other criteria. Now the behavioral economists quite reasonably say people aren't like that. They don't behave in accordance with all these rationality criteria. They're not mechanical, economic men or women. And so it doesn't make any sense from a descriptive point of view. But, and then they don't say this explicitly. But they in effect hold to this. That this is the way people should behave. So the goal of policy is to incentivize people to behave as rational economic agents. And we think that's too narrow. There are plenty of reasons why people might be inconsistent in their choice. That doesn't make them irrational. It may be that they're experimenting with different options. It may be that they haven't decided what they want to do. And so on and so forth. So we think from, in that sense, the behavioral economists have an incomplete revolution. The complete revolution would be to reject the rationality criteria even at the level of normativity as well as the descriptive level. Inclusive rationality is the idea that we ought to be broader about what we consider reasonable behavior. So that reasonable behavior ought to be the standard not technically rational behavior. And as I say, reasonable behavior could be simply that a person has not fully decided what they want to do. And that's why they're inconsistent. That's reasonable in a world of high information costs and things of that sort. And interesting, there's an economist by the name of Isaac Jilbaugh, I think is his name, who says his favorite definition of rationality is if you can give reasons for your choice that don't make people laugh. Then he says, then that's a reasonable choice that you've made. And he would consider that rational. I tend to go along with that joke. Some psychologists have actually, and we discuss some of this, have argued that there is a functional value to biases. That sometimes biases can inhibit people from making decisions that they otherwise would not want to make. For example, some biases cause people to procrastinate in their decisions. Now, procrastination could be viewed as a bad thing, or it could be viewed as time out. George Ansley, a psychologist, a very prominent psychologist, has argued that many of the biases have functional value and they are important and do contribute to people's broader rationality. That's a whole other approach, but I think it has a lot of vitality. I think that Knight's contribution is important here because most economists, both behavioral and neoclassical, like to model all conditions of uncertainty as if they were purely risk, so that the probability distribution is known. Now, this comes out implicitly in the idea, for example, that certain habits like eating junk food or smoking, etc., are bad for your health in the context of young people. In order to make an appropriate economic decision as to whether to eat junk food, for example, or not, in view of the health risks, one has to have some sense of what the health risks are going to be 20, 30, 40 years from now for a typical 20-year-old. This is not something which is determinable by a fixed and objective probability distribution. We don't know the future course of medical technology. For example, people in 1960 who were obese were in much worse health than people today who are obese because of all the drugs and various other medical procedures that we have now. Who knows what people who are obese in the year 2050 or 2060 are going to be like in terms of health, and therefore what the real risks are of these habits today. So, in one sense, the Knightian approach warns us not to assume that we know what the unique rational decision is for decisions for young people, for example, in smoking or eating junk food. It cautions us that there may be more than one, what reasonable decision to make under these conditions of radical uncertainty. There is a distinction that psychologists have made between sources of self-control, external and internal sources of self-control. I think the terms are kind of self-explanatory. So internal self-control, a person comes to the recognition that a certain behavior is bad or either their health or morally, etc. and they exercise willpower and self-control. Even before you get to self-control, there's something called a broader concept of self-regulation. Self-regulation is not the immediate sort of stealing yourself against the chocolate cake, but it is more subtle. You don't go to restaurants that serve a delicious chocolate cake or you put the chocolate cake on the top shelf where it's inconvenient to get and it reminds you not to get it. Things of that sort. So that's internal forms of self-control or self-regulation. External is somebody scolds you or there's a tax on the sugar or whatever coming from the outside. Now the relevance of this is that psychologists have done studies that indicate that self-external control and internal control are treated by people as substitutes. The more there is external control, the less they engage in internal control. So what's wrong with that? What's wrong with that is that it has also been found that the substitution or the reduction of internal control doesn't only occur in the area where there's the external control. It generalizes. In other words, people become more accustomed to leaving it to somebody else so that there have been experiments that show that in completely unrelated areas when people were subject to external control, their self-control was reduced and vice versa. I have to admit the research is not definitive but it could actually lead to a condition where people are less able to do for themselves than they were prior to the nudges and paternalism. The political economy of paternalism is fraught really because it oftentimes politically is a combination of what we call and has been called in the literature the coalition of the bootleggers and the Baptists. The Baptists mean the people who are truly concerned with making people better in some sense morally or here health-wise, for example. And the bootleggers are the people who have some special interest to gain out of the policy. I would add a third category and that's the category of the clueless voter. The clueless voter doesn't really know what the correct amount of junk food eating should be according to behavioral economics and its standards. It's clearly not zero because they admit that people get pleasure from this. So it would tend to be the case that in political terms it's never enough. In other words, you could always say that as long as people are eating junk food we should do more to intervene. And when the tax isn't enough then the temptation for harder forms of intervention are present. And so the idea is that because the standard is so hard to determine in practice that politically it will tend to be simplified into all or nothing and therefore become more severe in its paternalistic aspects. There are groups of economists who want to reanalyze and reevaluate the notions of rationality that have come down to us. They tend to be not in the mainstream by and large. I know INET, for example, has many economists who are interested in this sort of evaluation. The Austrian School of Economics has people and even behavioral economists. I mean there are some behavioral economists who are not particularly happy with the public policy aspect. So I think there is a growing group of economists outside the mainstream who are interested in reevaluating these ideas. I think ultimately the mainstream will get the message, but I'm not sure when and how. For a young person thinking of entering economics I think it's important that they understand and learn well the mainstream approaches. But that they take away from that some sort of critical faculty. That is to say they take away from that the idea that although this is the established mainstream approach it is not perfect, there are good arguments against it and maybe to think about how they and their careers could improve the approaches that economists take. But I think it's vital for every economic student to have a PhD, economic student, to have a good solid background on what the contemporary approach is. I know Milton Friedman used to say about Keynesianism. He said you have to understand Keynesianism better than the Keynesians in order to criticize it. So you have to understand standard new classical economics better than they do in order to criticize it.