 Good morning. I'm delighted to welcome you back to the second day of the McLean conference. We have, as you can tell from the brochure, a very intense full day of discussion. I'll not take any time except to introduce Laney Ross, who's an Associate Director of the Medical Ethics Program, a Professor of Pediatrics and at the University, the Bucksbaum Professor in the Departments of Pediatrics and also in Medicine. Laney will chair the first panel and I'm just going to turn it over. Laney. Thank you very much and thank you for getting up so early to join us. I was a little confused when Mark asked me to chair this panel, maybe because I'm a married to an economist. It counts. But as anyone knows, if you go into pediatrics, you don't understand economics. But I have the privilege of organizing this panel called Can Rational Consumers Improve Quality of Care and Restraint Cost? And we're going to begin with Peter Eubel, who is a former fellow here and now is the Director of the Rational Decision Making up at the University of Michigan. And Peter will be talking about his new book called Free Market Madness, a look at Rational Healthcare Decision Making. Peter? The average lifetime risk of breast cancer for a woman is about 13%. But if you ask women to guess their risk, they'll typically guess a number way higher than 13%. In fact, when Karen Lerman tried to educate women about their risk of breast cancer, she found she asked them what they thought their risk was, then gave them more information about it, retested them, found that they had accurate knowledge about their risk. She also found out that many of them lost interest in mammography. Which, if you think mammography is a good thing, would just suggest you should not educate people, because it doesn't help, right? But actually what I thought was going on with Karen's study wasn't that she educated people, it's that she had them guess their risk before she educated them. And let me just show you what I mean by that. We conduct this study where we ask women to estimate their lifetime chance of breast cancer. The mean estimate of the women that we surveyed was 41%, a lot, a lot wrong there. A lot of people put out 50%, which to us in this room means 50 out of 100, but to many people in the general public means, I don't know, it could go either way. Flip of the coin. Nevertheless, they put these numbers down on the page. For half the women we surveyed, we did not ask them to make this guess, we just told them the 13% number. And for both groups we then asked them what they thought about this number, how it made them feel about the risks. Now among the guessers, remember about 95% of people who guessed put a number in that was at least 5 percentage points too high. But when we asked the people who didn't guess, and we just gave them the 13% number, we said, if we had asked you to guess, would you have guessed lower, higher, or about the same? And on average, they think they would have guessed about the same. So that's called hindsight bias. When you give people the answer, the answer is a little bit more obvious. It also means that that 41% figure wasn't already in their heads. So why does any of this matter? But when we asked the guessers how that 13% risk made them feel, they felt relieved. I mean they just put the number 40 down on the page, or 50 down on the page, and now they heard it was only 13. But when we asked the non-guessers what they thought of that 13, that same 13% number, they were anxious about it. The guessers thought it was a low risk, and the non-guessers said that's a high risk. So 13% was a lot more of an unlucky number to the people who hadn't been asked to make some guess, kind of pull a number out of thin air. And it shows a little bit about the subtle things that can influence the way a risk might feel and the decisions you might make with that information in mind. A subtle thing like before I educate a patient, I ask her or he to make a guess about something. They might end up responding very differently to the information I give them and make a very different decision. Same information, same freedom to ponder that information, to decide what to do about the risk, and yet very different judgments. So freedom is a very good thing, I'm a big fan of it. I'm glad I grew up in the USA rather than the USSR. Freedom is also good as a means, however, for us to maximize our well-being. We can choose things that fit our own conception of good life. So that's why autonomy has been such a big deal in clinical ethics, right? As we recognize that one size doesn't fit all, the different patients have different preferences. But what I'm going to kind of talk about here is that sometimes the preferences we have are very unformed and easily changed by subtle things that make us question whether people are actually making decisions based on true preferences. And when they make choices that seem to harm them, we can't just always assume that they knew that that was what was in their best interests. And I think ultimately we need to balance freedom, autonomy with well-being when we think they are at odds with each other. Let me give you a couple examples of this. This is a study actually presented here a long time ago, but my slides broke down that day, so I can actually show it and you'll know what I'm talking about. We asked people to imagine that they had colon cancer, that there were two possible treatments for this colon cancer. Surgery A, I came up with that name. 80% chance of curing them without complications, a 16% chance that they would die of the colon cancer. A 1% chance they'd be cured, but they'd have a permanent colostomy. Or cured but have intermittent bowel obstruction. Or cured with a wound infection that takes a year to heal. Or now suffer from chronic diarrhea that might get them up out of bed once a week to raise to the bathroom. So that was surgery A. Surgery B a little simpler. 80% chance of cure without complications and a 20% chance of dying of the disease. So given this much time, this much information, how many people here would choose surgery A? And how many would choose B? So most of you chose A, somebody who chose A, why did you choose A? Less chance of dying. Why B then? So I'm going to choose B. A lot of side effects. So really this is a choice between give me colostomy or give me death. And somebody might have a preference that's different than someone else, so different choices might be better for different people. Well it turns out we asked the general public to make the choice between being dead or having a permanent colostomy, being dead or having chronic diarrhea. And we found that 90 plus percent of people said each of those complications was better than death. And so based on that, less than 10% of people should choose the treatment with the higher death rate, which is about what looked like happened in this room. However, the page after asking them to choose between death and colostomy, we then gave them that choice of which treatment to choose. And what we found was that the majority of people chose the treatment with the higher death rate. So they seem to tell us what their preferences were on page one, and they seem to make a choice that indicated very different preferences on page two. I think that, you know, I've done a bunch of studies to follow up on this result, and I can tell you what's really going on from what best I can tell is people know what the best choice is, but it feels awful to make that choice because the side effects are kind of disgusting, and they're icky, and they are just, it's emotionally, they're emotionally aversive. And so they end up often making a choice that they can almost feel their hand heading over to the other side of that page. This is how they'll describe it to me. If I understand a decision I'm free to choose, I should be able to pick the choice that maximizes my well-being according to my view, my preferences. I don't think that's always what happens. I'll give you just another example of just the subtlety of how things that can influence choice. Imagine you have, this is really a fun talk. Now imagine you have breast cancer, we've gone from prostate to colon to breast. Imagine you have a 6% risk of breast cancer in the next five years. That risk might make you worried, anxious, and it might make you think about doing something about the risk. In fact, you might want to take a pill like tamoxifen or reloxifine, you can cut that risk in half from 6% to 3%, and even reduce the risk of bone fractures. But at a cost, not just a financial cost, although there is a financial cost, of course, also a cost in terms of side effects. So this might increase your risk of endometrial cancer, cardiovascular events, cataracts, menopausal symptoms. So this is a choice that many women face, and we've been trying to develop decision aids to inform them about these choice and involve them in the decision so they can make a preference-sensitive decision, you know, on their own or with help. But what we wondered was, how do we give them all this information in a way they understand and that won't bias choice? So here's an example. We actually find in our research, these little pictographs help people a lot. Most people have a real hard time. I think half of people have a very difficult time understanding probabilities, and the other two thirds can't add simple fractions. So we often use pictures like this to try to help illustrate what these numbers are so people grasp them better. And so in this case, all I want you to notice is that we have 100 boxes out here. So instead of just telling them a percent, we show how many out of 100. But what we did in one study was we changed whether we showed people 100 boxes to illustrate the risks or whether we showed them a thousand. The same information, right? Just pictured a little differently. We also, in that same study, so that's where I say whether the denominator of the risk is out of 100 or out of a thousand. The other thing we did was we showed some people the low probability side effects first and that ended with the last one being the higher probability but milder side effect but more of those boxes colored in. And the other people saw the same information about side effects but in the opposite order. So everyone ends up getting basically the same information at the end of the day. But they don't end up feeling the same about that information. If you hear about a risk out of a hundred, you're less worried than if you hear about a risk that's out of a thousand because 13 out of 100 might be bad, but 130 out of a thousand, that's 130 people. And so people actually, that scares them more. If you first hear about the low probability events, the last thing you see is the high probability ones and what you remember in your brain there is you saw a lot of boxes at the end. You end up more worried in general about the side effects of the drug than if you had the exact same information in a different order. So it's really hard for people to use certain types of information to make important decisions in their life. They can be influenced by very subtle contextual factors. I could spend an hour just showing you all kinds of studies, the tiny little things I can change around that will have dramatic impact on people's decisions and judgments. That emotions play a big role when people are making decisions and even in these surveys I can trigger emotions by describing a health state. Now imagine you're in talking to a doctor who's just told you you have cancer. Think how those emotions might be influencing the decisions you make. That seemingly neutral ways of giving information, I'm telling you about the risks and benefits, right? That's a neutral thing. Oh, but if I give you the benefits before the risks, you would have made a different decision. So it's really hard to find neutral ways to help people make decisions. So I've told you a little bit about some complicated medical decisions that people make, rare, unfamiliar, scary situations. I'm going to tell you a little bit about decisions people make every day and this is going to be decisions about eating and I'm going to use decisions in very large quotation marks. Richard Posner co-authored a paper where he described the word overweight as a misnomer because we all know what makes us fat or not fat. It's how many calories you consume and how many you burn. It's not a very difficult thing to understand. And so people must be making choices in their life about how much to exercise, how much to eat that lead them to whatever their body weight is and so that's not overweight or underweight, it's the appropriate weight because they've made the choices. I actually think the word overweight is a nomer. I actually think that there are many people who don't consciously choose to be overweight and I want to show you just the subtle things that can influence how many calories we eat and what specific foods we eat. So if you went out to dinner last night after the tremendous conference, maybe you went out with a friend and had a dinner together with that friend or maybe you went out in a larger group. In making that decision were you thinking about how much food you wanted to eat? Probably not. But by being in a larger group of people you're likely on average to probably put away another 20% of calories. Why is that? Well one is when adults eat that way they stay at the table longer, there's more conversation going and they have just more time to eat. Another is it just energizes the table and when you're more energized you tend to clean up the plate more. You don't even have to stay longer though to eat more. They did a randomized trial in a daycare center where they went to snack time and in some rooms in this daycare center they put two kids for table and they put so many crackers out per kid, more crackers than the kids would put away in the little time in a break. And in other rooms they'd have eight kids to the table. I think it's one of the ones where you slip them in the chair and they can't even get out, right? But they put them, you know, eight to a table, same number of crackers per kid out in front, same amount of time to eat. What they find is that all the kids finish eating very quickly but the ones who are at the group table of eight, like 20% more calories. Just from being around more people. And those kids weren't consciously choosing to eat more, right? You know, we are a lot like kids in that way. So how much you eat? Pretty subtle. In fact, if you come to my house sometime in Arbor for dinner, I could put a buffet out there for you in my dining room and depending on what size dinner plate I put in front, I will have a big impact on how many calories you eat. Because, you know, a plate looks kind of empty when the plate on the right looks kind of empty. You're going to probably just put more food on there just so it looks like what a plate ought to look like. And then when you're eating, you will need to empty a plate a certain amount because it looks weird not to and there are also social norms, of course, about eating and mom's voice in your head, cleaning your plate. I can influence how many calories you eat when you come to my house and you won't have consciously decided to eat more calories. So are we deciding how much we want to eat every day? Absolutely not. In fact, Brian Wansink has done a whole bunch of experiments on this and one of my favorites, it took a month to figure out how to do this experiment. He rigged up soup bowls with tubes that went under the table to the bottom of the bowl and would keep refilling the bowl up with soup as people ate. There were some early incidents of exploding bowls of soup but he eventually worked out the mechanics of this and it was hidden and he had the story for why the bowl had to stay on the table and everything and he randomized it so half the people had a regular soup bowl and half had this refilling kind and the amount of soup different. I mean, there's one guy who put away a quart of soup. Totally oblivious to the fact that the soup bowl was just as full as it was when he began. Our eyes are a big part, you know, not just our brains telling us how much to eat. Our eyes inform our stomachs quite a bit in ways that we're unaware of and so therefore we might not eat how much we want to eat. Eyes are big in terms of when you have food in front of you, if you have M&Ms in a glass bowl with a top on it even, on your desk, you're going to be more likely to eat those M&Ms than if you have a ceramic bowl with a top on it because you just see it more and so you come up and you just put it in. It's mindless munching. We're not consciously deciding how many M&Ms we want to eat in a day, but little things can totally influence how much we do. So I've given a lot of information about how much we eat, but there's also how food tastes to us can be subtly influenced by really strange things. So there was a business professor in Texas whose family is from India. He had a bunch of people over to his house for a party and he served mango lassies and a lot of the people there didn't know what a mango lassie was and so he told some of them, oh, it's a healthy fruit drink from India. Yeah, they thought it was okay. He told some other people, you know, it's an unhealthy fruit drink from India and they loved it. He did another study where he gave people crackers. Now this is a marketing test, so he wanted to have them give feedback on the taste of these crackers. He told them this cracker has 11 grams of fat, nine healthy, two unhealthy and people thought it was kind of dry, you know. Same cracker, different people. Now it's nine grams of unhealthy fat, two grams of healthy fat and it was moist and delicious. So we actually, our expectations will influence how food actually tastes to us and so did we consciously decide to be that, screwed up about how we, you know, what we believed about food, I don't think so. And so it's hard for me to think that we are all making these rational choices that lead to what is a tremendous obesity epidemic in this country which has major consequences, right. So I think that we can't just stand by and let the free market work everything out when so many subtle things, including companies that have powerful marketing departments that understand our brains better than we do, are influencing our behavior. Let me give an example of why I think we also can't just expect to inform people out of these kinds of troubles. New York City has calorie count laws now for many of the restaurants. Let's suppose that you're going to eat and you've got these five sodas to choose from and you've got the calorie count above them. Which soda are you going to order? I'm kind of, I'm a middle child. I'm politically neither conservative nor liberal. I'm a flaming moderate. So I'm probably going to choose the middle one, right. Well, if you happen to eliminate these two options and just show me these three, I'll choose that one. So in fact, you start putting calorie counts out. If you put a couple really incredibly high calorie meals onto the restaurant, onto the menu, it'll make everything else look good by comparison. And that's in part how people are going to judge whether it's a healthy meal or not a healthy meal. Not like people know how many calories they ought to consume in a day. So I think we need to do more than just inform people. I think sometimes we need to persuade people. I think that the poison label works in part because that image is powerful. It's emotionally provocative. Maybe we need to think about ways of labeling food that will help influence judgments. We can keep people with some amount of freedom to choose what they want to do, but try to use taxes to influence what they do. And of course, that's had a big impact on cigarette smoking. The cigarette tax has a big effect on it. People can still smoke. They're not outlawed from smoking, but it's much less entry into the smoking market among teens when the cost of entry is higher. So these are the kinds of things that are tweaking our liberties while potentially improving well-being of the population. I practice medicine in the VA, and a lot of my patients have a hard time exercising. They don't live in safe neighborhoods, or maybe they have bad arthritis. It's hard to do walking exercise. I try to see if they can go to a pool and they can't afford going to a swimming pool. $4 a day doc, I can't do it. Maybe we can try to think of ways to make it easier for people like that to keep active and it would be a good thing. And wrapping up, right, Mark? That was that signal-wise. I think that when you go into a building right now, I have a hard time finding stairs half the time when I walk into buildings. I want to walk up three flights and I can't find it. Elevators are easy to find. You can't find the stairs. Or when you do find the stairs, the door's locked when you get up to the third floor, right? So that actually adds to my exercise. So maybe that's a good idea, right? Is it a bad idea to think that when we're putting up new buildings, we should have some regulations in there that people will walk up stairs and that will influence that behavior without forcing them to walk up the stairs. Maybe that's why I think a lot of good health is going to be about urban planning and how to have environments where it's easy for people to get out and move around. I think we must continually try to balance freedom and well-being. Our choice is not between autonomy and coercion. We must look at other ways of trying to influence people's behavior to improve their lives. Thanks very much. Thank you, Peter. It's scheduled. We're going to hear three commentaries on this, and the first is going to be from Kevin Murphy, who's a professor here at the University of Chicago, who's going to say, talk about, quote, an economist's look at rational health care decision making. Beth and Murphy? A little bit about the presentation in the book that we heard about today. And then I'll talk a little bit about some of my own thoughts and ideas. First, I think the book has somewhat of an odd title because it has very little to do with markets and has a lot to do with individuals, and I realize there's an interaction between the two. But there's a lot more to markets than just individual decision making. In particular, one has to think about the supply side. One has to think about competition and the role competition plays in shaping people's decisions. And in particular, the fact is that we rely on more the fact is that we rely on markets to substitute for our ability to know things quite frequently. I shop at the supermarket and rely on the supermarket to be a pretty good at choosing the kind of products that I would like to have and new products and evaluating those. I don't know much about those things. I don't know how much they should cost. I don't know on my own what those things should be worth. Competition and supply plays a very big role. Also, I think the book is very presumptuous in a lot of places. For example, it discusses savings and the tremendous and the fact that we've had success in inducing people to save more. I agree with what's said today. You can manipulate people and you can cause them to make different decisions by framing things in different ways. For example, making whether you save or not an opt-in or opt-out choice certainly does matter and the type of framing decisions that were made today do matter, but I'm not sure it's actually true that we're going in the right direction. I don't think there's any economic evidence that's 20-year-olds are saving too little for retirement. In fact, recent work on retirement consumption really has cast doubt on much of the research that people had used previously and one colleague of mine over in the business school has shown when you look at the composition of people's consumption, it's true consumption goes down on retirement, but that's mostly because of the fact that they stop commuting to work, stop buying clothes, consumption of things like cruises and golf and other things actually rise which is hard to believe that's a result of people having saved too little. I think even the presumption there. Same on cigarettes. There's a lot about cigarettes. You could try to show that people don't put enough weight on cigarette health information that the decline in smoking we see when people learn smoking is bad is too small. We have good estimates of what we think the value of life is from other experiments and other decisions that people make. We know how people respond to taxes. You could try to show, in fact, that people did respond too little or too much to cigarette information, but he doesn't do that. He just presumes that people should be smoking less than they are. I don't see how one comes up with that or points out and it's the same as the savings information. Secondly, most of the decision variables that he talked about today would exist in a world with no markets, in a world in which there was a single provider who told you these are the healthcare choices you have and whether it was a free market, no market, whatever, the individual frailties would still be there. Also, what are the alternatives? I mean, as economists, one of the things we always focus on is you have to have a choice. You have to say what's the alternative to the system? For example, there's another area we could think about substituting for individual choice in state education. We've decided that we don't want to see poor people choose where they go to school, so we'll provide schools for them. Are they particularly successful? We think that they get very good deal out of that? I don't think so. So you need to think about and ask the question, do poor decision makers fare better under markets or under alternative systems? Are you going to be more successful in a political environment than you are in a market environment? I think in many ways the answer is unlikely. Now, I want to turn again to rationality because most of what we talked about today had nothing to do with rationality. The other thing there are the kind of things like choosing the middle size. Amir, who's going to talk later, has done a lot of work in that area and really has a theory that really tries to explain why people make that. If you really want to tackle this problem, we should have a theory of rationality. Why are people as rational or not as rational as they should be? A good example is what does it cost you to be off by 10% in the position of a product? You know, so you know kind of on average how much it's worth but plus or minus 10%. How much should you be willing to pay to become perfectly informed? The answer turns out to be roughly about one half of one percent for something that depends on the elasticity of demand but something like one half of one percent was what you'd be willing to pay to eliminate that 10% error. It's a very small amount that's willing to pay. Small errors in judgment have relatively small consequences. The other side is, you have to remember rationality is endogenous. People choose how rational to be depending on the incentives that they have and the more you do to make choices for them, the rest rational they're going to have an incentive to be and one of the things we know is that rationality is an element of human capital and human capital is complementary between the two domains. The more decisions I have to make that require me to be rational the better decision maker I'm going to make on other margins. As you reduce my incentive or need to make decisions on one margin you'll reduce my incentive to make choices on others. An interesting thing today was there was a discussion of hindsight bias. Turns out there's an enormous hindsight bias research on decision making. We look at decisions where people make we know they made poor decisions or people are poorly informed and don't make good decisions but how do they know things about which they're poorly informed? You can't propose a solution that they can't implement ex ante. Good example is we people look back at the mortgage crisis and say well geez, boy people should have read those contracts. Well people sign lots of contracts in their life. You can't look after the fact and say that's the one you should have read, right? Nobody reads any of them. The only policy you have that you can follow is read all of them. Well that's a pretty costly policy. There's an enormous hindsight bias in all of this. You look back and say wow people don't know anything about this. Well they don't know anything about almost everything. And to say they should have invested, they should have learned more about the probability of breast cancer. Well I don't know if they should have. They're everything they buy in the supermarket they don't know about. So you have to have a policy that somebody can actually implement. You have to say okay let me take a decision they made at random and let me see if they would have paid to learn about that. You can't look at particular decisions. It's kind of ironic. That's the own form of hindsight bias. I don't know what you're going to understand that. The lifestyle changed. Talked about obesity. Bob Topel and I did a lot of work on trying to figure out how much reduction in death rates are worth. So I went back and looked at our work and asked the question how much would it be worth to get a 10% reduction in both your lifetime, 10% reduction in your annual cancer death rate and a 10% reduction in your life. And in order to get that you'd have to change your behavior every day of your life. So on a daily basis how much was that worth? Turned out it was $1.34 a day. So if you enjoyed eating a cheeseburger or whatever the hell else it was that made you reduce your health by $1.34 or more a day you're better off. Better off doing it. Completely forward looking. Nothing else. Now we have proposed solutions. So let's think about a tax. So should we put a tax on people to make them make better decisions? Well I have a problem. I can't hit a golf ball straight. So one of things you could do is tax me for hitting a crooked. And if you tax me for hitting a crooked I will learn how to hit it a little straighter than I would otherwise. But that's not clear that makes me any better off than I was before. Decision making is costly. Rationality is a choice. You have to decide how rational you're going to be. You're going to have to decide how much you're going to invest in being rational. You have to ask what if you had such a tax system what would it actually look like in practice? You can't presume that the government's going to be omniscient and generate a tax system that's going to maximize people's preferences. Why in the world will the government do that? Why would that be in their interest? What model of government would give you that as a choice system? In fact every model of government I know would fly that they wouldn't want to do that. They don't have those incentives. Why would that be what they want to do? Now the question is can we use markets? And I think one of the things about markets that really you have to think about and this is completely missing from the discussion is that I benefit from being in a market because other people are also in that market. So for example I don't shop much between jewels and dominates. I really don't know what the prices are. I go to jewels as close to my house. That's where I go. I used to go to dominates when they were closest to my house. So I guess I'm like him. I like his critique. I just do whatever is easy. I don't make conscious decisions. But I depend on the other people who do. The other people who do. The little lady who looks in every you know looks in the paper every week and ties it aside who's cheaper. That's what I rely on. She helps me. One of the problems we have in the healthcare market right now is we don't get that benefit. Insurance companies go out and negotiate deals but they don't negotiate deals for me. They negotiate deals for themselves. They don't help me get a lower price. In fact what they do is negotiate a discount off the list price. So if I'm a guy playing the list price if anything my list price probably going to go up as a way of reneging on the deal with the insurance company. Sure I'll give you 50% off. I'll give you 50% off a higher price. So the market we have doesn't really help in terms of the external benefits to to shopping. The same as the problem we have with employer provided care. I think the tax incentive for employer provided care really inhibits the growth of a more market driven solution for healthcare. I'd like to see us move away from the employer provided care system. I think that would really help us generate a greater market. So what's the bottom line? I agree 100% that people are are fallible. People don't always make good decisions. They often I can manipulate their decisions. I like to think of that as advertising works or marketing works. There's no question that marketing works. But I don't I don't think that tells us that people are irrational in any sense or that people aren't doing as well as they can. I think the specific examples of savings and cigarettes and the others are not compelling at least the evidence that was presented in the book. I didn't find it compelling at all in that direction. I think we have to I think if you want to go down this road let's try to build a theory of how rational people ought to be. And I think what I've learned from economics is that it doesn't pay to be hyper rational. That hyper rationality is really very costly. That I could be a little better decision maker. I could read more contracts. I could shop at Jules and at Dominix. I could I could try to prevent myself from being influenced by framing and other things. But that would be counterproductive. Why do people rely? I get another example from the book. People rely on price as a signal of quality. Well why do people do it? Because it works in equilibrium. That is on average in equilibrium it does work. So I put them in the lab and I give them a situation where it doesn't work and I say look they're fools. Well that's fine. But you know that doesn't mean it wasn't a good idea. Same is true with the contracts. I don't sign contracts. I don't read contracts or I do sign them and I don't read them. Interesting today when we came in to sign up there was a consent form for videotaping the conference. It's interesting. They didn't ask is it okay if we videotape? They say could you sign this form so we can videotape the conference? It was an interesting frame I thought because it sort of it sort of implied what the default was. You're supposed to say yes. Right? I mean so that was a good example of that. And that works. And I'm not going to deny that. Although I don't see how that quite gets us to where we end up in the book. Those are my comments and we'll surely have more discussion. Great. So I have the privilege of introducing myself. I'm Laney Ross, one of the associate directors of the McLean Center. And the title of my talk is How Would a Rational Parent Make Healthcare Decision? And my one confession is that I wrote the title before I read the book. I understood from reading the book. Premise one, when people operate their unconscious behaviors too often cause them to act against their own best interests. Premise two, knowledge doesn't always lead to quality decisions. Even when people know what the risks are they may still base their decisions on how the risk feels. And so the conclusion is if we only look for the logic underlying people's behavior we will miss the illogical forces that influence people's emotional reactions to risk. So I thought about that. And here's what I realized. How would a rational parent make healthcare decisions? They would make decisions for their children almost as poorly as they make for themselves. There would be all these mis-predictions of future happiness. And that was the end of my talk. So I decided I would write a different talk. And what I'm going to talk about today is the meaning of Ubal's free-market madness for pediatricians. And one of the things I'm going to say is even if pediatricians don't understand economics we're actually ahead of the curve after reading Peter's book. So I changed the premises a little bit. The first remains the same. When people operate their unconscious behaviors too often cause them to act against their own best interests. And then the second premise is that the limits of human rationality of implications for the proper limits of free markets. And so what Peter then works on is what markets should look like when they're designed to take account of human nature. So I wanted to look at a couple of problems and solutions that Peter looked at in his book. His first was make the default option the preferred choice. And he said for example assume that you'll put money into a retirement account or in healthcare there's a whole debate about whether we should assume you'll be an organ donor whether you need to opt in or opt out. His second example is looking at studies of social networking which shows that obesity, smoking, drugs are contagious no longer just H1N1 and this gets in pediatrics at least to the importance of peers and we all know how much influence children have on each other. And his third problem and solution is the limits of self-control and he talked about maybe we should have for example weekly paychecks and food stamps rather than monthly. And in other words what he called quote soft paternalism. Now pediatricians have always been aware of free market madness. Our default option for example we tell parents that immunization is a mandatory. The mandatory is you can't go to school. Well it turns out that virtually every state has an opt out system. You just have to know about it and what you'll find is that your pediatrician doesn't know about it but there are religious and philosophical exemptions for both immunization. We also do newborn screening is mandatory and in fact here we may go overboard because not only is it mandatory we actually don't even tell parents we're doing it and so in order to opt out you actually have to know that it's going to be done and that you have the right to say no. Now again when Peter talks about the importance of social networking pediatricians have always been conscious of this. We know that we need to engage adolescents in productive after school activities because otherwise if we leave them to their own influences sex drugs and rock and roll. We also know that there's the need for parents to ensure that their child is quote hanging out with a good crowd. We also know that the friends of fat children are often fat so the social networking fatness really is contagious. Of course to accomplish that we need to promote healthier choices in school cafeterias despite what Reagan said ketchup is not a vegetable. With the limits of self-control it was also interesting could pediatricians are there. We need to balance promoting the child autonomy or developing autonomy often thought about their developing independence and setting limits. So children need to be taught these skills of self-control and to practice the skills and my favorite example actually isn't about adolescents it's about young children and how you put them to sleep at night and many parents when they have kids who they can't put to sleep the first thing they do is give them a bottle and that's clearly not the right answer because the fact is we're all going to wake up several times a night and if the only way we know how to self-com is by having a bottle both the tooth decay and our obesity epidemic started here. So Peter also talked about the fact that we have limits on self-control and he talked about how it can be really hard at the end of a really stressful day or a week in order to be able to have self-control and so that's one of the reasons why pediatricians we espouse the benefit of a two-parent home I'm not here making any comments about whether it's a married couple a heterosexual couple but the two parents often are better than one that not only do children need time out but so too do parents and all other adults so what does all this mean for pediatrics because it seems like we already know all of the things that Peter's trying to tell us to do in this book well it's interesting because the traditional conception of pediatrics is that parents are supposed to act in their child's best interest but we actually don't hold parents to a best interest in part because we don't know what's really best and in part because even if we knew what was best once you have two children what's best for one child might not be best for the second and so there needs to be some degree of compromise so actually we don't intervene in families until parents are actually abusive or neglectful and this hands-off approach maybe too libertarian for you both given the free market madness environment and so one thing you could argue is that we should have greater state input to ensure that parents make better decisions or another way to think about it is what role can the pediatrician play to help parents make better decisions and then to follow through so one of the things that this made me think about was that Peter was talking at the policy level and I want to take us back into the clinics and talk about the level of the doctor-patient relationship and so there's this new movement called motivational interviewing which is a book recommendation on the bottom and what this whole theory is and I actually should thank one of my students in the audience, Jen Walters who pointed out this book to me how to get patients or parents to do what they ought to do which is really what they want to do but they just don't realize it told you if you want to be a paternalist you're blowing a pediatrics well they don't realize it unless you help them discover that that's what they really want and so this methodology which was used by psychologists for mental health counseling especially addiction but has been tweaked for use by other healthcare professionals and it's actually been evidence based has shown that this methodology actually works and leads to better healthcare and so what's the methodology it's actually quite simple there are four principles one is to resist the right reflex don't go into the doctor-patient relationship and correct or argue with parents about everything that they're doing with or the patient's own motivations for change in other words what you need to get to it is the patient's own reasons for change and not your reasons as a physician that are more likely to trigger behavior change so you have to listen with empathy active listening help patients parents and our teens come to realize that they already know and hold the answer to change we all know how to lose weight take in less calories or exercise more right but we have to empower the patient and part of that is exploring with the family how they can take an active role in their health and how they can bring their own expertise they can't afford the $4 pool a day they may have other ways where they realize can they walk around the mall early in the morning the museum open up early enough for people to get exercise and things of that sort so working with families to make them think that it was their idea to do this change rather than us telling them they'll get the best out of the West so Peter's free market madness defines the problem and offers solutions at the societal level and as he writes if we care about people's well-being we should pay attention to how people behave and take measures to regulate markets in ways that promote people's health happiness and social functioning I think we can think about it at the clinical practice or at the occupational relation level that much of health care today involves outcomes can be greatly influenced by lifestyle or behavior change motivational interviewing is often effective in evoking behavior change when education and exhortation have failed so my last thought is both approaches and I would as a pediatrician actually say both types of advocacy are needed thank you. Our last speaker in this panel is Professor Meir Kamenica who is in the economic section of the GSB and he will be talking about economics and psychology of health. I also picked the title before I read the book so the title you can ignore so I'm going to begin pretty much the same way as the last talk and just tell you what my reading of the book is and then after that I want to focus on one particular part of it namely to what extent can we account for the rise in obesity by thinking about malability of choice and lastly talk briefly about market based approach to health care. The first part is what I read to be the argument of the book basically the first input is that there are decision making anomalies so bounded rationality, malability of choice, those are decision making anomalies and those coupled with a theorem which is kind of implicit in the book and that consumer rationality implies that free markets are efficient and vice versa if we have these two inputs that people are boundedly rational and the consumer rationality is equivalent to efficiency of free markets then we reach the conclusion that free markets are inefficient or in the language of the book they're mad. So this is what I read to be the basic structure of the argument in the book and what I want to focus on is this middle part to what extent does consumer rationality imply that free markets are efficient and to what extent does the bounded rationality imply that free markets are not the best way to go. It's important to realize these are two entirely independent claims. There's a claim that if consumers are rational then free markets work well and there's a separate claim which is the more important one for the purposes of the book that if people are bounded rational then free markets are going to perform poorly presumably relative to some alternative. What most existing literature in economics is focused on is the argument of the first kind and I just want to question whether in fact those arguments necessarily imply that we should conclude that just because people fail in their decision making at times free markets are not the best way to go. So here I'm going to echo Kevin a little bit and point out that most of the data in the book is on decision making rather than market failure and what's implicit and I think the greatest thing to put more thought into is to what extent does in fact failure of decision making imply that market's going to perform poorly. If you think about a really simple decision and Kevin and I didn't collude to talk about local convenience stores but it's just an easy market to think about. Think about just how much milk the guy that runs my local convenience store decides to buy every week and how much milk the consumers decide to buy. Now it turns out if you write down the mathematics of this problem they're really hard. It's a stochastic dynamic programming problem given that the demand goes up and down and milk has an expiration date and the probability of purchase depends on this. I assure you if I wrote down the equation that describes the problem and gave it to the guy that runs my local convenience store he would do less well than when you ask him to compute 10% of a thousand. Okay? The consumers also have a problem that's really hard. Okay? How much milk to buy? Well, you know presumably if you're in steady state and you eat certain amount of milk you drink certain amount of milk every week it's easy but you might have visitors and so on. It's a hard problem and things work out fine. I really don't think that we need any government intervention or regulation about how much milk should be in the convenience store. So we really need to ask ourselves even when the problem is hard and people can't solve it does that mean that markets are necessarily going to perform poorly? And there's a sense in which markets are as magic as they are. There's two people who, neither of whom is associated with free markets Danny Kahneman and Dick Taylor devised a game for a long time ago for a bit of their research they eventually abandoned were people who were not experts in deciding whether to enter or not into markets had to play this little game where there was 15 of them and each had to decide to either enter or not and there was some capacity of the market which was less than 15 so if too many entered they would all lose money if sufficiently few entered they would make money and it's a bit of a conundrum because if they all enter they all lose money if none enter they're forgoing profitable opportunities so what they did is they played this game and they varied the capacity of the market and they played this very simply these were just their students and they would raise in the car to enter or not and they played this many times very in the capacity and shockingly at least to Danny and Dick just the right number of people enter each and every time and Danny Kahneman wrote to a psychologist this looked like magic because there was no obvious reason to see why this would work out in that particular time it did and it's not because I'm sure I teach MBAs I can tell you would not be able to solve for the equilibrium of the game but with this game it works out fine okay and this is not something so this is a 25 year old game but this kind of research continues there's a paper which was just written a few months ago by really needily and and a co-author where they have people completely inexperienced subject played this quite complicated matching game where people get particular value from being matched with one kind of a person different value from being matched with somebody else and this is true for everybody and it turns out every time they run this game on first try it converges to something which you can expose once it's done check is a stable outcome meaning there's no pair that would be better off with each other in whom they're with and to this date after they've demonstrated this experimentally they cannot come up with an algorithm that can find that interior solution okay there's extreme I'm getting the technicality of this irrelevant there's some extreme solutions that are easy to find that's not where people end up they end up in a more equitable stable match we do not yet know how to write down a computer algorithm or set of equation that can help us find it but you put people in this environment and they end up there okay so markets can have this power to lead us to outcomes we like even when we don't understand the process through which they do that now I would like to go back to the first part that if consumers are rational markets are efficient okay this is I've now talked about why the converse is not necessarily the case this is actually a rather incomplete statement or even you know of the view that people in the economics department hold this statement by itself is not true if consumers are rational and we have very well defined property rights with no externalities we can write complete contracts there's no adverse selection there's no more hazard which pretty much never holds okay then free markets work perfectly fine right so we know that but this is even assuming perfect rationality we still need all these other ingredients alright so applying this to health care even if consumers are rational we still want to worry about well how much is adverse selection we know in insurance market you can get very bad equilibria even when everybody being rational so compulsory insurance can be a good idea we don't need to talk about malleability of choice to reach this conclusion they're clearly missing markets in insurance contracts there's lots of contracts which from an economist standpoint will be the best kind of insurance contracts to sign nobody's offering them right now we don't entirely understand why but I think clearly we want to we want to explore why I have to why my premiums can go up and I can't insure myself against that there's no insurance against increasing premiums I'd like to have those there's obviously moral hazard in doctor's choices this means that the market might fail we want to get through this that means government can do any better there's an issue of property rights and advertising I mean you rightly point out that people are susceptible to advertising but I think the becker-murphy framework for thinking about those things is the right one to point out why it can be a bad thing if we're involuntarily exposed to exposed to advertising and of course there's all kinds of asymmetric information certification there's reasons why we may want we may want the government to be involved in telling us which medicine actually works rather than let anybody sell whatever they'd like we may want to have certification of medical programs and so on the most important part I think of healthcare market is that we have social preferences I think this is what makes this market complicated and hard to think about from a pre-market perspective we're not willing to let poor people die when they can be treated very very easily okay and we're not willing to let charities and charity hospitals be the sole recourse we are simply unwilling to say there's an old person this one particular old person did not say for retirement no look we're just not going to treat you we have as a society seem to have a preference and I I share that preference for providing certain kinds of healthcare as a right okay we feel the same way about education we think we need to provide people with elementary school and high school education regardless of whether they're willing to pay for it and this makes markets complicated that doesn't mean that we should just therefore just say let government run everything but it does mean that now just the simple libertarian argument is not going to it's not going to go through because preferences are complicated okay so one thing to note is and this is just Kevin spent a lot of time talking about I'm not going to talk that much the markets fail government might also there is a there is another things to know which is when government fails markets might also do it the argument goes the other way as well so for instance I talked to some physician friends that told me about Medicare billing and told me then act and I'm really here less informed than pretty much everybody else in the room but I've been told that if you do two procedures on the same day the procedure you get reimbursed less than if you separate them now that means that if you want to replace two kneecaps or something it is in the interest financial interest of the of the health provider to actually split those and to try and tell you it's actually better if we just do one now that you recover and then do the other one now important thing is even if doctors are well meaning even if the system is well meaning we've learned that in fact we're influenced where our decisions are often influenced for reasons that we don't recognize so I'm willing to believe that nobody's going to consciously impose unnecessary separation procedures but if it's in our interest to do something we start finding ways to rationalize now the question is how would market solve this there's very we understand why negotiation between the government and the healthcare providers might lead to these inefficient pricing because we don't have market for that but if we have a for-profit insurance company doing it I do not I know of no theorem that tells us things are going to work out better so I think we really have to in every particular domain think about the alternatives and focus on the role of political economy we need to have a model of how the government is making decisions before we can decide whether we're going to do a better job or not okay let me briefly talk about one part of the book which you know I think was was when I thought a fair bit about which is to what extent this malleability of choice affect obesity so the fact there's malleability of choice is really uncontroversial claim we know that and we know that pretty much every setting we know that if you ask people would they like to pay the last two digits of their social security number for this bottle of wine okay take the last two digits of security number transform that into dollars here's a bottle and would you pay that much okay now let's bid for it if you have if your social security number ends on two high digits you're going to bid more even though obviously that's relevant now the reason for this there's some informational stories that make sense as to why this should be the case and those often explain also default effects and many marketing interventions the fact that choice is malleable is almost prerequisite for the existence of marketing campaigns my favorite marketing intervention is that you know there's this nice controlled experiment if you play german music in a grocery store you sell more bottles of german wine and french wine if you play french music you sell more french wine and german wine okay so choice is malleable now most of my work on this sometime ago has been precisely why this is a phenomenon which doesn't mean that we should question markets but rather is a market induced phenomenon there's ways in which markets provide information that makes this responsiveness to cues a good idea on average alright but let's think about the particular research which was in the book namely research is largely by brian once about how you could if you wanted to nudge people into obesity at least for a day okay so we saw the bottom of soup size of plates I'm not going to go through this again there isn't I think a crucial question which is missing from that research is what the long run effects of this are you can certainly get me to eat a lot at a particular sitting through these interventions can you get me to do so over the course of my lifetime I think we need an NSF grant that's quite large to run that experiment and we may now say okay we don't know that but let's look at the data on obesity here's what it looks like here's this is from the CDC and as you can see let's focus on the obese line that's the middle line is something happened in 1980 so we're much more obese now than we were in 1960 but it's not like it's a nice steady trend the obesity was basically constant from 1960 to 1980 it had a very very small increase and then it just exploded over the next 20 years I think this is important data to keep in mind because what would have happened you think it once had run his experiments in 1960 would he have gotten different results I don't think so choice has always been malleable it's always been easy to change our decisions and yet something changed so if we want to understand the rise in obesity we can't go to static factors to explain them so I think there's a question there as to what's happening and let me react also to the proposal in the book which is the desirability of taxes there is a subtle argument which is missing from the book which I've seen before about made by Matthew Rabin about why taxing goods if you think they're bad might be a good idea there's a sense in which it may be first order gain for those making a mistake and a second order loss for people who are optimizing but apart from that strengthening of the idea for the taxes I think the political economy of this is that if we start taxing potato chips we're going to lose the political capital to tax things that we really ought to tax like driving and pollution I mean if you ask me if I get to increase taxes a certain amount that's all that's politically feasible what should we tax I mean it seems really really easy for anybody that's looked at the numbers as to where the great externalities are however serious the problems of people eating too many potato chips if they are might be they really really pale in comparison with the terrible externalities that come from driving in various forms pollution, accidents, time there's just all the reasons why those markets work very poorly because property rights are ill-defined pollution property rights are also ill-defined and we're destroying the world much more quickly with these other problems and I think this brings me back to something which is kind of a repeated point which is we really have to think about a political economy of the situation before we can make concrete proposals lastly let me finish by telling you how I think what I think is the right way to think about health here if you're going to take a market-based approach first is to acknowledge what the complicating ingredients are so first the fact that we have these social preferences the reason why we can't just say forget about the government let just everybody buy whatever they want to consume we don't want that for whatever reason we do not want people who can't afford basic care to get them so we have to acknowledge we have to start by specifying our objective function so to speak this is a complicate mark because you get little or no feedback if you get a surgery you probably will never find out how good a job the surgeon did it's the kind of market where there's very little feedback so it's very hard to figure out if it's operating well there's highly asymmetric information vis-a-vis doctors and patients vis-a-vis customers and insurance companies and we know when there's asymmetric information you need to be careful about designing markets and I think we should acknowledge that decision-making is made under stress we want to keep that in mind as we think about structuring those decisions but it's not obvious how that affects the design of the markets per se so I think a good analogy is to think about at least the way I would solve a simple problem in Newtonian mechanics start with the standard assumptions we understand really well as economists start to figure out what the system should look like if there aren't any complicated ingredients and then go ingredient by ingredient and see what if any remedy there is for violation of the assumption so if we have these social preferences how could we fix that and how could we fix that and what is going to be the political economy of fixing that and is some other system going to do better than do better in the market I think this quickly leads to important questions which have been completely missing from the debate on healthcare reform and they pop up very very quickly if you start this way if you start by saying let me begin with the standard approach and try and figure out what does the sort of standard model get wrong you get questions like we talk a lot about the uninsured what are they under-consuming are they under-consuming healthcare or insurance I've never heard this question asked when we talk about the uninsured what are we worried about that people aren't getting enough health or that their payments are such that with a margin utility of income is high they have to pay a lot of money those are different problems they have different solutions until we know what is it we are worried about we can't possibly come up with a solution another question you ask it seems rather obvious that something like pregnancy should be covered by insurance nobody is bringing this up but it is quite often if not always a procedure which is expected the only way that economists know how to think about insurance is four things which are shocks so suppose that you have a woman who decides to become pregnant plans this in advance there is no economic argument for why she should be insured against this and yet it seems sort of if you have the current approach it is crazy of course you want to cover that it seems almost unseemly to think about taking away payment for pregnancies from insurance same thing for annual physicals those are things that you should expect vaccinations now maybe this isn't about insurance maybe we really just want to make sure everybody consumes this care but we are talking about it as if we are discussing insurance and I think we are discussing about what we should have government pay for instead of private people pay for but it is kind of individual pay for but we need to know what we are talking about before we can reach conclusions the obvious one is why would you ever want to have employer based health insurance I think that is not a new one and it becomes obvious once you take this approach that there is a real need for long term contracts and insurance so what we are talking about is there a way to make those more readily available that is all I have to say where to begin right there is a lot to talk real quickly I would say that markets are great well first off if I knew how few books I was going to sell I would fight harder for a different title for the book that is hindsight best the second thing is markets are a great way to help figure out how to match supply and demand and my book was really more about why are we demanding what we are demanding why are we deciding to buy as many Cheetos as we buy versus not buy them and so milk is a great example of what you go and buy in a grocery store but some of the other things people buy which they might actually be just as happy if they had never bought them and what marketing forces influence that to get people to make these decisions and then when they bought things did they understand what would influence how quickly they went through supply of chips I think given that as a clinician I see a lot of people struggling to try to lose weight and I believe them when they say they want to and I also see all the forces corralled against them I see that many of them have started becoming obese when they were in grade school and there's all kinds of market forces that have influenced that and I know the biology of the set points of once people gain a certain amount of weight it's very hard to lose weight and those things I guess are important as a society for us and figure out what to do about and I don't think we can expect the free market just markets on their own to control them now we have to balance them I think it's a great point against do we have more important things to do the externalities of pollution being a great example and I can't tell you how much political we have probably don't have enough will for either of them so we're going to be okay and the second thing I'd say is again going back to the hindsight bias Kevin you talked about with mortgages there were people of course who were worried quite worried about that before everything came crashing down and if you were going to pick one thing to be very worried making sure people are making wise decisions it might be the largest purchase that they make in their entire lives for most people that have the potential to bankrupt them and consequently have major economic impact on society as a whole that might be a place you'd look you probably wouldn't be looking at the imperfect decisions people make about how many gallons of milk to buy and certainly look at whether they understood the terms of their mortgage before they staked their future on hoping that that interest rate wouldn't go up and interest rate by the way that they don't even understand what it means so but anyway I'm very interested in hearing other people's questions and thoughts Dan? In terms of the approach I suggest I just don't see what the alternative is I don't know how to start thinking about the problem other than by enumerating the things which are different from a problem we understand well so there's a problem we understand well which is how things work when all of these complications are missing so I can at least give myself a path toward some toward if I want to think about the problem if I enumerate those those departures now I completely agree with you that this problem is very hard so that if I try and fix an anomaly in a way which is unwise or hasn't been tested we could get all kinds of problems but at least this method gives us a way to proceed forward so we can say okay this solution for this for let's say this particular moral hazard problem doesn't work because we paying for performance might be a really bad idea and we can still go back to the drawing board and still know what is the set of departures that we have to address and it's really meant to be a framework for thinking about things rather than a set of suggestions I don't know if that answers your question but I guess I would say there's a couple things one is I guess I'm not sure that what you're saying is what is so different for example you had those four points that you put up in in terms of what you took out of practice for pediatricians it seemed to me those flowed directly out of kind of the rational model that you know if you look at the Becker-Murphy advertising model basically says exactly what you were trying to say there if you're going to try to do it in a way that has the most complementarity with the action you're trying to take and do it in a way that has the least cost to the person you're providing it to I didn't really see it and I guess the same is here I viewed what Amir is saying in his prescription was what are your objectives what is the problem you're trying to tackle let's write down what those problems are and what is the system that was going to fit that you know for example paper performance we've known that there's big problems with paper performance not just in medical area we know that happens in education we know that happens on the assembly line we know that happens in corporate board rooms I don't see why what have we seen that tells us it's different in medicine there are problems with paper performance everywhere I guess I'm confused as to precisely is the thing you're saying that's so different about the medical problem and the other problems I'll take one back to Amir's question is are we trying to provide insurance or are we trying to provide care is this really an issue about whether we shoot a market to provide it or whether we should ensure that everybody has access to it or everybody ends up receiving a certain level of care those are not the same questions you could have a system in which you denied certain people care and it wasn't a market or you could have a market and provide everybody care it's a lot like the education issue about whether you should have government provision government funding those are different questions I think you always have to get the objectives right and your problems identified before you're going to come to any kind of rational decision about what the market is going to look like those terms are both defined too slimly to extend to the real much more complicated life of individuals I'm thinking of some very recent work by two young delorians in economics Marcia Sandman and Jesse Stiglitz where they argued that the typical way in on it model of rationality doesn't fit the complexity of people's real choices which in their view are still rational people argue that all of them only concern about their own health outcomes and yet 20 mid choices they have to do with well even concern here about long term environmental impact or what long term, far impact on other people those choices aren't necessarily either irrational or out of concern with our own wellness and so I guess I want to make a point that I think the definitions that I have offered about both rationality and wellness are just too slim to embrace the entire complexity of real people's real decisions please and it was a very good comment I think that was a very good comment and I don't think any of us gave thorough thick definitions of wellness and rationality but I agree you have to be careful about being too narrow in looking at either sorry I'm just going to I guess I would try to relate it to another thing that came up in the book which is an interesting question you brought up sort of the fear of flying and the fact that you have small probabilities and people but have an enormous amount of fear one question you should ask is should that be based on the actual amount of risk or the perceived amount of risk I think the right answer is it's got to be based on the perceived amount of risk you don't say well nobody's really going to die so why should we try to prevent it you got to say the angst that people feel is real I think that's a good example what let's say for example I'm afraid to fly because I'm afraid that I have a bomb on the plane and we can take actions to reduce our perceived risk let's say that in fact the probability is so small that it really shouldn't be very important to me but it's something that worries me I think we should in fact then invest in trying to to make those people feel better because that's a real gain that people get I think people really are really are anxious and so that should be based on their perception whether or not that perception in some abstract sense is rational that's what I'm saying I would sort of include that fear as in some sense counting the same as if it was based on some objective probabilities it doesn't really matter that it's subjective I wrote on a Cambodian airline and you know at the beginning of most flights when they go through all here's how the seatbelt works here's the mask and nobody watches to make their saying this again they didn't do that on that airline and they felt like oh no do they even care comment about what Mary just said because I'm Bob Taylor from Ohio State University it seems to me that what you were saying Eric is that that's precisely the reason markets work is because people don't have to have clear understanding of their own rationality their own reasons for doing things and the market accounts for that now that we don't understand is that wrong? I misunderstood you I think that I think markets may work even if people are unaware of how rational they are or not I think there's a sense in which they may we from a theoretical perspective we know that we often need at least some participants to have some sense of what's going on before we can get good outcomes but that's pretty much the only ingredients we often need in the sense of input of rationality it is far from necessary that everybody involved knows exactly what's going on before we get outcomes we like now there's some exceptions and I think you know depending on the nature of the market there is more or less self-correcting mechanisms like there's a market which you know you may not think of it as a market but one of the legacies of the University of Chicago's approach has been to expand the notion of the market which is like the marriage market if you don't know what you're doing and you end up marrying the wrong person it's going to make your life miserable there are few ways in which other market participants being informed is going to help you there it's not if you don't know what's good for you you're going to get a bad outcome and that doesn't change the fact that thinking about you know what exactly are the requirements of information requirements in markets that make them operate well is a question we can't address yeah I mean I guess one way to think about it is under what conditions does other people being better informed help me versus maybe even some cases hurt me and I use the example of where to shop and I don't comparison shop and I take advantage of all the people who do there are other markets where that doesn't help so much and I think that's one of the things to think about in structuring markets are there alternative ways of organizing markets in ways that we get more of that kind of positive feedback as opposed to negative feedback and also I think and I'll try to stress what I said earlier I think it's also important to realize that how rational people are is endogenous and if you say I'm going to look out for you people are going to take less precaution for themselves and you got to feedback that you have to feed that back in terms of how both individuals and markets operate and failure to do that could lead you to do some do some I think silly things but I just say go back to your slide on the obesity the rise of obesity since 1980 and it's a you can tell these great tales about how you don't have to be that rational for the markets do some things really well I totally agree but you can also have some very powerful parts of human nature Brian wants to stuff has been around a long time but now it's going to be a little bit more exciting with some other social forces maybe more effective marketing by food companies maybe and obviously distorted by subsidies for sugar which you know I understand but all these things are colliding to where I don't think people have settled on a better equilibrium for their overall well-being to be that to be overweight in the way they are and to be obese and to have the disease early disease that we're starting to see now diabetes coming in and people but is it the market or the people haven't worked that out well people haven't so but it's the people who are the demand side of it right they're deciding how much to buy and and so the market then helps figure out how to meet that demand and there's what's the evidence that food marketing for example is that anything to do with this as opposed to food getting cheaper and that's sitting in front of a computer rather than those going out part of the portion I agree I mean do we have any evidence that actually food marketing it's changed in a way that it explains the change over this period of time or people's inability to make good decisions has changed over time as opposed to just saying it's other background variables that have changed well I mean the cost of calories has gone down cost of calories has gone way down and the ease of eating has also gone down and it's easier to eat you don't have to prepare food as much right so there's not the cost of consumption has gone down so what does that mean we are necessarily rational in how much more we've consumed as a consequence and there I just say I'm not convinced so what's the evidence that you know and then also the health that the idea that the medical system is going to bail me out at some point if I'm overweight have ways of treating me and other things that also got to feed in what's the evidence that that people have gotten fatter than they should I'm just asking what's the barometer I know foods gotten cheaper so people should eat more exercise got more expensive so people should exercise or they should eat the same amount and save and now they have more time and more money to do other things how do we know what they should have done it's hard to know I agree but it's also I just say to then assume that therefore they've done the right amount is itself a huge leap of logic I'm not saying that I don't know there are people who have gone back to the stuff on savings because that's where I want I think for people for a long time we're convinced that people were saving too little and I think the more we understand that system it's not clear that that's true I think that that's the danger in this or you know cigarette smoking I mean I haven't seen the evidence that says that the decline in smoking due to rising the increased realization of health care cost is too small does anybody establish that that people responded too little to the health information relative to how they responded to taxes say I don't know I haven't seen anybody establish that the amount of information is that most people can absorb but I can say we've been audio taping encounters between urologists and men newly diagnosed with prostate cancer and listening to the audio tapes has been quite an experience they're told at minute one that they have cancer and then by minute one and a half they're given a very jargon filled technical discussion of how they arrived at their particular Gleason score I have to read these things several times to understand what the doctors communicating then they get a brief description of what's for waiting why it's probably not best option for them then a discussion of radiation treatment an acknowledgement that I'm a surgeon I'm biased in favor of surgery but here's what surgery can offer the decision is often made it is very far from the ideal that you're putting out here there's often not set up follow-up appointments for people to have time to think this through and ask questions they're not encouraging them to even go see radiation oncologists to figure out if that makes sense it's a real challenge to figure out how you can actually get someone to better participate in that decision I guess I'd like to respond just a little bit it's interesting the description you gave of the optimal way in which a physician should deal with a patient because it sounds very much like if you listen to a financial advisor talk about how he should work with his clients it's exactly the same description you gave and it's back to my point earlier that there's a lot of commonality here of what makes helps people make good decisions and when do people feel good about the same decision even if it was arrived at in different ways and both are important for that the other thing about you made a point that people can't absorb information and I'm always struck by one of the examples from my own research I did a lot of work on rising returns to college and return to college started going up about 1980 economists really became aware of it about 1985 and when you look back at the data the number of people going to college actually started going up about 1981 so people actually knew a lot at a long time before economists knew about it how I'm not quite sure it's a little bit like Amir's example from the lab I mean that market seemed to have responded in ways that you know we didn't even have the information process from the data yet and people where he seemed to be aware of it at the time and it's often interesting to see that that happened so I think we often sell people short what they can actually do so can't be an academic nowadays so we should be responsive the actual fear or the actual risk these costs are coming correct we have to feather those costs to justify the various problems in people's decisions that optimize outcomes for given and possible correct I agree with that in some sense though there's even a presumption that it's a mismeasure that there is an information problem it's just that people may be different fears of the you know the type of probability maybe has a direct effect I mean it's like I always give the example of my students of like the horror movie or the sad movie people are crying in the movie it's like why are you crying they didn't really die it's just a movie you know but it really does effect you know people get disturbed by movies even though they're fully aware that it's actually only a movie so you wouldn't say that that's an example where somehow we have to show them that it's not real they know it's not real but it still has a real effect so there are two hypotheses one they don't understand the probabilities or two they know they understand the probabilities but just like the movie they get disturbed by what they think about or what they fear and if they're on the plane constantly thinking about it but they don't think about it when they're in the car even though the risk is a hundred times as great well maybe we have to live with that so I'd say you have to consider those two possibilities the true they're either uninformed and we should try to correct it it may just be the actual impact on people of the same probability of death is different in different contexts which I have every reason to believe it very well could be well I want to thank the panel for a very interesting conversation