Discover Your Inner Economist | Tyler Cowen | Talks at Google

Tyler Cowen is a professor of economics and director of the Mercatus Center at George Mason University. He co-runs and writes for MarginalRevolution.com, the number one economics blog according to BlogPulse.com. Steven Dubner, who co-wrote— the co-author of Freakonomics, called this blog awesomely good every day. Cowen also writes a monthly column for the business section of the New York Times. And contributes to many other publications. "Discover Your Inner Economist," subtitled "Use Your Incentives to Fall in Love, Survive Your Next Meeting, and Motivate Your Dentist," is a quirky, entertaining book in which New York Magazine writes, "Cowen ventures where no economist has gone before." I, for one, am certainly interested to see if he can help me with getting Googlers to show up on time to my meetings. We thank Tyler for being with us today and look forward to hearing just how to use fundamental economic principles to get more of what we want out of life. So, thank you. Tyler.

Thank you for the kind words. Today I'm going to talk about your favorite topic, namely you, that is Google. I opened up the New York Times this morning and I saw the following headline. Probably all of you saw it. It was the lead story on news.google.com. It says, "Up to $25 million in prizes is offered for an unmanned landing on the moon." And the group backing the prize is, of course, Google. So the idea is to have an unmanned lunar landing by the year 2012. And depending on the details of the landing, Some amount will be given in prizes. The basic prize is about $20 million. This, of course, is about incentives, and incentives is the main topic of "Discover Your Inner Economist." So the key question is, when is it a good idea to use a prize as an incentive compared to a grant? Let's just define terms for a moment. What's the difference between a prize and a grant? A grant basically is paying researchers for inputs and a prize is paying researchers for outputs. So if I think back to the very first job I had, it was working in a supermarket. I wrapped up produce, fruits and vegetables, and they paid me only on the basis of my inputs. That is, I was paid by the hour. Every hour I spent wrapping lettuce, I got, I think, $6.75 in those days. No one asked how good the lettuce was wrapped. If it was bad enough, they'd fire me, but otherwise I was paid by the hour. I was paid for inputs. Another kind of contract is when you're just paid by outputs. I was writing a piece for Forbes on how to make your meeting— your business meetings better. And in order to get paid for this piece, I had to actually produce the piece. I couldn't send them a bill saying I had labored over the piece for 27 hours and here was my hourly rate. The deal was, "Produce the piece, we pay you. Otherwise, you get nothing." So the interesting thing about these prizes is that in the contemporary world, we're seeing a lot more use of prizes and less emphasis on what are called grants. Grants again are paying researchers for input. Here's some money, go work on a problem and you get the money upfront. A prize is saying, "Here's some money but you only get it if you solve the problem." In economics, there's someone named Michael Kremer at Harvard. He's argued very eloquently for using prizes for new vaccines for the developing world. There's now a pilot project for these prizes. Most notably, the X Prize Foundation, which was behind Google putting up the money for the lunar landing prize, has set up a whole range of prizes for a number of things including fuel-efficient cars, cures for cancer. The X Prize Foundation, it even has a prize for the best prize. So if you come up with a really good prize idea, you can get a prize for them for this. I don't know if there's another prize for the best prize for a prize for a prize. But anyway, prizes are becoming more important. So let's put on our economist hat. Let's use what I call our inner economist and think about when are prizes going to be a good way of getting something done. And I think there's a basic trade-off in using prizes in terms of incentives. And I'll give you what I see as the main benefit and compare that to the main cost. The main benefit is what I call either division of labor or decentralized knowledge. And that is, let's say that you put up a prize, $20 million for an unmanned lunar landing. You as the prize giver, you really don't have to have any idea who is capable of winning the prize. You don't have to judge applications. You don't have to referee proposals. All you have to do really is announce the prize and be credible with it. So you're economizing on knowledge, which is one of the most important things in our modern economy. So if you don't know who's capable of getting the prize, the point is this is no longer important. It's not a limitation. And this is a huge advantage of prizes over grants. The other huge advantage of prizes related to economizing on information is quite simply you only have to pay if the person wins the prize. So if the person does not succeed in creating this unmanned lunar landing, you don't have to pay. And viewing this as an incentive for the person who is trying to strive to create the unmanned lunar landing, it's obviously the case that there's a stronger incentive with a prize than with a grant, because you get the grant up front no matter what. Then you report at the end of your 3 years, "Oh, I tried really, really, really hard." "Sorry, we just couldn't get to the moon, but we published some lovely papers in academic journals." And a lot of donors don't really like that sort of outcome. So the prize has these really neat qualities: decentralized knowledge, and there's a better incentive to actually get the thing done. So if that's the case, you might wonder, why don't we use prizes for everything? Why do we ever pay people according to inputs rather than outputs? And the reason there is what I call something like risk aversion, or you might call it also liquidity. And that is both the great benefit of the prize and the cost of the prize is you don't pay off until a person puts that unmanned lunar landing on the moon. So in the meantime, here you are. Let's say you have a great idea for how to get to the moon, but A, maybe you don't have enough money. B, maybe you don't have money at all. And C, you might worry about the risk. So there you are, you could work, say, your 7 years trying to fulfill this project. But if you don't make it, or even worse, say you're on the verge of making it and someone else gets to the moon a week, a day, an hour, a minute before you do, what happens to your prize? Often you get zero. With this Google Prize, I believe there are some runner-up prizes of $5 million for people who don't take first prize but manage to do something of value. But say you don't get one of the 2 or 3 runner-up prizes. Again, your reward is zero, and this is a very risky endeavor. So you might imagine if someone asked you to work at Google and they just said, "Well, show up for the first 6 years and we'll see how things go. If you do something really great, we'll give you a prize." Some of you would take this offer, I believe,. But actually, a lot of you would not. In the meantime, you need to eat, you need to pay your bills. And even if you think you're really, really smart and productive, there's still risk, if only because someone else beats you out. So when judging when a prize is efficacious, again, it's the basic trade-off: on one hand, risk and liquidity, and on the other hand, there's this issue of it being a better incentive and not requiring the prize awarder to know who's actually capable of winning the prize. One other thing I'll note about prizes, when it's good to use them. When a prize is efficacious, it depends a lot also on the time horizon for when something can be done. Some people at the X Prize Foundation have told me they think prizes are most efficacious when what can be done is over a time horizon of about 4 to 8 years. Let's say you gave a prize for something that maybe could be done next week. This kind of prize doesn't make sense because if it takes years of work to achieve the prize and someone maybe wins it next week, the point is they've already been working for 4 or 5 years on their achievement. So if they're already working, what's the point in putting up the prize? Whatever they come out with next week, we would have gotten anyway. So a prize for the very near future doesn't make sense. It's also the case, a prize for the very distant future You know, let's give a prize to the person who comes up with a perpetual motion machine anytime in the next 26,000 years. It doesn't cost anything to announce that prize, so maybe there's no reason not to. And if we had such a machine, we could afford to pay off. But it's not really a useful prize either because the timescale over which it can be achieved, will the prize still be there? Will the people who read about the prize today still be alive and so on? It's pretty much a meaningless prize. So there's an optimal time horizon for prizes which some people have told me is in this range of 4 to 8 years. So to put this unmanned landing on the moon within 4 to 8 years or 10 years, it doesn't sound like a crazy dream. Since we have in fact put actual human beings on the moon, it sounds like something that could be done. Another thing to be noted about these kinds of prizes is you really need to be able to specify the outcome very exactly. Because you need to award the prize if it's deserved and otherwise not. So as I read the prize in the newspaper, you get the prize if your lunar landing vehicle can move on the moon 1,312 feet. If it can move 1,311 feet, as I understand it, you don't get the prize. You might get a second place prize.— but you don't get any prize at all. So you need to specify very exactly what wins the prize. But of course, you don't have to specify the path there. And that's part of this neat principle of prizes being able to use decentralized information. Now conversely, we've talked about some reasons why you might use a prize, some advantages of prizes. If we're going to ask the offsetting or corresponding question, what are the advantages of grants? We can find some related features to decide when should you use a grant, like give people like me money and let us sit on our behinds and think really hard and who knows what will happen. When should we use a grant rather than a prize? One time you should use a grant is when it's relatively easy to judge who has a lot of talent and who is likely to achieve what it is you want achieved. So if there are only, say, 4 or 5 mathematicians in the world capable of proving a certain theorem. Maybe only they are smart enough or only they have enough background. You might say, "We don't really need a prize. What we can do is we know it's up to these 4 or 5 people. Let's target those people. Let's give them some money. Let's give them a grant and let's just see what happens." So when you know in advance who are those who can do it, that militates in favor of a grant. Uh, it's also the case that when it comes to incentives, you're more likely to favor a grant if you think people are motivated by ideas rather than money. So let's now compare these two incentives. If you're given a grant, as we know, you keep the money no matter how much or how little you do. So the money is no longer an incentive once the grant is awarded. What's an incentive is you get to put your idea into motion. You get to invent something neat. You get to invent, you know, an automobile that will get 100 miles to the gallon, whatever. Maybe you're motivated by love of discovery. To that extent, you would tend to favor grants. But alternatively, let's say that people are motivated by pure cold hard cash, like you'd pull out of your wallet. That would tend to favor prizes because the prize pays you off a lot of money. But also, let's say people are motivated primarily neither by ideas nor by cash. But they're simply fame lovers. They like publicity. They like to be on television. They want a lot of attention. That in most cases militates in favor of the prize rather than the grant because the prize itself creates a lot of media. You know, here's the New York Times article about Google. It was the lead story on news.google.com. It was in every major newspaper. And that hasn't even happened yet. So if it happens and there's the $20 million awarded, Whoever wins the prize will get a lot more attention than if it had just been someone with a grant because you have this whole machinery of the X Prize Foundation and Google behind the publicity for making this achievement. So depending on your theory of human motivation in a given instance, if you have fame seekers, go for the prize. If you have money seekers, go for the prize. If you have people who love just the process of creation, then you might tend to go for the grant. Grants also make sense when people, the inventors, are liquidity constrained. They need the money to get going, to pay their rent, and when inventors are liquidity averse. There's one other way of thinking about grants that I think is quite subtle, and it's worth pointing out, and it gets to a key question of how do we really define the difference between a prize and a grant. And on first blush, the difference between a prize and a grant It sounds fairly simple. You think that if you put this unmanned vehicle on the moon or invent a car that gets 100 miles to a gallon, uh, that that's a prize. Whereas if the National Science Foundation sends you a check for $100,000, that's a grant. But in reality, a lot of grants have some incentive features of prizes. It's just a bit hidden. So let's think about getting your National Science Foundation grant for $100,000. You get the grant, you cash the check, it's in your bank account. But it's still the case that if you do something wonderful with that money, if you figure out a new way to scan, scan the brain using neuroimaging, you become a famous person. You can write a book, you can go on the lecture circuit, you might win a Nobel Prize further down the road. So grants arguably are connected to some later prizes. And a person who wanted to defend grants might say something like this. They'd say, "Look, I'm all for prizes, but today we already have a system of prizes. They're just not all explicitly defined and announced. What happens is that if you get a grant and do something wonderful with the money, you will win a lot of implicit prizes. You'll be a lot more popular. You'll have higher status. Your speakers' fees will go up. And a system of grants actually has some of the incentive benefits of both grants and prizes combined. So that's a very good point. Question? Sure. The question is about the Netflix Prize, which I believe is for $1 million. If anyone can come up with a better algorithm for taking people's tastes in movies and predicting which movies they'll then want to rent from Netflix. And there's a prize for achieving this. You have to beat, I think, Netflix's own algorithm by some certain percentage. Tyler Cowen. Let me turn to this question now. And this is the case of how much effort should your prize be trying to induce. And if you want to run prizes in a good or socially optimal way, it's important that you induce a lot of effort, but it's also important that you not induce too much effort. So let's say, for instance, that you offer a prize for a million dollars., and people spend $10 million worth of their time and energy trying to win the prize. We economists, we sometimes call this winner's curse. That is, people are over-optimistic about winning the prize. So if there's a million-dollar prize for a better Netflix algorithm, and people out there are spending $10 million of their own money and time, it would seem that this is an over-investment in energy in winning the prize. And even if you got some good solutions, If what you're offering a prize for isn't that valuable, it still seems from a social point of view wasteful, but from Netflix's point of view, a good way of tricking people into basically working for them for free. So when you're asking, for what things should we give prizes? Not predicting like, when will prizes work? But ask the normative question, when should we give prizes? We have some fairly simple intuitive answers based on the idea of incentives. We should give prizes when the achievement you're giving the prize for has a lot of external social benefits. So if we think about Netflix getting a better algorithm for predicting tastes in movies, I think there are some external benefits. So if my wife Natasha in the audience, if she has a better idea what she will rent next on Netflix, she can tell me, she can tell 5 friends, a number of people are better off even if they're not customers of Netflix there's some social benefit, but I'm not sure that's an area where the social benefits of the achievement are very high. But basically, the most socially optimal prizes are in the areas where the social benefits of the achievement relative to private rewards are extremely high. So that would tend to be what I call basic science. So if you look, say, at the people who discovered the principles behind quantum mechanics, most of them— or none of them got rich. Well, if they did get rich, it was not because of their discovery. Einstein, for the theory of relativity, did not get rich. Later in his life, he, he did have opportunities to earn money, but he didn't get paid much for the theory per se. People working in the foundations of biological science, if you make a breakthrough, it will have implications for many different ideas, but it's a form of very basic science. So in general, people working in science at the most basic of levels tend to earn the least relative to the social values of their contributions. So one principle is the optimal social prizes are simply for areas that are most basic to science and will translate into the greatest number of discoveries from other people. There's also a question that relates to the incentives of the donor. In what kinds of areas should we offer prizes? And there are two different forces pulling here— pulling us here in different directions. One argument would be that you should offer prizes in an area that's relatively neglected. So for instance, there are some prizes now for cancer research, cure— cures for different kinds of cancer. And you might say, "We don't need prizes in this area because there are already many different people working on cures, say, for different types of cancer." And therefore, the extra momentum of a prize is relatively small. Whereas if there's some area that no one is thinking about, say a cure for some kind of disease that hardly anyone has heard of, you could argue that there the prize is especially important because otherwise it simply won't get done. But on the other hand of the equation, there's this desire which a lot of the donors have and a lot of us have just as spectators, that the prize be given for something interesting or fun or sexy. So if someone tells you about a moon landing We get excited. If we're old enough, we think back to 1969. We've read Robert Heinlein. We've read science fiction. Moon landing, it's fun. We'd watch it on the news. We'd watch it, you know, on Google and so on. So sexy prizes in some ways tend to have the greatest value. They have the greatest bang for the buck and they have the most publicity. So do you want a prize in an area that's ignored or do you want a prize in an area that's sexy? It may be that the answer goes back to what's your assumption about the researchers If you think that researchers are motivated by fame, you may want prizes in areas which are sexy. If you think that researchers are just motivated by money, then you may want prizes in areas which are relatively ignored. Now also with prizes, I think a key question about prizes has to do with the incentives of people who offer prizes and not just the incentives of people working for the prize. So we've talked about scientists or researchers striving for some achievement. But also the other side of the equation is that people who offer grants or people who offer prizes, they have their own reasons as well. And we can think of these reasons as falling into a number of basic categories. People who offer prizes or grants, they might care about results. They might care about the fun. Of giving the grant or the prize. Or they might care about, you know, the fame. Or in the case of a corporation, they might care about favorable public relations. So if Google backs a prize and someone puts this vehicle on the moon, it's good news for Google. And Google is seen correctly as having done something very meritorious, very socially worthwhile. So think of results, fun, and fame as some of the rewards to donors for putting up prizes or putting up grants. But with regard to these rewards, prizes, and grants, they work very differently. So grants have a neat feature that they will mobilize a lot of donors because a lot of donors or backers or supporters, they like the idea of discovering something unknown. So if I think of my own activities in life, one of the things my wife and I do is we support some relatively unknown painters in rural Mexico.. And we offer them what are in effect grants, like grants for doing their painting. And I at least, I'm not sure about my wife, but I have the view, possibly deluded, that these painters will go on to be very famous someday. And I take some pleasure in the fact that I will be able to say in my deluded expectation, "Well, I spotted these painters when they weren't so well known." And that makes the entire venture more fun for me. I'm more directly involved. We call the painters about once a week. We talk to them. There's progress reports. There's steady feedback. We buy pictures from them. The pictures come in the mail. We call them up again, hear about the family. It's an ongoing process where we have feedback pretty much every week, every month of our lives. And we have a sense of having been involved at the ground floor.. And that I at least find gratifying. And I think a lot of other people who support projects do as well. A possible problem with prizes is you don't really get this benefit. So if you offer a prize, you say, "$20 million if you put a vehicle on the moon," and then you just sit back and you wait. And in the meantime, what happens? Well, you might watch people trying. But you're not so directly involved. You're not the person or the institution that spotted them in advance, right? And when the achievement comes, then they have done something great and glorious and you recognize it. So there's less of a sense of sharing in the glory on all the way up. So I think this is one reason why prizes are not as popular with donors as they should be. It's simply the sense that you make one announcement and you wait for a long time and nothing happens and you get no feedback and there's a kind of scarcity of attention on the side of the donors. Where are you? You're just sitting there. And a lot of people would rather have this more involved relationship with very regular, very direct feedback. I think another limitation of prizes from the point of view of donor incentives is that when it comes to a prize, it's very clear who is offering the biggest prize and who is offering the second biggest prize. So, okay, Google has put up between $20 to $30 million of a prize for an unmanned lunar landing. And that's great. Let's say that I, as an upper-middle-class individual, that I too were to favor a lunar landing. And I would make my own announcement that whoever puts the Sunman vehicle on the moon and moves 1,312 feet, I will send them a check for $50. Or if I'm feeling generous, $1,000. That's the Tyler Cowen unmanned lunar landing prize. It's pretty clear from all of your reactions and from the fact that I'm not doing it that this really isn't very rewarding from my point of view. I'm always second dog relative to Google. And odds are no one will offer this kind of second-tier prize. But that's not so much the case with grants. So let's say someone is offering a grant to Damien Hirst in England to paint very famous pictures. And then I offer grants to my painters in rural Mexico. I don't really feel bad that the grant to Damien Hirst is bigger. Like, I redefine the terms of competition so that my painters in Mexico, well, they'll turn out to be more famous. They do something different. It's a different kind of art. Of course you don't expect them to sell at Sotheby's. I can still take a lot of pride from my grant to Mexico. And the existence of a bigger grant to Damien Hirst, the famous painter, the guy who puts, you know, the dead sharks in the formaldehyde in England. The fact that he's more famous, it doesn't really take away from my grant much at all. Because I still get the affiliation I wanted. So when it comes to prizes, like what's the big prize and what's the little prize, it's in a way too clear. So once a big prize is announced, the expectation is we won't be following up with a lot of little prizes for the same thing. So with prizes, it's harder to mobilize additional capital. But grants, you can mobilize additional capital because Damien Hirst gets his grant, my Mexican painters get their grant, you can support some other artist of your own, and there's sort of always more to go around. So this is another reason why, from the donor's point of view, a grant in many cases will be preferred to a prize. So those are two reasons I think why prizes are not more popular— no one wants to be second dog, and a lot of people want to be involved from the ground floor. But those same reasons— I'll get to your question in a minute— the same reasons that explain why prizes are not more popular, they're also an argument why it's especially important to have a lot of prizes. Because what a prize does is it steps into a gap. It steps into an area where we ought to have a particular kind of incentive but the incentive isn't there because donors, supporters, backers are for these reasons somewhat shy to commit their capital. Question? You asked what will happen if we have bidding wars for the prizes. These major prizes are very new, but I'll tell you what is my guess for the equilibrium. My guess is that Microsoft would find it bad public relations to try to upstage Google directly. So if Microsoft were to offer a prize, it would not be for putting an unmanned vehicle on the moon and having it move 1,312 feet. That Microsoft might decide to offer a bigger prize. But they'll try to pick an area, or they have already in the case of vaccines in the developing world. They will try to pick areas that maybe sound more noble or are somehow different. And I predict the final equilibrium is you have a kind of carving out of the space of possible sexy prizes. Where for each possible prize area, there's one major prize giver and not that much backup support. And I think that's what the equilibrium will look like. I think that's what we're tending toward. But not that someone will try to double Google and then Google will double them and you'll get into this bidding war. I think that looks not noble enough from a public relations standpoint. There's always some other area where Microsoft could choose to give a prize. So anyway, if this is the basic setting, we have donors motivated by results, and fame. And that means that we don't actually have as many prizes as we ought to. If we go back in world history, we see a very interesting tendency. If you go to the 18th century, for instance, prizes were typically the major means of awarding scientific achievements. There's a very good book, I think it's by Dana Sobel, it's called Longitude. And the man who discovered the system of longitude, he wasn't given a grant, but he won a prize. So in the 18th century, prizes were the norm. Today we're seeing a resurgence of interest in prizes. But through a lot of the 19th and 20th century, there was an enormous increase in grants relative to prizes. And I think from the point of view of history, this is poorly understood. But if we're going to understand why we might want to choose a prize or a grant as an incentive, we need to figure out why is it that prizes fell out of favor. And I have a hypothesis about this. I don't really think it's backed by firm evidence, but my hypothesis is the following: that the rise of the, what I would call the modern university, the modern university turned out to be a very effective way of keeping donors interested in research. And the university was a vehicle of bringing supporters together with researchers, researchers, and through the mechanism of being an alumni, having gone to Stanford, MIT, Harvard, wherever it is you went to,— that you felt an automatic affiliation with the people doing research at that school, and you were more likely to support them through your original university because you felt connected to them, and that prizes didn't offer a comparable benefit. And I think so the rise of the modern university led to this decrease of attention to prizes, increase of attention to grants. And then the question is, if that's the case, if that's what happened in the 19th and 20th centuries, which I find plausible, Why today are we seeing increased interest in prizes? And I think there are a few reasons. I think one is simply that a lot of philanthropists are giving to areas other than universities. They may feel disillusioned with their alma mater. A lot of people who are doing philanthropy, they want to somehow be innovative. But I think also the growing wealth of the extremely wealthy is a factor behind prizes. That part of what makes a prize work is that you are the big dog on the block. You have the one large prize and the prize has to have a large enough dollar value to capture people's attention and to motivate the people to perform the task. And to get people to put something on the moon, a prize of $50 is not going to do it. So you need companies, foundations, individuals who have enough wealth where they can offer prizes of $20, $30, someday I think we'll see a prize for $100 million and afford to pay it. So since in the United States the wealthy have become relatively wealthier over the last 15 years, this has in my view favored prizes. And I think also the growing celebrity factor in our culture has favored prizes as well. That a small number of foundations, companies, and people can become known as the institutions which give prizes. So the Gates Foundation is known in this manner, the X Prize Foundation, Google is now known for the Lunar Prize. And the fact that media coverage is so intense means that this has become relatively more rewarding relative to the process of offering grants, sending it back to a home university. You know, every week getting a letter in the mail saying like, "Here's what's new at Princeton." and taking some reward in that. But that all of a sudden seems less attractive or less sexier, less interesting than it used to be relative to these very expensive, very fancy prizes. Question? Well, there's both. X Prize Foundation is of course funded by donations. It's not commercial. Google's philanthropy I don't quite understand. I know there's a wing of Google which does philanthropy but on a commercial basis. That's a very interesting blend of models, and I think we'll see a lot more of it. But whether you're philanthropic or commercial, there's still the question: should you give a grant or should you give a prize? And there's still the trade-off of these different incentives when making that decision. Now let me just give you a 4-minute version of the talk I had been planning to give you, which is not the talk I gave you at all. But I thought as the, as the guest, it's always polite to talk about the host.. And I saw Google in the newspaper about prizes, so I thought we needed to talk about prizes. What I was gonna talk about is how to save the world and how to give away your money. And here's a 4-minute version of that talk. First principle: Don't give money to beggars. It's a bad incentive. Giving money to beggars encourages begging. It's like a prize for begging. The incentive is pointing in the wrong direction. Give your money to people who are not expecting it, people who are working or taking care of family members. Second principle of good charity: ask what is scarce. What is scarce is usually your attention and whether or not you care. Do you remember the New York Times headline from June 2005? It was, "Hundreds are feared dead in Chinese floods." How many of you actually remember this? Not many. This showed up on page 5. It wasn't a big deal. It wasn't a major story. Most people didn't care. The thing about charity is no matter how much you're giving away to charity, a lot of the time you don't care either. You're much more wrapped up in your daily life. Often the exhortation is to care more, care more, care more, but I think this is often counterproductive. We're just not capable of caring more. We're self-focused. It's often more useful just to admit to yourself, "I don't always care that much," and just say to yourself, "Hey, I don't always care." What happens then? Once you make that admission, charity then becomes an area you can think about again, because if you're always telling yourself, "I must care, I must care," and you don't, then thinking about charity makes you feel bad. What you do is you decide not to think about charity very much, and you don't give a whole lot. So wake up in the morning and say to yourself, "Hey, I don't care that much." That's the first step toward giving more. It's counterintuitive, but I think it works. When it comes to getting charitable requests in the mail, this is sometimes called direct mail. Direct mail is pretty expensive. If you get a good package in the mail, the nonprofit that sent it out to you might have spent —up to 50 cents or up to $2 preparing that package. That's a fair amount of money. So, the thing to do is just make it easy for the charities you want to find you. It's easier for you to find them. So, get on their lists, become, you know, what's known as a regular donor. Get off the lists you're not using, and you'll save everyone a lot of money. So, one thing this means is that if there's a charity you like and you send them $30 and you never send them anything else again, and they keep on sending you these mailings, you've actually drained them of money. That's counterproductive. So, that's not necessarily the thing to do. On the other hand, if there's a charity you really don't like, maybe a political cause you think is evil or misguided, send them a check for $10, write them a long letter on fancy stationery saying you're about to go vacation on your yacht, give them all changes of address, and just let them keep on sending you mail for the rest of the existence of that charity. But the point is, when we get to this question of scarcity, what's scarce there is them finding you. The charity finding you will make that easy. You will do a lot more good if you find the charity, and then I become a very loyal giver. Basically addict yourself to the charity. Just some very simple questions I often get: Christmas gifts, in my opinion, is a form of charity. They're pretty much a waste. The rate of regifting is very high. There was one economist, he did a study He estimated that $4 billion in economic value is lost every year from Christmas gifts. And he assumed if you get a gift that the person spent a dollar on, it's worth 83 cents to you. To me, that's crazy. If I get a gift worth a dollar, I feel lucky if it's worth 30 cents to me on the dollar. So the odds are we're wasting, I guess, at least $20 billion a year on Christmas gifts. Now, it's not a waste if you enjoy giving, if it's a signal to a friend, but don't think of it as a form of charity. It's not.. It's really you selfishly investing in your networking, which is fine. Just don't substitute it in for charity. People sometimes ask, "Should I give to the cause of the day? What about, you know, 2004 when there was the big tsunami and everyone went around to all the offices taking up collections?" We have very limited evidence on this, but the evidence is that giving to like a cause that pops up very suddenly, it does not detract from other giving. So probably you should give, but keep in mind that giving to the cause of the month, it tends not to be very effective because the infrastructure isn't there and a lot of the money is wasted. But the key point is, if you're gonna give to the cause of the month, it's like a snowball effect. The person next to you in the office gives, and then the person down the hall gives, and so on. So make sure you're giving visibly when you give to the cause of the month, because that will cause other people to give too. But if you're choosing between your loyal charities and cause of the month as alternatives, then stick with your loyal charities. People sometimes ask me, finally to close, "What about fair trade coffee?" You know, you buy fair trade coffee and supposedly behind the production of that coffee they treat the workers better. I have mixed feelings about fair trade coffee. A lot of smart economists are spending a lot of time building mathematical models about when does fair trade coffee work. But I have a simple thought experiment. Let's say instead of selling fair trade coffee, they took one brand of coffee, they and they labeled it "Exploitation Coffee." And they sold you that and they promised that the workers who made that coffee would be treated actually a lot worse and paid less. And that was Exploitation Coffee. A, how would you feel about that? You might be put off. But it's the same thing as Fair Trade Coffee, it's just we're changing the framing. And the second point is B, actually if you're going to buy coffee anywhere, there's arguably a case that you should buy coffee from the workers who are being paid the least because if a lot of other people do the same, you'll bid up their wages. So probably what fair trade coffee is doing, to some extent it's helping some poor people, but it's helping the poor people who are already earning a bit more, and it's hurting the much poorer people who are producing what I call the exploitation coffee. So, you know, buy that coffee if you like it as coffee, but don't think of it very much as a huge attempt to be charitable. Anyway, that was the talk I was gonna give, but you got the 4 or 5 minute version of it. And you got Tyler Cowen on prizes. And we still have a little more time for questions. Thank you all for coming, and I urge you all to discover your inner economist. Thank you. Question? That's a good question. The question is, how do we know a charity is actually supporting people rather than spending all the money on their expenses? There are a number of ways to research charities, and there are ways charities are evaluated. So start with Dare I suggest at Google, and see what comes up. But I would say if you're still truly uncertain, then maybe you shouldn't give money to that charity. I have a simple favorite alternative, and that is to simply send money using Western Union to strangers on the other side of the world. I ran an experiment on my blog where I put up a request for people from India who did not have a lot of money. I said, "Write in to me." I will not do anything to monitor your request, and I will pick 10 of your requests and simply send you money by Western Union. There's of course no tax deduction, but this is what I call zero overhead giving. And I don't doubt that some of the people who wrote in were not necessarily poor, or didn't need the money, or they made up a reason, so there was some wastage. But I actually think the wastage was fairly small. And if you're really not sure about any charity after having done your Googling, I would just say experiment with zero overhead giving. There are some people in New York who would be happy to take your money. But again, don't pick the beggars. Just look for someone who might need the money and give it to them. And you don't get the tax deduction, but again, if you have to, just give proportionately less and you'll still probably do the world more good. Question? The question is, Carnegie gave a lot of money to set up libraries around the US. Is there a way to do that with prizes? Well, I view the X Prize Foundation as trying to set up a large series of prizes for different kinds of creations. And their hope also is that people will copy them. The X Prize Foundation on its own probably won't have enough resources to set up a prize for everything we want a prize for. But there's a demonstration effect with giving. Just like Bill Gates's giving spurred Warren Buffett to do something similar. What we're seeing is more and more publicity for prizes is leading to more prizes basically. So we're seeing it to some extent. We don't really need more paper libraries and we, of course, have the web as an online library and now we're seeing this proliferation of prizes. Yes?

It was your blog that talked about the X Prize being set up. Can you share something about that?

That was a post written by my co-blogger Alex. He wrote a post on the X Prize being set up as an insurance contract. What the X Prize Foundation did— this is now my hearsay from Alex— is they bought an insurance policy for one of their prizes where if they have to pay off the prize, the insurance company pays them. The insurance company thought that they would never be able to have the prize conditions be met, and so they were able to buy the policy quite cheaply. So from the point of view of the foundation, this was a great deal. For the creator, it was a great deal. For the insurance company, of course, they lost money. And that's what happened there. You can Google back to that, looking at one of Alex's posts, my co-blogger Alex Tabarrok. Probably not gonna work again. Probably not gonna work again. Less likely, right? So we should keep quiet about it, perhaps. Yes? The question is, if someone begs, you can give them food, money, or attention, what should you give them? I would say it depends on what the alternative is. If the alternative is doing nothing, then it may even be worth giving the beggar some money. But food may be better or attention may be better. But if the alternative is giving that same item to a non-beggar, I would say giving to the non-beggar typically dominates. There's also the problem of what's called aesthetic overload. There have been some studies done. If one beggar approaches you, there's some chance you'll give money. If 40 beggars approach you all at once, or 200 approach you, the chance you'll give money is very small, because it seems to you that the problem is a hopeless one and you can't make a dent in it. Now, whether the beggars are visible shouldn't matter. There's still millions of beggars out there. But you wanna choose those giving strategies where you will get intermediate feedback, credible and truthful signs of success, so you're not feeling aesthetic overload that your efforts are hopeless, and you want to keep on affiliating with your cause and tell yourself, you know, true and focal stories. About the successes, so you will in essence addict yourself to giving more. And my sense is, viewed in those terms, beggars are not the way to go. That if you give to beggars and you have bad experiences— people wrote into me, you know, stories they would give money to beggars and then the beggar would rob them and so on. It tends to put the giver off of charity more generally. So that's another reason to give to non-beggars rather than beggars.— if it's an either-or. But it always is. You can always just wait and then give what you would have given to the beggar to someone else. Unless it's highly perishable strawberries. You have strawberries, you're out in the heat of Calcutta, and you don't want them, and you give them to the beggar or they go bad. Sure, give them to the beggar. But that's usually not how it is. And money for sure is a durable thing. Yes? How do you feel about commercialized charity like yogurt caps for breast cancer? I don't know about that particular offer, but very often those are done for reasons of marketing, and what actually goes to charity is unclear. And even then, it's still going to some other nonprofit which will have its own expenses and costs and processing, and it's very hard to know where the money is going. So, my belief is there are in fact a number of very good programs of that kind, and if you know a particular program is good, that's great. But if you're just doing it blind, I would treat it as acting on a lark and not as checking the box, "I did my work for charity today," because it probably isn't. Yes. That's a good question. The question is about paranormal prizes, the offer of $1 million to anyone who can produce, you know, replicable evidence of paranormal abilities. I suspect that's more like an anti-prize. The people who give the prize are pretty convinced it can't be done and they want to prove it can't be done. It's not that they hope to cultivate paranormal activities. You can imagine some kind of genetic engineering for paranormal activities. To try to win the million-dollar prize. But I don't actually think that's gonna happen. It's more like a dare. And you can always say, "Hold up, look, we're offering a million dollars and no one can do this. Surely there is no evidence for the paranormal." And it's a kind of marketing ploy. So a lot of things that may look like charity or look like incentives, they're not always incentives. They're very often marketing ploys. And that's fine. I have nothing against a marketing ploy. I don't believe in the paranormal, and if I argue with someone, I like being able to say, "there's a prize for a million dollars and you still can't prove this stuff. Look, it must be simply not true." But that's how I tend to view it. Yes. Sure. The question is about Nobel Prizes. And sometimes people who do, like the X Prize Foundation, sometimes there's an insistence that these prizes be called awards. Some people use different terminology. But of course, the Nobel Prize is called prize. So it's a prize of sorts. The Nobel Prize is essentially for lifetime achievement. It is not for a specific contribution. Or you could define the specific contribution as something like, you know, making 9 Swedish people very, very happy with what you have done. But if the Swedish people are good judges of literary merit or economic merit or whatever, then that will work just fine. When I think of my own field, economics, I'm not sure how much the Nobel Prize has mattered. It causes a lot of people to lobby for the prize. So many more top economists visit Sweden. Which is fine. But I don't think they really do better work. Maybe the Nobel Prize causes them to take more chances. Because to get the Nobel Prize, you can't just play it safe and be number 2 in a field. You have to be number 1. You have to be a pioneer. But I tend to think that it's very often the immediate incentives in front of our face. Competition with the guy down the hallway or just the desire to solve the problem. And the notion that you might get a Nobel Prize 30 years from now, I think even for people who otherwise have long time horizons, that's somehow too abstract to get the adrenaline pumping. And I don't think it matters that much. I think it's a good public relations ploy for economics.. It puts us on the front page of the New York Times once a year, which of course we deserve. And again, it does a lot of social good and I don't see any harm in it, but I'm not sure it really functions very much as an incentive. Yes?

We actually have something internal to Google that is a prize for an award. It's much more like the Nobel Prize. It's nonspecific. It's just for doing good work or enhancing the business versus the specific of accomplish this very specific goal. And, you know, do you— how would you expect, you know, some large prize like this to affect sort of a, you know, inwardly facing environment like a workplace? And do you think that something like that could be successful? And have you heard of any other companies or organizations doing similar sorts of things?

There's the question of awards or prizes at Google which are given not for very specific contributions but for overall achievement. Awards of that kind tend to work well when the level of morale is high. Here's the possible problem with such awards. People in a workplace— I'm not saying here, of course— but they often tend to think that the prize or award is simply given out to the favorite of the boss, and that it's a way of reallocating resources within the firm without having to very directly cut the wages of some people and raise the wages of others.. And to the extent that's all it's doing, then it tends to have a negative effect on morale. If there's a high enough level of trust in the workplace that that's not how people view it, and I'm assuming that's the case, then it is a real incentive. And corporations do that all the time. Academic departments do it all the time. They don't always call it an award, but it may be like an office with a window, you know, it's promotion to full professor, it's extra travel money. It's another research assistant. And they're not— these things are not given to you for some very specific— the department chair doesn't say, "I'll give an extra research assistant to anyone who publishes a book with Harvard University Press." But rather, if you have some good publications, you'll get some extra rewards. And that kind of system works very well. And the fact that it's not pinned down in advance, you can take advantage of subtleties and framing effects. And if there's enough trust in the workplace, use those in your favor rather than against you. Yes, over there. How do you use incentives to fall in love? I met my wife Natasha almost 5 years ago and I fell in love and she's in the audience today and we're still in love. And the first step toward that is to buy "Discover Your Inner Economist." She's a character in the book and her daughter Iana who didn't always do the dishes, but Jana is a character in the book as well. And in part it's my story, but it's also some general tips. But I would say one way to start is to use self-deception in your favor, but avoid bad self-deception. And people very often, they think they know what they want much more than they do, and their worst enemy can be themselves. So figure out the ways in which you're your worst enemy, stop self-deceiving in those ways, and realize you don't always know what you want. That would be tip number 1. Tip number 2 is buy and read the book. I think with that our hour is up. But thank you all for coming and for your questions.