 So my topic for this afternoon is game theory, or I like to add kind of subtitle to it, a game theory and entrepreneurial twist. So I think game theory is an area where I think it is potentially very useful to us as Austrian economists, and I'm going to build the case that there is a way that we can think of game theory as being really a part of praxeology and the overall logic of action, but the way that it is typically applied is perhaps problematic. So I think that it's probably best described. It's one of those cases where in theory there's no difference between theory and practice, while in practice there is. So we need to get this right, because after all, the purpose of a theory is not to build some beautiful construct that people are good at math. Wow, that's a beautiful model. We're not looking for beautiful models that we can't take pictures of. We are instead looking for descriptions of reality. How is it that reality actually works? So this is also where it drives me crazy whenever you hear somebody say, oh communism is great in theory but just doesn't work in real life. We fundamentally misunderstand what theory is for in that case. The theory is there to explain real life. If the theory does not work in real life, you need a new theory. So my hope then is here to kind of lay out how game theory is often used to answer the fact that this often doesn't work in real life and how can we then use game theory more usefully that will actually be descriptive of the reality we see. So first, let's kind of give some legitimacy to game theory. I don't have to try too hard. Murray Rothbard actually talks about game theory very briefly in man economy and state and one of the appendices for chapter one. He says outside of economics, which he breaks down as crossover economics and catalectics, the rest of praxeology is an unexplored area. And then he lists some points where it kind of starts have been made. And he says a theory of games has been elaborated. So it seems that Rothbard considers a theory of games to be part of praxeology. It's about the logic of action. So it's not just about how can you win chess or how can you play a better D&D game or something like that. I'm also interested in that last one so we can have that discussion later if you want. But that's not what the lecture is about. What is game theory then if it's not about these types of things? So we're imagining now our cases where strategic interactions matter. That is where when I think about the action that I want to take, I need to take into account the actions that other people are going to undertake because the end result is going to be a combination of the choices I make and the choices that others have made. So I'm going to start then with a classic example. So here we have the prisoner's dilemma. So we have two people, Bonnie and Clyde. I would recommend if you're sitting in the back and it's hard to see the slides, definitely bring up the slide stream. I was doing that this morning. It works really well. So mises.org slash ppt. I definitely recommend that. I try not to use slides, but I haven't figured out how to describe this thing in words that would actually stay in my own mind, let alone everyone else's. All right. So the prisoner's dilemma, this is a classic story, goes something like this. So Bonnie and Clyde are partners in crime and they're going around robbing banks and to make this a crime, they're robbing Rothbardian 100% reserve banks. They're actually thieves, right? Taking money away from people. So eventually law enforcement catches up with them. We'll say that they're private law enforcement. So these can be the good guys in the story. Catch up with Bonnie and Clyde or take them into the police station and they separate them. So Bonnie goes in one room, Clyde goes in the other, and this is the deal offered. And so the police say, I'm going to go from Clyde's perspective, says to Clyde, so over in the next room Bonnie is either confessing to the crimes that the two have committed together or she's not. If she is confessing, then, and you don't confess, we can put you away for 10 years. We're going to get enough evidence from her to convict you of very serious crimes. So if she confesses and you want to confess too, we can engage in a plea bargain. We'll cut that down to five years for you. On the other hand, or if Bonnie is not confessing in that next room and you don't confess, we still have enough evidence to get you on some lesser crimes. So say two years of prison is what you can expect. But if you help us put her away and you confess while she doesn't, she goes away for 10 years while you get to just go home. You get to go home scot-free. So Clyde thinks about the situation. What does he want to do? Well, I'm going to make the assumption here that prison is bad. This is a case where higher numbers is not good. So Clyde considers each possibility because he doesn't know what is Bonnie doing next door. He can't see, he can't hear what she's doing. So he runs through each possibility. He says, well, perhaps Bonnie is confessing. In that case, if I confess I only go to prison for five years, not great, but better than the 10 years that I would get if I don't confess. So if Bonnie's confessing next door, I also want to confess. On the other hand, if Bonnie is loyal and true and not confessing, then I definitely want to sell her out because then I don't have to go to prison at all. That's the best possible scenario. So Clyde comes to the conclusion that it doesn't actually matter what Bonnie is doing in the next room. His best choice is to confess. Bonnie in the neighboring room goes through the same line of reasoning, faces exactly the same incentives that Clyde does and therefore reaches exactly the same conclusion. She also wants to confess regardless of what she thinks Clyde is going to do, whether he is loyal or disloyal or what have you. So we then have a simple conclusion of what is actually going to happen. We call this the Nash Equilibrium, named after John Nash who came up with this concept. It's actually a concept from math that he just applied to this type of situation. So we end up with the result that both of them confess they both go to prison for five years. So I want to point out just a few traits of this that make this what we call a prisoner's dilemma. So a prisoner's dilemma is not a prisoner's dilemma just because it's about prisoners. We use the same terminology in lots of other situations and it has a few key elements. The first element is there's exactly one Nash Equilibrium. I will show a case later where sometimes you end up with more than one or it's not so obvious what the prediction is going to be. Having one Nash Equilibrium makes it very obvious what the prediction is going to be. It's a very clear prediction here. Everybody confessing five years in prison apiece is what we would expect to happen. And so that's the first point, exactly one Nash Equilibrium. A second trait about a prisoner's dilemma is that Nash Equilibrium is supported by what we call dominant strategies. A dominant strategy is where whatever the action is you take is always the best action regardless of what the other person does. So in this case as you reason through we figured out Clyde actually doesn't care what Bonnie is doing in the next room. His choice, what it should be, is always the same. He should always confess. That's always going to be what's best for him. So that would be a dominant strategy. So Clyde has a dominant strategy of confessing, Bonnie has a dominant strategy of confessing as well. And then the last trait of a prisoner's dilemma which is why certain economists love these types of examples is that that Nash Equilibrium is lousy. Now lousy is of course a technical term. No, we would say it's not Pareto optimal if you want the technical term. But what do we mean by this? I like lousy better, I think it's more intuitive. What we mean is that there is another possible outcome that they could obtain if they just played the game differently. And they both agree that this other outcome is actually better. So in our case both of them confessing is that Nash Equilibrium would predict that as we went through the reasoning, we have a good reason to expect them to do that. Yet at the end of the game Bonnie and Clyde both say, gee, if only neither of us had confessed, we would have saved three years of each of our lives in prison. We would have been better off if we could have ended up in that bottom right corner instead of the top left. So now that we have this conflict between what is individually our best choice confessing and what is collectively the best choice, ah, now there's room here for economists to say this is really interesting. We want to find these types of cases because we can use this. So let's look at a more economic case. So we have something called the tragedy of the commons. So the image here is that Bonnie and Clyde were taking them out of somewhat modern America. We're putting them into a medieval village. In this medieval village we have hovels all surrounding the central grassy area. And it's grassy areas where you take your sheep, you take your cattle, and you graze them there. So each of them have a choice of either using the commons or not. The problem is that if everybody uses the commons, we end up in the top left here, we end up just destroying the commons. They get over grazed or all of the grass gets eaten, it devolves into a mud pit over time. So now it's true in the short run you do get the benefit of your cattle, your sheep eating from it, but in the long run it becomes totally useless. Now it's also possible that neither of us would use the commons. So we have just this pristine wilderness in the middle of the town, and it's nice to look at, it's not particularly useful for cattle since we've decided not to do that, right? But at least it looks nice, it looks better than a mud pit. And then we have the other possibilities where one uses the commons and the other one doesn't. Well in that case whoever uses the commons uses it enough to actually get use out of it, but not enough to destroy it, which is of course beneficial. Meanwhile the person who doesn't use it clearly gets no use from it at all. So let's reason through this again to figure out with Bonnie and Clyde whether they would use the commons or not. So I'll start with Clyde again. So Clyde can either use the commons or not, so he thinks through, and so well if Bonnie uses the commons and I don't, then I get nothing from them at all. If I go ahead and use them I at least get that temporary benefit. Sure that the commons get destroyed in the long run, but I at least get something from it in the meantime. Meanwhile if Bonnie doesn't use the commons, well then I definitely want to use them because I can use them forever in that case and continue to feed my cattle, my sheep, and so on from the commons. It's extremely useful for Clyde. So Clyde then reaches the conclusion, it doesn't actually matter what Bonnie decides to do, he wants to go ahead and use the commons. Bonnie goes through the same line of reasoning, faces the same situation, and therefore makes the same decision. She also wants to use the commons. So what happens, we end up with the desolation of the commons over time. Now nowadays when we apply the tragedy of the commons you normally aren't thinking about common grazing land in the middle of a medieval village. We think about things like say fishing in international waters or these types of things. The way that I get benefit as a fisherman is by pulling fish out of the ocean, but if too many of us do this, fish populations decline and then we don't have any more fish. That's a problem. So it's the same type of situation, just change the animals you're dealing with I suppose. So that would be another application. If we look carefully, we see this is also a prisoner's dilemma. There's exactly one Nash equilibrium, a very clear prediction. It's supported by dominant strategies. It doesn't matter what Bonnie's going to do. Clyde's best choice for himself is always the same. And that outcome is lousy. That is, we all agree that the bottom right, the pristine useless wilderness would be nice to get some use from it, but at least it looks nice. I'd rather have it look nice than be a big mud pit. So we agree there's a better outcome out there. Another possible example we could see is the case of public goods. So in this case Bonnie and Clyde are thinking about how their getaway car is going to get away from their various escapades. So they need to have a stretch of road built, a stretch of highway, and they're thinking about how they can possibly fund this. So with public goods, first what a public good is by definition, public goods have two traits. The first trait is non-rivalry. That is one person's use does not interfere with somebody else's use. So I think probably the best example in my mind of a non-rival good would be something like radio waves. So I can tune into a radio station in my car and that doesn't prevent somebody else from tuning into that same radio station. In fact, we could all tune into that radio station as long as we're in that area. We don't get in each other's way. We don't get congested or anything like that. So that would be the first trait. The second trait is non-excludability, which is kind of what it sounds like, that you cannot exclude someone from getting the benefit if they want it. So the problem you run into with public goods is the possibility of free-riding. So Clyde is making the decision, should he help fund this highway so that they can get away more easily? Or should he just free-ride and let somebody else, in this case Bonnie, be the one to fund it? Well, if both Bonnie and Clyde decide to fund the thing, then they can split the cost between them. And for the sake of the argument, let's just say that Clyde is perfectly willing to pay half the cost of this highway. He thinks the benefit he gets from that is more than half the cost. But he would not be willing to pay the full cost. That's too much. And Bonnie similarly has exactly the same preferences. Exactly the same preferences is, of course, realistic. So Bonnie also would say, hey, I'm willing to pay half the cost. So in that bottom right corner, we have these beautiful highways that have been funded that everybody can use. Wonderful. On the other hand, if everyone just tries to free-ride, everybody waits for somebody else to pay for the thing, and then we end up with, well, who's going to build the roads? Well, nobody does in this case. This is of course why libertarian society is impossible. It's not my position. That is what you'll hear. Who's going to fund the roads? And this is exactly the reasoning. Of course, then we have those mixed cases of, well, what if Clyde free-rides while Bonnie funds the thing? Well, in that case, Clyde clearly benefits. He gets to use the road and doesn't pay for it at all. It's pretty nice. On the other hand, Bonnie really feels like she's been taken advantage of. She had to pay the entire cost of the road, but the benefit she gets from it isn't enough to really justify paying the entire cost. So what does Clyde decide to do? Well, he says, if Bonnie is going to free-ride, I want to free-ride too, because I don't want to pay the entire cost myself. It's not worth it to me. On the other hand, if Bonnie is going to fund the thing, then I want to free-ride because I'm perfectly happy if she pays the entire cost. So in either case, Clyde wants to free-ride. Bonnie wants to free-ride. We kind of see how this goes. And what happens? Who's going to build the roads? We see no roads built at all, and it makes it very difficult to get the getaway car to, in fact, get away. Meanwhile, they both agree there's a better outcome out there. If only they could find a way to split the cost of the roads between them, then they would be provided and they would be perfectly happy to go along with this. One more example. All again, prisoner's dilemma. One more example here, a depression trap. This is one argument you hear from, say, Keynes makes an argument kind of like this, and it has to do with wages in the midst of a depression and what happens. So in this case, Bonnie and Clyde have decided to give up their life of sin and instead are going to become entrepreneurs, employers, and they have employees. And then things start not doing so well in the economy and they have a choice. Either they can increase wages or they can keep their wages low. That's where they are now. If everybody keeps their wages low, we maintain the depression. Employees don't have enough money to stimulate the economy. This is not an Austrian argument. Okay, forgive me. This is the argument you hear presented. If everybody keeps their wages low, there's no income to support consumption. The economy stays depressed. On the other hand, if everybody increases wages, now all of our workers have money. They can now go out and consume more. That stimulates all of our businesses. There's no economic expansion and everyone's happy. Now, what about those cases that's kind of mixed where, say, Clyde keeps his wages low while Bonnie increases hers? In that case, Clyde, by keeping his wages low, gets to keep his costs down. If Bonnie's increasing her wages, her employees start spending more and Clyde gets to benefit from that. He's selling to them. So there's a profit then for Clyde, but Bonnie, on the other hand, takes a loss. She has to pay more in wages, but she's not getting any of the benefit from Bonnie increasing. So Clyde facing the situation says, well, if Bonnie's going to keep the wages low, I don't want to take the loss by increasing my wages. I'm going to keep them low. If Bonnie increases wages, I'm very happy to make the profit. So, either way, I'm going to keep wages low. And we end up stuck in this depression that we cannot escape from unless there is some way to get everybody to increase wages at the same time. So the typical conclusions we come to here, we, not including me, they come to here, is that in prisoners' dilemmas, we see that people are choosing optimally from an individual perspective. They very clearly are doing the right thing individually. At the same time, we're getting bad outcomes that everyone involved agrees is bad from a social perspective. So the obvious solution, we need the government to step in and solve these problems. Have the government step in and push up minimum wages so that all the wages rise that leads to an economic expansion and I occasionally have students that write papers about the minimum wage and they will often make that argument. And then I kind of feel like a failure. It's an argument that you hear. So we need the government to step in, push minimum wages up, that would allow us then to break out of this depression, lead to economic expansion. We need the government to step in and fund public goods with taxes. After all, all we have to do is make Bonnie and Clyde each pay half the cost of the highway and build it and they agree this is better than what happened. And we just regulate the use of the commons. So decide who gets to use it or how much, to ensure that we don't end up with desolation. So regulate how much fishing people can do, only allow certain amounts or sizes or what have you. And of course, we need to prevent prisoners from confessing. Okay, you don't often hear that particular argument. We might acknowledge that there's another player here that might possibly matter, society as a whole or something like that. All right, so this is the way we typically will see prisoners dumb as used by those that would talk about game theory or present game theory. So I want to question this. By bringing in concepts related to the Austrian view of entrepreneurship. So first, think about Mises' view of entrepreneurs as being ones that have good foresight. So according to Mises, as Dr. Klein mentioned earlier this week, Mises' view really what makes an entrepreneur an entrepreneur is that they're acting under uncertainty. So to be then a successful entrepreneur, you're the one that has keen foresight. You're better at seeing through that uncertainty to what is going to happen than others are. All right, so successful entrepreneurs then have this trait of having very keen foresight. And then we also have the Kersnerian view that it's about entrepreneurial alertness. That is that we see those opportunities that exist and we seize those opportunities. So it's just having good eyesight or this alertness. So how would this then apply in a world of game theory? All right, let's revisit the prisoner's dilemma. So I would suggest that those with this entrepreneurial mindset that are alert, that have good foresight, they would not just declare the game lost and say the only way to win is not to play. Instead, they would change the game. After all, it's perfectly possible if we are in fact rubbing banks together that we might possibly think that at some point we'll get caught. So we plan for this. How can we set things up so that when we get into that situation that we actually end up with a better outcome, that we all agree is better? How can we change the game to make things better? And in fact, the reality is that game theorists and I'm talking about mainstream game theorists already know that there are a number of solutions to the prisoner's dilemma. There are ways we can tweak the game here and there to improve the outcome. So one example is if you just repeat a prisoner's dilemma multiple times, then that allows us to use punishment and reward strategies in later rounds to enforce better outcomes in early rounds. So we could play one of the most famous of these strategies is Tit for Tat, which is kind of a very warped version of the Golden Rule. The Golden Rule is doing to others as they would do to you. No, that's not it, is it? As you want them to do to you. This is inspect. Tit for Tat is doing to others as they have done to you. So what happens here? Okay, we are partners in crime. We think we're probably going to get caught more than once. So here's what we're going to do. If you rat me out that first time, I'll rat you out the second. This possibility of punishment then changes our behavior in that first round so we can end up with much better outcomes. So this is not something specifically Austrian. This is something where I went to a very mainstream graduate school and were taught about this. This is something that we know exists as a possibility. It's somehow we don't make the connection to maybe people are actually creative and will think of this themselves. It's only those of us in the ivory tower that think of these things. We get to watch the rest of humanity make bad choices until we force them to do otherwise. So let's re-envision Game 3 with this in mind. So the key assumption that sneakily was underlying all of the stuff we did in those four examples of the prisoner's dilemma was that the game is fixed. Of course, in reality, if we have entrepreneurial players that foresee the game coming, this tends not to be the case. We find ways to rewrite the rules of the game to make it more advantageous potentially. So what we need to do then, if we're going to use Game 3 what I would argue as well, we need to ask the right question. The right question is not, what's the Nash Equilibrium? The right question is, how might entrepreneurs try to improve the outcome of the game if it is possible? And if they're not doing that, if we do actually see the Nash Equilibrium happen, why aren't they doing it? What is it that's keeping them from being able to achieve what we all agreed to be a better outcome? Now, it might be that there's some hidden costs involved with that outcome that we don't notice. So in fact, that outcome isn't as good as we think. That can happen. After all, if we believe in subjective preference, I'm not the one playing the game. They know more about other preferences than I do. Or maybe there's some other rule that's preventing them. There's something in society as a whole that's not allowing them to fix the game. I would also want to point out, just for a moment, I mentioned before that, so you will sometimes hear Friedman's axiom, the test of a model is its ability to predict. On that basis, the prisoner's dilemma should be just rejected. Because I have run this as an experiment in many of my principal's level classes. The number of times that I've seen them actually, my students actually choose the bad outcome that would be predicted by Nash Equilibrium is extremely small. Virtually all the time, they find some other outcome. Which is wild, right? No, I'm not threatening them with prison. So I might need to up the stakes a bit. That might be a problem. I'm tempting them with bars of candy or something like that. But I love doing this because we play the game before we describe what the outcome is supposed to be. So students don't know what they're supposed to do. And then when they inevitably play the game wrong, then we can start doing the survey. So, okay, why didn't you do what you were supposed to do? Now, sometimes the students say, oh, I just misunderstood how to read the table. So you, in fact, would have betrayed your partner if you knew how to do it. But a lot of the time it's things like, well, actually I don't actually like chocolate that well. Okay, what happens if our prisoners aren't actually that concerned about prison time? Okay, now five years in prison may not be that bad. Okay, so we need to make sure we get things right. And if we're seeing where things are going wrong, again, there's something wrong with the theory. We're missing something, some fact about reality. Okay, so I suggest that these are the questions we need to ask. First, how would entrepreneurs try to improve the game assuming that we actually have the structure correct? And then why aren't they doing it if they're not? Okay, so I want to give some examples of using game theory wrong and then look at how we could potentially use game theory right. I suppose I'm reversing what Hayek's order suggests we should do. Well, so first, using game theory wrong, we can look at all those previous examples. The depression wages or forcing wages up or solving the commons, the tragedy of the commons or public goods problems. These are the ways we would use game theory wrong. We just take as given, this is what the game is, this is what the outcome would be. To fix the game we need to have some outside force come in. And I would suggest that Austrians are not immune to this. When you do look through the Austrian literature, it does seem that we do agree with Rothbard that part of praxeology is game theory and we do occasionally appeal to it. So there's this one paper by Krillian Dempster from 22 years ago now, which I have a paper about how monetary expansion affects the quality of entrepreneurs. I did not originally respond to their paper when I wrote this and sent it to the quarterly Journal of Austrian Economics, but one of the reviewers of that paper said, oh, you really need to deal with the Krillian Dempster paper. I thought, wow, I need to find out what that paper is. So I went and I found it and I found their argument. I was not satisfied with it. So eventually the associate editor of the quarterly Journal of Austrian Economics at the time started referring to my paper as my Krillian Dempster paper. It became such a big part of that paper was dealing with their arguments. All right, let's take a score one for peer review there. All right, so here's the Krillian Dempster argument. They're trying to explain why is it that entrepreneurs would invest in a business cycle causing the boom-bust cycle. After all, the idea here is if entrepreneurs are so good, this is what we call the rational expectations critique of Austrian business cycle theory. So Austrians love entrepreneurs. We think they're very good, have good foresight, that kind of thing. So if entrepreneurs are so great, they'll understand that investing amidst the boom phase, they're leading into the bust phase. They know the bust is coming. Why would you do this? Why would you invest in something that you know isn't going to last, is going to fall apart? Well, one of the arguments that's made is the one that Krillian Dempster make, where they suggest it's really a game theoretic problem. So imagine here that we have two firms, well, I guess two sets of firms, we're going to focus on one firm that is firm X, and then we have all other firms are also making similar choices. And your choices in the midst of the early phases of the boom or interest rates have been pushed down by credit expansion, do you decide to increase your investment to maintain its previous level? So what's going to happen? Well, the argument is, for firm X's point of view, if they see other firms increase investment, and they also increase investment, then their profits are going to be basically the same as all the other firms. So relative profits between firms will be equal, but then we also see the boom-bust cycle. Everybody's increasing investment, creating that of the boom phase, which must inevitably collapse. On the other hand, if they maintain investment in this case, then they're going to find that at least in the short run, their profits are less than the other firm's profits. They're not taking advantage of these low interest rates that they could in fact be taking advantage of to improve their profit at least at the moment. So in the case where others are increasing investment, firm X will want to increase investment as well, so they don't fall behind the other firms. Meanwhile, if all other firms just maintain their investment, firm X now has an opportunity to take advantage of these low interest rates to enhance its own profit, whereas if it just maintained investment too, it would just earn the same profit as everybody else. So better to outshine your competitors. So again, firm X would decide to increase investment. So it doesn't really matter what the other firms do. Firm X gets pulled into increasing investment, as do all of the other firms, resulting then in the boom-bust cycle. We should recognize the structure of this game. It has three elements. One, Nash Equilibrium, dominant strategies that Nash Equilibrium is lousy. We have relative profits equal and a boom-bust cycle, versus relative profits equal, no boom-bust cycle. I suggest probably most firms would rather not have boom-bust cycles, which means this outcome that we have is lousy. There's another outcome out there that is clearly better than this one. So it's a prisoner's dilemma, and we're using the standard prisoner's dilemma kind of reasoning. So how would I criticize this particular usage of game theory? I actually have a couple points. So the first is that the short-term perspective is kind of weird, and that we're really only looking at the short-run profits that we can make during the boom phase forgetting that the bust phase is going to come. That's kind of odd. Secondly, the emphasis on relative profit is kind of strange. It's like I'm only worried about how my profit compares to the people next to me. It's kind of a keeping up with the Jones's theory of profit maximization, which is kind of odd to me. And then the last point, which I think is the more fundamental for our talk here, is that we're not asking the right question. The question we need to ask is, why don't firms organize to prevent that increased investment? Why don't they find some kind of solution that is going to push them toward the better outcome? And that's the question we need to answer, or why doesn't this happen? And that is a question that, at least in my reading of their paper, I didn't see an answer to why they couldn't organize to fix this problem. If you want to know my answer to that, read my paper. It has to do with entrepreneurial quality. Now about using game theory right. So we've seen how game theory can go wrong and often does go wrong, specifically with the use of the prisoner's dilemma. But how can it go right? What do I have in mind for how game theory should work? So let's take a look. So I suggest that game theory will inform when entrepreneurs are going to seek creative, institutional solutions to nasty problems. So the prisoner's dilemma is a nasty problem. It's not immediately obvious what can be done. But entrepreneurs do things that aren't immediately obvious. That is part of their traits. They have good foresight. They're alert to what they could potentially do to solve nasty problems. That's why they earn a profit. So that being the case, I would suggest that all we can use is game theory to figure out when they're going to look for solutions. Even if we can't predict the solutions themselves, we at least know this is a situation where somebody will start looking for a solution. So one of my favorite examples where I saw entrepreneurship in action. This was several years ago. There's a British game show called Golden Balls. And the way this game show played, you had a couple players that made choices throughout the show and they won some pot of money. And then at the end they played a game called Split or Steel. So Split or Steel involved the two players that had the choice. They had two Golden Balls in front of them. Inside of one of them it said Steel. Instead of the other it said Split. And they could choose which of these balls they wanted to choose as their last play of the game. And here was what their payoff structure was. So in this particular case it was Nick and Ibrahim were the two people at the end of the game. And the rules of the game said that if both players play Steel then everybody walks away empty handed. At which point the winner is the BBC. They saved a lot of prize money. On the other hand, if both of you play Split then you take it as 50,000 pounds that particular episode that you split the 50,000 pounds evenly so you each walk away with 25,000 pounds. Not bad. On the other hand, if one plays Steel while the other plays Split then whoever plays Steel walks away with all the money. Well whoever played Split walks away with nothing. So here we have that structure just laid out right inside of a table format. So if we go through and we analyze how would we expect our players to behave in this situation. Well first let's analyze this from Nick's perspective. I like to always do whatever player is choosing the row because that's how I remember it. Alright so Nick says, well if Ibrahim plays Steel what I do literally does not matter. Because if Ibrahim plays Steel and I play Steel I walk away with nothing. If I play Split I walk away with nothing. So it really doesn't matter in that case. On the other hand if Ibrahim plays Split and I play Steel then I get 50,000 pounds. Or as if I play Split I have 25,000 pounds. So there is some reason then to pick Steel over Split. Now a note here this is not strictly speaking a prisoner's dilemma as choosing Steel is not strictly a dominant strategy since in the case where Ibrahim plays Steel it doesn't matter what Nick does. It's still somewhat like I guess we'd call it a weakly dominant strategy. It's not obviously the best. Sometimes it's okay but oh well. Alright meanwhile Ibrahim faces exactly the same incentives has a reason to Steel, no reason to play Split so what we should predict is that the BBC will win the game. Both of the players walk away with nothing. Well if they both play Steel. But here this episode Nick was very entrepreneurial and he did something really bizarre. So in this game one other distinction that makes this different from the typical prisoner's dilemma is that they let the two players talk to each other about what their strategy is going to be. Now normally this is just kind of a lot of show boating of oh you know I'm a trustworthy person I would never betray you that kind of thing what game theory is called cheap talk. That's normally what you see. Nick decided to look like a maniac. So here's what he did. He said I'm going to change the game. So I'm telling you right now Ibrahim I am going to choose Steel. But after I choose Steel I will take that 50,000 pounds that I've won and I'll split it with you. I'll give you 25,000. And Ibrahim I believe literally said you're insane or something similar to that you're totally nuts. And Nick said no I'm a trustworthy person and I will play Steel. What a bizarre thing. But look at what he did. He just changed the circumstances of the game. So now we think about alright if we think Nick is in fact this crazy that he's willing to go on national television and say that he's willing to steal from the person he's playing with then what should I do if I'm Ibrahim? Well Ibrahim thinks through it. He says well if I believe him then I really do need to play Split. This is something that wasn't true before. He never had a reason to play Split before. Now Nick created a reason for Ibrahim to maybe play Split. Now it's still true he might think well okay Nick is crazy I might just pick Split just for I don't know for fun in that case I still want to play Steel. So Ibrahim now is caught in this dilemma this is much more dilemma than the prisoners faced what do I do do I believe Nick in which case I definitely want to play Split or do I think he's lying which would be a bizarre choice to lie about being a bad person in which case I want to play Steel. Alright so you can watch his face I really recommend looking at this video he's so torn until finally he commits it's okay I'm going to go ahead and pick Split. So this is the outcome that we are then promised Nick says I'm going to play Steel Ibrahim picks Split it turns out Nick is a liar Nick played Split well go back and revisit the game after he has changed it what are his incentives now? Nick just made it so it literally doesn't matter what he plays as long as he's honest about the sharing it doesn't matter what he picks anymore so you may as well go ahead and pick Split you're going to split it anyway and then you save yourself having to write the check after the game or something like that so we end up then with a better outcome so what we would have predicted 0 for each isn't what happened we got 25,000 for each because Nick was entrepreneurial he knew how the game worked he had a plan for what he was going to do and he changed the way the game was going to be played another example which I think is much more economicy would be Jesus Huerta de Soto money bank credit and economic cycles talks about credit expansion how do we explain how credit expansion works and he imagines a case like this where we have bank A and bank B where each of them have the choice to either expand credit or not to expand credit and here are the results so if neither of them expands credit so they operate on a 100% reserve basis or something like that then we end up with both of them surviving and both of them earning a relatively small profit the bank fees that they can charge from depositors and that kind of thing on the other hand if both of them expand credit then we end up also both are surviving and they're in a larger profit as they now also make money off of the lending market now where we end up with trouble is when one expands and the other doesn't in that case the one that expands ends up losing a bunch of reserves as Dr. Patrick Newpin mentioned earlier this week and ends up failing so whoever the expander is in this mixed situation ends up failing while the other bank survives and so then what are the optimal choices? Well bank A says well if bank B doesn't expand then I don't want to either because if I do expand I'm going to fail and failure is not good for a bank on the other hand if bank B expands then I want to expand because earning a large profit is better than just surviving so we end up what bank A wants to do depends on what bank B wants to do and bank B faces the same situation here this is not a prisoner's dilemma in fact I believe fails on the first two traits we don't have just one Nash equilibrium we actually have two the first one that is where nobody expands the other where everybody expands we call this a coordination game other games that play out this way are things like which side of the road do you drive on it doesn't actually matter you can drive on the right, you can drive on the left as long as you know what everyone else is doing so in Alabama drive on the right in London drive on the left neither is necessarily better than the other so this is a coordination game what we want to do is get everybody on the same page is ultimately the goal and that is what we end up with here so what is then how does quarter disorder use this is conclusion is that prudent bankers would not want to expand because of this fear of failure so we would have if we have a banking system that is governed by prudence we would not see that much credit expansion in fact we would see very little if any meanwhile imprudent bankers would be the ones saying I'm willing to risk the failure for those big profits but every bank does have an incentive to try to coordinate to make credit expansion happen why go back again we can either both survive with a small profit or a large profit so between these two Nash equilibria one is actually preferable so this isn't quite like left side of the road right side of the road it's better okay that is like right side of the road left side of the road it's called the right side of the road people I'm an American alright so what does that mean bankers then want this credit expansion to happen if we can make sure everybody else is involved so we need to change the game to get rid of that upper left outcome where there's this reason to be scared of credit expansion and this is why he suggests bankers will push for central banks this is how bankers change the game so what does he suggest he suggests the game looks something like this so we change the game this way that is we have a central bank that is going to bail out anybody who's in danger of failure so the result then is that for bank A when they think about what's going to happen they say well if bank B doesn't expand and I don't expand I get a small profit if I do expand I get bailed out and I still get a small profit there's only no reason not to expand on the other hand if bank B does expand there's definitely a reason to expand because that large profit is possible meanwhile bank B goes through the same line of reasoning they're not really worried about failure anymore the central bank will bail them out to prevent that failure from happening so therefore all they have to worry about is the possibility of that large profit so what do we just do we just eliminated the less desirable Nash Equilibrium and now we're down to just one the more desirable one from an extra-economist perspective and this I would suggest is a good way of using game theory because what did we do we laid out the situation we said here's where we would have problems potentially and here's how entrepreneurial banks would try to solve this situation in this case they'd appeal to the political system to enforce the outcome that they like better so there are other possible solutions when we look at some of these other games so things like a general prisoners dilemma I already mentioned repeated games are a way that we can solve this there's also perhaps the use of enforcers I like to call this breaking the confessor's kneecaps okay I didn't go to prison but it turned out there's a third person involved in our criminal enterprise who's going to take my knees out when I leave the police station that is going to change the outcome most likely and then the tragedy of the common so one thing this is something this is not specifically Austrian just privatize the common so that tends to solve this problem if it is my field I don't want it to devolve into a mud pit so break the thing up give each farmer an area of the commons and they were going to tend to try to manage it well we also have Eleanor Ostrom's work on managing the commons where she went through and looked at cases it was one of the odd things about the tragedy of the commons is that you feel like the conditions should apply a lot of the time but yet we don't see a bunch of mud pits around so why is that is kind of what Ostrom suggests this is one of those cases where the theory is very strong and also very wrong didn't predict what actually happened and what she found was that communities are actually pretty good at managing the commons if we recognize she found some conditions that are necessary and they're not like the conditions of many of these highfalutin mathematical theories to be correct these are things that we do actually see happen and can believe happening we have a relatively well defined community of people using the resource small fishing village in Nepal it's believable we know who all the fishers are going to be we have a resource that we all recognize is important well yes we're all eating fish every day we all recognize that's important and then if we have some way to come up with rules that we can then enforce these are not hard things to have happen and what Ostrom suggests is that that's why we see that actually happen instead of mud pits and then with public goods we have options, things like matching funds as an opportunity so here what I mean by matching funds I think thinking about Kickstarter is probably a good way to get this so for those who are unfamiliar with Kickstarter I don't know if there's anybody that would be unfamiliar with Kickstarter at this point but what it is it's a contingent payment system so people come up with some kind of project that they want to do and then they put it up on the website and then you can make pledges to make sure this project happens it's often used for creative types of activities like making a documentary about something but I need a bunch of money to make that happen so then you ask people to pledge and normally those that pledge get some kind of benefit it might be you get your name put in the credits or maybe you get if it's a book a copy of the book maybe a signed copy if you pledge more that kind of thing and the way it works is that if whatever the project is doesn't get fully funded then nobody has to pay so it's this nice contingent payment system which allows me to say if this happens I am willing to pay $40 for this book if it doesn't happen I wouldn't be willing to but I can pledge the 40 and then we'll see what happens and that is something that we've seen be very successful at funding would often be considered public goods Reading Rainbow I like to talk about here I remember this from my childhood because I remember the 80s I was a show encouraging literacy and it was on PBS funded then largely by taxes and also by donors well they decided they wanted to try to make kind of new content available to classrooms so LaVar Burton started this Kickstarter program to try to get this out into classrooms and raised $5 million on Kickstarter people were willing to chip in to make this project happen and this didn't require that we get some grant from the Corporation for Public Broadcasting or something like that people were willing to chip in to make it happen as long as they knew others would and Kickstarter was the way of guaranteeing that so this is another case where we can see an entrepreneurial solution that has now been institutionalized by a website if you want to fund a project use Kickstarter and you can see whether people are willing to fund you or not or in my case if you want to fund the project up on and offer to do it so wrapping up then I want to suggest or remind you there are two ways we can think of game theory we can use it the wrong way saying things like all we're going to end up with lousy outcomes from individual choice so the obvious solution is to eliminate individual choice or we can use it the right way and point out that lousy outcomes attract entrepreneurs so we need to ask what are they doing to fix these problems and if they're not why not what's getting in their way to keep them from fixing these problems thank you