How game theory and competition explain politics | Interview with Dr. Bruce Bueno de Mesquita

00:37Why political science?

Jed Macosko: Hi. I’m Dr. Jed Macosko at AcademicInfluence.com and Wake Forest University. And today, we have another special guest with us and this is now Dr. Bruce Bueno de Mesquita, who will be talking a little bit about how he got interested in his field and how he has become one of the most influential people in that field. So go ahead, Professor Bueno de Mesquita.

Bruce Bueno De Mesquita: Hi. Thanks for having me on. Well, I originally got interested in studying politics for no good reason at all. I took a summer vacation when I was 17 or 18, and during that time in my travels, I ran in repeatedly to a fellow from Pakistan who was a professor of politics in Pakistan and who regaled me with the wonders of Pakistan and the evils of India, as he saw it.

The two countries then fought a war, I got intrigued and as I got intrigued, I, first of all, figured out that India was not inherently the evil side. They each had a case to be made. And so I decided I would study the subject. I had experimented with Chemistry, which I found boring, with Economics and with Literature and so forth.

In any event, once I got into studying politics, I then read a book called The Theory of Political Coalitions, that was perhaps the first game theoretic mathematical model of politics that had been done or certainly one of the first. And I was blown away by the realization that there was an error in one of the main three results in that book, a theorem, and that I could say it was wrong, not as a matter of opinion but as purely a matter of logic. And that was just, to me, incredibly exciting to realize that you could talk about politics in a rigorous, theoretically driven way that wasn’t about partisanship, it wasn’t about personal opinions, it wasn’t about the way you wished things were. And that got me off and running, and I guess I’ve never turned back.

Jed: So you say that you found a logical error in what you were reading in one of the three main points.

How old were you at the time when you found that out?

Bruce: Oh, I was probably, I don’t know, about 19.

Jed: Wow.

And were you already at the university?

Bruce: Oh, yeah. Yeah. I started college early.

Jed: Okay. So you were learning all this stuff about economics and chemistry, but it was really political science that got you most interested and it was probably because you could see how, even as a 19-year-old, you could make a huge difference in what people were saying.

Was that how it all unfolded?

Bruce: Well, I didn’t really think about what I could do in terms of having any impact in a field. I didn’t really know enough to know what that would have meant. What I felt was that this was a subject that had not been studied in a really, in my point of view, good way and it could be. And I’d been exposed to something that was doing it right and that was very exciting. It was a new way and I thought, “I’d like to be part of that.”

Jed: And so, is that what you ended up doing? Is going into the game theory side of political science?

Bruce: Yeah. And this was a very primitive game theory. It was what’s called cooperative game theory. The thing that was key for me is that once I’d got even a little bit of exposure to game theory, I couldn’t imagine how you could possibly study politics or any social phenomenon without studying game theory, because game theory is about how people interact when they’re being strategic, taking into account what they think other people are going to do if they do this or that. It’s like playing chess. And that’s what almost all social phenomena were about. And so once I saw that and I saw how politics and other subjects had been studied up to that point, I said, “Well, it doesn’t make sense. This makes sense.” And so I just was tremendously excited. It’s a long time ago, I’m still tremendously excited.

[chuckle]

05:28National interest?

Jed: It’s carried you through this far. So for those of us who don’t even know what a strategic particle is, tell us a little bit about how seeing political science through the lens of these strategic things, the strategies that go on, changes everything and got you so excited.

Bruce: Okay. So, for example, people, when they talk about foreign affairs, which is a subject I teach, they talk about the national interest. And so when I teach about international politics, I ask my students what they think is meant by the national interest? And maybe I’ll ask you that, “What do you think it means to say the national interest?”

Jed: Yeah. I would just say that it’s in the nation’s best interest and it might include things like the price of oil. If the price of oil is low, that’s in our nation’s best interest because we buy most of our oil from overseas or maybe if it goes high, it’s in our best interest because now we are fracking and producing petroleum products right here in the United States. So depending on the year, what’s in our best national interest might differ. Is that right or am I totally off?

Bruce: No, that’s completely wrong. Sorry.

[laughter]

Because the notion that there is a nation that has interest is incoherent, people have interests and one of the things that we know about adding up people’s interests, aggregating them is something called Arrow’s Impossibility Theorem. And in a nutshell, what Arrow... Later won the Nobel Prize in part for this work, died a year ago, unfortunately. What Arrow worked out was that under a set of reasonable assumptions about what constitutes fairness, it is impossible to construct a set of rules for adding up people’s preferences that satisfy the conditions.

So for example, you might think that low oil prices is “In the national interest” and you might think that’s better than higher prices with fracking, but worse than lower prices with fracking.

And I might have a different ordering on those things, and when we... If we have at least three choices and at least three people making the choice, then it is entirely possible to have a cycle to violate transitivity. We as a group, like A better than B, and B better than C, and C better than A, even though we each individually have a coherent ordering so what we...

Jed: So it’s like rock-paper-scissors, just going on an endless loop. Okay, interesting.

Bruce: Yeah, exactly. Or as I put it to my students, is it possible? I’ll put it to you. Is it possible, without tricking anybody, without changing anybody’s mind, for two-thirds of say the American public to want to spend more money on defense, and for two-thirds of the same American public to wanna spend less on defense, and you would naturally be inclined to say, “That’s not possible.”

Jed: That doesn’t add up, it doesn’t seem like it’s right. [chuckle]

Bruce: And yet it does, and it’s easily shown. If we add a second dimension and we link two issues, for example, how much to spend on defense and should we have a freer trade or protectionist trade, fairer trade or some other dimension. And we have three groups or more making the choice, then we have, each has a position on the two axes, there’s a status quo somewhere in between them, then there exist, if we draw an arc through the status quo that is centered on each of the three group’s preferences, and we get overlapping petals, overlapping portions of the arcs where two-thirds wanna spend less on defense, two-thirds wanna spend more on defense, two-thirds want to have free trade, two-thirds want to have fair trade without tricking anybody.

Jed: Interesting.

Bruce: So then what’s the national interest? It is however, the competing politicians succeed in defining the agenda over which the competition takes place.

We know that there... And we have some theorems that prove there is a rational path from any point in that two-dimensional issue space to any other. And in fact, as long as there is at least one more chooser than dimensions, there’s a rational point to anywhere in the space for any number of dimensions.

Jed: Wow.

Bruce: So then what’s the national interest?

Jed: I’m totally befuddled now. [chuckle]

So once you confuse your students like you just confused me, where do you go from there? How do you bring them back home?

Bruce: So now we go home by talking about strategy. So how do politicians construct, for example, winning coalitions? How do they manipulate those issue dimensions to produce outcomes that favor them instead of outcomes that favor their opponent? That means they have to think about not only what they want, but what their opponent is likely to do and what is the best counter-move to what they think the opponent will do, knowing that the opponent is thinking about what is the best counter-move to what they’re going to do.

Jed: Wow. So it really is like a chess game, like you were saying earlier.

Bruce: Yeah. It’s multi-dimensional chess often with many players.

11:43First two contributions

Jed: Interesting. So this is what fascinates you about political science. This is what has kept you enthralled in the field all these years, and [chuckle] I can see why. It is like the most intense game and it’s in real, with real stakes, with real people and real issues.

Bruce: Yes, exactly.

Jed: So what do you add to this game? What have you added... Have you helped people understand how the game is played? What kinds of things have you done?

Bruce: So I feel I’ve added two important things and I’m hoping the book I’m just finishing will add a third important thing.

So over the years, I worked out a dynamic game theory model for any number, any finite number of players that allows me for any specific issue to predict how it will come out in the future, how it’ll be resolved, and to figure out how to engineer that outcome, how to shape it.

Jed: Wow.

Bruce: And so in my academic life, I’ve written a lot about that. In my consulting life and so forth, I’ve helped, for example, previous US governments to solve some fundamental puzzles. I’m being modest or whatever, I’m very proud of the fact that I did a... Different people have different points of view about this, but I did a study for the US Government in 2007 that persuaded the intelligence community that Iran was in fact not trying to build a nuclear weapon and was unlikely at that point over the next, it was five or 10-year span, was unlikely to build a nuclear weapon.

That was then made more broadly known by the intelligence community writing a National Intelligence Estimate, which was released to the New York Times and tied George W. Bush ’s hands from potentially bombing the Natanz Nuclear Facility and beginning a war. So I thought, well, that was a pretty good thing.

Jed: Wow. Yeah. That’s really good.

Bruce: And I’ve had opportunities to work on other very high profile foreign policy problems where using this model, I was able correctly to anticipate and reshape outcomes on major events. So that was something, and the idea...

Jed: That’s great.

Bruce: In starting to do that was to show that you could contribute to politics and to reshaping policy, not based on your opinion and the reason that the government trusted my work is because they knew I wasn’t going to give an opinion. Every word I would utter in a briefing was driven by what I could point to in the output from this Game Theoretic model.

And then with some colleagues a number of years later, with Alastair Smith, former student of mine, and my colleague at New York University, James Morrow also former student of mine and professor at the University of Michigan, and Randolph Siverson, a friend of mine, not a former student. He’s older than me, recently retired from UC Davis. We developed something that’s known as the Selectorate theory, which is an attempt at a fairly general theory of politics that starts from the perspective that we shouldn’t be talking about regimes as autocracy or democracy, that these are essentially meaningless terms. They’re meaningless in the sense that if you ask me, which is more democratic; France, England or the United States. How would I answer you? And if you ask me, which is more autocratic; Saudi Arabia, North Korea or whatever, how would I answer you?

So we developed a theory, which has now expanded to a third dimension, that was... Where every organization, and therefore every government, could be described in two dimensions, how many people had a say in choosing leaders, and how many of those people’s support was essential to keep a leader in power? And from this very... And where we started with the premise that what leaders care about is surviving in power. They don’t care about the national interest. They don’t care about the welfare of this or that citizen. They care about staying in power and conditional on succeeding at that, on having as much discretionary control over money, over revenue as possible.

So in a Game Theoretic setting, we built from that to prove why... That if you’re a small coalition, what we call, the winning coalition in Selectorate, you’re small coalition, large selectorate government, then you’ll have much more discretionary money. So there’ll be more kleptocracy. If the coalition is small, you have to concentrate spending on private benefits to the coalition members rather than public goods. If the coalition is large, it’s too expensive to provide a lot of private goods, so you shift to public goods. So we showed that corruption depended on the size of the winning coalition.

Jed: Wow.

Bruce: Kleptocracy depended on the size of the coalition and the selectorate. The welfare of members of the coalition has a functional form in the theory that looks like the Nike swoosh that is when the coalition’s very small, their welfare is high. As the coalition gets bigger, their welfare drops. As it continues to get bigger, their welfare becomes higher...

Jed: Wow.

Bruce: And eventually becomes greater than the maximum, the local maximum when it’s small. Two things were going on simultaneously as the coalition got bigger, holding the selectorate constant, private goods were decreasing, reducing their welfare, but total spending on keeping the coalition loyal had to be increasing. So the pie from which their goodies were being drawn was getting bigger. So they were getting a smaller proportion of that pie as private goods, but the pie was bigger. And so that smaller proportion could have been a bigger quantity.

And so this turned out... For example, the World Bank has used this in making decisions about loans to countries, to look at the details of their governance in these terms to work out who is likely to default, who is not likely to default, and so on and so forth. And others have used this to calculate what the different sorts of political risks are in doing business with this country or that country based on the structure of these institutions, where you could then answer the question, “Well, how big is the winning coalition in Britain compared to France?” Well, that’s based on the particular rules they use. And so you can say which one is going to have more public policy orientation, which is gonna have more of a corruption orientation, even among countries that are basically honest.

Jed: Right.

Bruce: Likewise, you could make these statements about North Korea and Saudi Arabia and what have you for the same reason, because now you... We’re not talking about types of governments, but a continuous space where the even subtle differences have direct predictions about actions and in how they conduct foreign policy, how they conduct education, how they conduct healthcare, how they conduct immigration, emigration, across the board, pretty much almost every important political subject.

Jed: Wow that is fascinating.

Bruce: That’s really cool.

20:27The third contribution

Jed: Yeah, two really big contributions to the field, and now you say that there’s a third one coming out.

What’s that one are gonna be about?

Bruce: Yeah, so the third one is quite esoteric, I’m finishing a book called Popes Versus Kings, the creation of...

Jed: Interesting.

Bruce: Of Western exceptionalism, in which I’m using game theory to explain why some parts of Europe are more prosperous than others, why some parts of Europe were more aggressively involved in the scientific revolution than others, why some parts...

Jed: Those were the kings. Right, and then the rest of them were the popes right?

Bruce: Yeah. Yeah. Yeah.

Jed: I can see where you’re going with this.

Bruce: So the thesis is that contrary to the standard views, what was important was three deals, an agreement called the Concordat of Worms signed in 1122 and two predecessors to that in France and England in 1107, that just said says, “Popes get... Or the church gets to pick Bishops and kings get to say, I agree to that nominee or I don’t.” And what the concordats did is that they then added... And if the king doesn’t agree, then the income from the diocese in question goes to the secular ruler until a bishop is agreed to.

Before that it had continued in the church’s hands, whether there was a bishop in place or not. This completely changed incentives, this incentivized secular rulers to secularize, and in order to... That is to get bishops who would be loyal to them rather than to the church. And the way they could do that was only if the diocese in question were wealthy enough that the Pope wouldn’t want to lose the money. And so would offer a bishop more to the liking of the king than the other way around, so as to get the income. And that meant that kings had an incentive to stimulate economic growth, whereas before...

Jed: Wow.

Bruce: They made money by waging war or marrying the right person, now they could make... They could be rich by stimulating the economy, which they had not tried to do before. And the church had the opposite incentive now. They didn’t have it before, once these deals were in place, where they were in the diocese that were subject to these rules, the church now had an incentive to limit economic growth, because of a diocese got too wealthy, then the church lost power over it, because the Bishop became an agent of the king instead of an agent of the church.

And so I can show with evidence that in dioceses where the bishop was expected to be loyal to the church, that was a... Before becoming a Bishop was a priest, a monk, and so on, that economic growth was less than half of what it was in the diocese that bishops whose prior jobs had been working for the king, the king’s chancellor, his ambassador, his teacher, whatever. That was not true before the agreements were signed.

Jed: Wow.

Bruce: I can show it was not true before. And then they saw the strategic incentives to stimulate growth and to stymie growth, they each acted on it.

Jed: The strategic particle right there.

Bruce: Yeah the strategic particles. And as it turns out, if you look at the difference in per capita income in Europe, across Europe today, and you look at the prevalence of secular or religiously-oriented bishops in this period, you’ll almost perfectly predict the difference...

Jed: Wow. Unbelievable.

Bruce: In per capita income today. So that’s pretty cool too.

Jed: Oh my god.

Bruce: That was a lot of fun.

24:28This next book

Jed: That is so cool. But it is esoteric because the first two things you mentioned have ramifications in here and now decisions that are being made, whereas this new book is just explaining the past, it doesn’t really...

Bruce: Not really, so the last chapter...

Jed: Okay, well tell us, tell us about it.

Bruce: The last chapter, which is called today, zooms in on stripping out the religion aspects and monarchy aspects of... What was happening was there was competition put in place between the secular and the religious domains between two power centers, and so the last chapter focuses on how regulated, that’s what the concordats did was regulate that competition, how regulated competition can produce better social outcomes. And that’s pertinent to choices today. And that’s what the last chapter is about.

Jed: So it’s basically a lesson from history, that the last chapter is saying, “Let’s take a lesson from history about what worked and let’s apply it to today.” Is that what you’re saying?

Bruce: Yeah, I’m saying something slightly different from that, I’m saying take a lesson, use history as the source of data to extract a lesson about what sorts of competition produces, what most of us would think of as good outcomes and what sorts of competition doesn’t. And let’s see if we can steer people strategically in the direction of the effective forms of competition. Which we can learn from history, but I don’t see it primarily as a lesson of history, rather. History as the data source.

Jed: And you really boil down what happened back in the past, into it’s just essential strategic particles that we can then harvest for our use today so it will look anything like the past, but it’s essential.

Bruce: With the important caveat that we’re not just looking at what happened in the past and boiling it down. We are taking the logic of these agreements, as strategic documents and looking at how they changed incentives. So the focus is on how different incentive structures produce different results. And so it’s not just...

Jed: Yeah. That’s an important distinction. We’re not just looking at the past, looking at it through the lenses of somebody like you who from a 19-year-old was seeing these logical connections, where other people couldn’t see them. So we have to look at it through those lenses, is what you’re saying.

Bruce: No, I wouldn’t wanna go so far as to say I saw things that other people didn’t see at 19. But historians have missed this version of the impact of the...

27:24COVID

Jed: Yeah. I’ve never heard that before. So that is really really fascinating. Well, I’d be remiss if I didn’t ask you your opinion about how COVID-19 is changing everything, because it is such an upheaval across the entire globe.

And through your eyes, I wanna know what are gonna be the long term effects of this? What about the countries that have done well? Are they gonna be like the Northern European countries of your Kings Versus Popes book? And the countries that have done poorly like the United States, Brazil, and other places, are we gonna be the ones that are like Southern Europe, that have lower standard of living? Tell us about what do you see in the future.

Bruce: This is a great question and I’m going to try to answer it in the context of the selectorate view of the world.

Jed: Okay.

Bruce: So in the selectorate understanding of organizations, leaders, politicians, always want to shrink their coalition, if they can. And there are some circumstances that compel them to live with a bigger coalition than they would like. Sometimes those were inherited, and sometimes those are the consequence of of external exogenous random shocks.

So for example, major earthquakes or tsunamis, major natural disasters, often drive large numbers of people into refugee camps where essentially they are given unwittingly freedom of assembly. And so they can coordinate and rise up against the government and compel it to adopt a larger coalition, produce better public policy.

The COVID pandemic has driven people disproportionately indoors, the opposite, away from artificial freedom of assembly, away from even genuine freedom of assembly to isolation. And that makes for an environment I believe that will make it easier for politicians to shrink their coalition and pull back on the ability of people who are unhappy with governance to coordinate with each other.

Jed: Okay.

Bruce: Fortunately we have technologies such as we’re using so that we can coordinate remotely. But for many, especially poor people, these technologies are not readily available. And so they are not able to promote and defend their interests as well as they otherwise could.

So I anticipate that one of the consequences of the pandemic is somewhat more authoritarian government, somewhat less accountable government, and a somewhat diminished world politically as a result, which will in turn create a new tension, because once we have control over the pandemic, then we can expect that more people will want to organize to change the pullback that has occurred. And in some places, they will succeed, and in others they won’t. The theory predicts where they will succeed and where they won’t. That’s a pretty technical conversation. Yeah.

Jed: But there’s really then no difference between the countries that have handled the pandemic well and have low death rates compared to other countries, or?

Bruce: Yeah. So those who handled it well, or will have had a much shorter period of lockdown, and therefore will not have the reduction in their winning coalition, that those like the United States that have handled poorly will have.

Jed: Okay. So we can...

Bruce: So those who handled it well will come out looking better, politically.

Jed: Yeah. Well, that’s for sure. That’s what I was thinking you would say something about how the people in those countries would trust their government more and that might have an effect on public policies or whatnot, or maybe it would give the government a bigger upper hand and less public policies in countries where they did better because people are now trusting them more. I just didn’t know which way it would go but now I can see what you’re saying.

Bruce: And notice the perverse incentives that this creates. So if you can survive politically while handling the pandemic poorly you improve your future political prospects more than if you handle it well. It is the nature of large coalition, more democratic, loosely-speaking, governments is two things better public policy, and higher turnover in leaders.

Jed: Yeah.

Bruce: Autocrats have lousy public policy because they just have to satisfy their inner circle, and they typically if they make it past the first couple of years, once they know where the money is, they typically die in their sleep or get overthrown when they’re believed to have a terminal illness like when they’re very old.

32:33Sign off

Jed: Interesting. Fascinating stuff today. This has been so helpful, Professor Bueno de Mesquita. I can’t thank you enough for enlightening me and hopefully everybody who watches this video about what’s going on in your particular area of political science. Thank you for spending the time with us today.

Bruce: Thank you.

Jed: We really, really appreciate it.