Game Changers series: all videos

Henning Stein:
So our guest today, Professor Jason Brennan, is arguably the world’s leading proponent of an idea known as epistocracy, rule by the knowledgeable. And in his book Against Democracy, Professor Brennan argues that the political system as we know it today, really incentivizes voters to be ignorant, to be irrational, tribal in a sense. And we are here to discuss what an alternative system might look like, and whether recent events have further underlined why change is actually needed. So welcome to Invesco Game Changers.

Jason, many thanks for joining us. Maybe before we start, I have to tell you that Against Democracy, it’s really one of my favorite books. And it captures perfectly my own feelings about everything that’s wrong with politics in the present day, right? And they say, you should never meet your heroes, but this is a real honor for me. So let’s deal with the fundamentals first. What do we mean by epistocracy?

Jason Brennan:
Yeah. Well, thank you, by the way, for having me. And that’s a very flattering thing for you to say, I really appreciate it. So I’m also honored to be here. The word epistocracy was coined by the philosopher David Estlund back in about 1997. And it’s supposed to mean a political system more or less like the ones we see in OECD countries, a representative political system. But the difference is that in one way or another, by law political power is apportioned according to demonstrable, measurable competence, or knowledge, or goodwill, or good faith.

So as opposed to in a democracy, supposedly what makes something a democracy is that every person has fundamentally the same amount of basic fundamental political power. Now, no actual democracy is truly equal in this way, all of them exclude children, all of them have issues where richer people, better looking people, people with the right name, people with the right race are more likely to win power than others, so no democracy is actually truly equal. And in fact, in most democracies, the leaders know more than the average person. The prime minister knows more than the person at the pub.

But in an epistocracy, somehow by law, you shape the vote in order to enhance the power of knowledge. And there’s different forms of this, there’s many possible variations, some of which are more objectionable than others, some of which are more defensible than others. So the most basic form, which is not the form I actually advocate, would be you only get to vote if you could say pass a test, like a kind of voter knowledge competence exam, the way you might have to get a driver’s license. So by default, no one gets to vote. But you get to vote if you can pass or demonstrate competence. That’s probably not a defensible version.

The famous philosopher and economist, John Stuart Mill, and even the philosopher, John Rawls at times, advocated plural voting, where everyone gets one vote but some people get more. In Ireland you sort of have that, because members of certain colleges get extra votes with regard to placing seats in parliament in their upper house.

Henning Stein:
I didn’t know that. Oh, interesting.

Jason Brennan:
Yeah. There’s about six seats in the upper house, which are decided by, I think is it University College Dublin, or one of the universities there. But the system I think is probably the best, which we’ll probably talk about it more at length later, is what’s called enlightened preference voting, which is a system in which everyone is allowed to vote. Everyone really votes in a sense as an equal, but when you vote, you do three things. You tell us what it is you want. Whatever the thing we’re voting on, you tell us your position on that. You tell us who you are, we collect your demographic information, your race, your income, your ethnicity, your religious affiliation, where you live, your employment status, your educational status. And finally, we give you a battery of say 40 basic questions about basic political knowledge.

When you get those three sets of data, what you’re able to do, and any statistician can do this, the government can do this, the data can be anonymized, made public, is estimate what the public would have wanted if it were demographically identical but fully informed. So I call this enlightened preference voting. It’s a way of estimating what the public would want if it were knowledgeable, even though it isn’t. And the nice thing about that is that in that way, we’re all still sort of equal. It’s not really that I get extra votes and you get fewer votes, but rather, we are extracting knowledge from the populous in a better way. So these are some of the possible variations. But the idea of epistocracy is in some way weight votes according to knowledge, and some versions are maybe more defensible than others.

Henning Stein:
Right. So when I read Against Democracy, that was after the 2016 presidential election. And for me, it captured really powerfully the problem at the heart of the political system. And I find it really amazing that more commentators haven’t taken this view. And how can people ignore that the political system is, I would say, completely broken in some cases, in some countries? Simple fact that a non-politician in America became president should have been a wake up call, positive or negative. And has this picture changed at all in four years’ time, do you think?

Jason Brennan:
Well, one thing I can do is check and track the responses to the book. And granted, in democratic theory, which is a subfield of philosophy that studies democracy, or subfield of political science that studies democracy, there is a selection effect where the people who choose to specialize in that are also the people who love democracy the most. In the same way that for whatever reason, philosophy of religion is dominated by evangelical Christians. It doesn’t have to be, but it is.

Nevertheless, the book had generally fairly positive reviews, positive responses. Even when people didn’t necessarily like my solution, they did like the critique. It was translated into 10 languages. I just got the Ukrainian version yesterday. It was a best seller in Germany, the German version was. So I think people are open to it. There’s lots of media mentions of the idea. Every day there’s media mentions somewhere in the world, in some newspaper, of the idea.

I think people are open to it because of the problems you’re mentioning. I mean, Erdoğan in Turkey basically said to the Turkish people, “Should we remove some of the constitutional constraints on my power?” And they said, “Yes.” Brexit, it sounds mean to say this, but the truth is if Brexit voters had been better informed, they would not have voted for Brexit. It’s not just speculation. We have actually very good evidence of that because of polling data, about what leave voters believed, what remain voters believed, and what the truth was. And we know that leave voters were in general and on average, very far off from the truth on a lot of relevant political information, such as where investment is going, what sort of money is coming in and going out, what percent of the UK is made up of EU immigrants.

And we know… I’ve done the regression. You know that if you could just have waved a magic wand and make everyone fully informed, they would have vote to remain rather than to leave. The Five Star Movement in Turkey, all sorts of these issues. What’s happening right now in the US, where… I wrote a post today on my blog talking about this point, because back in 2016 and 2017, when I first started arguing this, I used Trump as an example. And I often vote Democrat, but I’m not a committed Democrat. I really don’t like either party. But I didn’t like Trump, and I knew there was something unusually weird about him. And I kind of pointed to him, and I had a lot of democratic theorists respond to me by saying, “No, you don’t understand. The reason people voted for Trump is because they were hurt by globalization, and they’re rationally trying to find someone who will protect them from the damage done by globalization.”

And back at the time, I said, “It’s not clear that’s true. We have quite a few economics studies saying that’s not the case. In fact, the average Trump voter’s fairly rich. The average Trump voter actually receives a higher subsidy, not a loss, but a subsidy to their income from globalization than I do. In fact, the people who are voting for Trump are usually not harmed by globalization, but benefited from it, so that doesn’t seem to be right.” But now, that kind of argument is even less tenable, because you have a massive percentage of Republicans who believe in the QAnon conspiracy theory, even the ones who don’t explicitly affirm it, because they don’t know what it’s called, believe many of the beliefs in that theory. A lot of them believe that the election was rigged or stolen, or there’s a massive belief in election fraud and a bunch of other utterly bizarre beliefs that really can’t be defended.

And so I think the democratic theorists who back in the day were inclined to defend the voters who voted for Trump and say, “Well, they had their reasons,” I think are now more likely to come over to my side and go, “Yeah, there’s something deeply bizarre about their behavior. It’s not that they’re rationally trying to promote their interest. It doesn’t really appear that Trump has been good for them, and it doesn’t appear that they care about that.”

Henning Stein:
And do you think your theory also applies to countries like Germany or Switzerland, where I’m based, where we have a lot of consensus building, all the parties are represented in the cabinet and it’s less of a partisan atmosphere, I would say, compared to the US, right? So does it come closer to your idea of epistocracy?

Jason Brennan:
I think some countries do democracy better than others. Switzerland, in a way, does democracy very well. Though, in a way, they don’t do that much democracy. I mean, Switzerland’s secret is so much of politics is devolved to the local level where people have a stronger… Their votes matter more at the local level, they have a stronger incentive to be informed, it’s easier to be informed. If we’re talking about fixing Main Street, I can just see that Main Street is broken.

And then at the national level, they use a lot of referenda, where they have low participation rates in the referenda. And in fact, the people who participate tend to be the best educated and most informed. And also Switzerland, it’s an unusual country. Comparing Switzerland to the US is a mistake. It’s more like comparing Switzerland to say Fairfax County, Virginia, a highly educated upper class area. In Germany, I do think that empirically speaking, having more political parties is better than having fewer. And so this is a reason to maybe use proportional voting systems rather than first past the post, like they have in the UK or the United States. First past the post voting systems empirically predict that you’ll have two major parties, it’s unstable to have more than that for a variety of mathematical reasons we don’t need to get into. And that’s what you tend to find. Even when you have more than two parties in a first past the post system, usually one is just in the process of supplanting the other and it’s disappearing.

And Voltaire had a good analysis of what’s going on here when it came to religion, and it applies to parties as well. Voltaire says back in his letters on England, I think that was the document where he wrote this. He’s like, “If there’s only one religion in an entire geographic area, the people in charge of that religion want to dominate and crush everyone to prevent any opposition. When you have two religions, Catholicism versus Lutherism, they fight it out hardcore because you’re almost on the verge of winning, and you really demonize and hate the other side. And you have so many people on your side, you can afford to sort of ignore and demonize these people.”

But Voltaire points out when you have 30 religions, 30 versions of Christianity, you can’t form your own enclave. You just can’t escape it. You have to work with other people, you have to tolerate them, you have to understand them, you can’t see them as enemies. And so when you have 30 sects of Christianity, everyone kind of accepts each other and goes about their business, and it’s fine. There’s something like that when it comes to politics. When you have lots of very small parties, it’s required that you basically don’t demonize others and work with them. When you only have two parties, it’s really easy to think our side is wonderful and we care about justice, the other side is pure evil, they’re stupid and evil. We’re good and smart. They’re stupid and evil. You can’t pull that off in a proportional voting system when there’s so many parties,

Henning Stein:
Right. And Winston Churchill, I think once said that, “The best argument against democracy is a five minute conversation with the average voter.” So without insulting any voters, of course, you basically paint a picture of hooligans, hobbits, and vulcans in your book, right? That’s how you describe voters. And I’m particularly interested in the idea that hooligans, somehow 5% I think you say, manage to drive a much of our system, despite being relatively uninformed. And meanwhile, the smart, super hyper rational vulcans, academics, probably like you, they effectively drown out. So do you think most voters today would fail Churchill’s test?

Jason Brennan:
I think they would, and especially if you do the five minute conversations broken up into one minute at a time over multiple years, then you’ll really see how bad it is. So to sort of summarize that for people who haven’t read the book, a hobbit is… If you’ve seen the Lord of the Rings films or read the novels, hobbits don’t really care about the outside world, even though we see some having an adventure to save the world, they’re supposed to be unusual. They just care about living their genteel British countryside lifestyle. The analogy to that is the typical non-voter in a Western democracy doesn’t really care that much about politics. The main reason they don’t vote is because they’re not interested in it. Because they’re not interested in it, they are very few political opinions, if any political opinions at all. They’re not very ideological. They barely ever participate, and they just sort of want to go about their day-to-day business. They’re also radically uninformed. And that’s roughly about 50% of Americans. in Switzerland, it might be maybe 35%, but it’s roughly about 50% in most countries.

Hooligans, if you’ve ever been to a football game in countries where people really care about football, like for example, I went to a football game in Brazil that exemplified this, they’re very much into their team. They love their team, they can remember lots of facts about their team, but they’re incredibly biased. So an example would be, in the United States, Tom Brady of the New England Patriots was accused of deflating footballs. Everyone from New England, including me, says, “No, of course he’s innocent. He’s just the greatest football player of all time.” And everyone else, who’s of course, jealous of his greatness, was like, “Of course he did it.” We’re getting the same information, but we’re processing that information in a way that flatters what we want to believe. And really, in the best evidence is that’s roughly the other 50% of voters in modern democracies.

Vulcans, these hyper rational, non-ideological, people who just believe things on the basis of the evidence, and they change their mind as soon as they get better evidence, and they believe things only as strongly as the evidence allows, they’re kind of an ideal type. I don’t think I’m one. I’m not saying that vulcans should rule, I don’t think there really are any. The point of them was to say many democratic theories or arguments for democracy take the form of people are hobbits, but if we get to participate, they’ll turn into vulcans and democracy will work. In fact, democracy, and every other system, is the rule of hobbits and hooligans. What do we do about their hooliganism? And what do we do to diminish that?

When you talk to the average voter or the typical voter in the United States, this is what you will find. They are radically uninformed. They know who the president is, they generally maybe know who the vice-president is, they don’t know which party controls Congress, either house, they don’t know who their representative is in either house of Congress, or they don’t know who their two senators are, they don’t know the name of their governor, they don’t know which party controls the state, they can’t name any social statistics within say five percentage points, they don’t know the unemployment rate, they don’t know income things, they don’t know about recent changes. They maybe remember about six months of economic performance, which if they do know who the incumbent is, they’ll blame it on that person, regardless of whether they’re responsible, which is sort of like saying, “I had a bad day at work, so I go home and kick the dog.” It only makes sense to blame them if they’re actually responsible for it. They don’t know much about other countries. They can’t find…

If we’re at war with a country, the majority of Americans can’t find that country on a map. And by the way, usually the Americans who can are against that war, et cetera, et cetera. Similar statistics apply to other countries in which we have statistics. We have better statistics about the US than others, but we generally find this is the same. But it’s even worse than that. It’s not just they’re radically uninformed, most voters don’t really have much in the way of political opinions. There’s a really good book on this that summarizes all the research over the past seven years called Neither Liberal Nor Conservative by Kinder and Kalmoe. And one of the things that they look at is surveys that were started by a guy named Converse, where you take a bunch of voters and you ask them their opinion on a particular topic. And then you ask them their same opinion on about that same topic a year later, and a year later, and a year later.

And he would calculate what is the correlation between these beliefs over time. And in Converse’s original study he found for most Americans it was 0.00. When you ask someone their opinion on the spot, they don’t like to say they don’t know, so they’ll make something up, but they really don’t have opinions, they’re just making something up. And even among the people who do appear to have opinions, it’s much more common for someone to do this. Like, “I am committed to being a Democrat because I derive social benefits from being seen as a Democrat. Other college professors like Democrats and they hate Republicans, so I better be a Democrat so they like me,” and that’s kind of what we’re doing. And then I find out what the Democrats believe, and I say I believe the same thing. I might even convince myself to do it.

As soon as the Democrats change, I will change too. And if you ask people, these are real studies, if you ask people, “Why did you change your mind?” They say, “I’ve always thought that.” For example, with Trump, when he became the presumptive nominee in 2016, before he was the presumptive nominee, the overwhelming majority of Republicans said they were in favor of free trade. Within about a month of him being the presumptive nominee, the overwhelming majority of them said they were protectionists. When people asked these Republicans in polls and surveys and scientific studies, “Why did you change your mind?” They said, “We didn’t. I’ve always thought that.”

If that seems weird, think about sports again. Tom Brady, when he is the quarterback of the New England Patriots, Paddy down at The Lansdowne Pub in Boston, says, “Tom is the greatest quarterback of all time. He’s the greatest football player of all time.” But Paddy, you know about him, that had Tom just decided to leave for money and go play for the Jets, Patty would have said, “I always thought he was always overrated. I’ve been saying that for years.” Paddy’s sincere, but he’s wrong about himself. And that’s kind of how we think about politics, and that’s how democracy works.

Henning Stein:
That’s interesting. You stress in your book, Against Democracy, that actually it’s more about enhancing democracy then dismantling it. So with that in mind, how might epistocracy work in practice?

Jason Brennan:
Right. There’s a couple of forms that are more defensible than others, and let’s talk about one of them, the idea of enlightened preference voting. So the enlightened preference method that I have in mind, it’s not completely original, in the sense that it’s a research method that political scientists and economists use right now. So here’s a question, I’m of Irish descent and I’m high-income, and I’m highly educated. How did those factors affect my voting behavior? Well, if you wanted to know that, you couldn’t just look at things like how do say Irish people vote, because that’s one of many, many factors that influence my voting behavior. What you need to do is get lots of information about people’s demographics, among lots and lots of different people, plus you need to know what they know as well. What sort of information do they have?

Once you have all this data about lots of people you can then using basic statistics, isolate variables. Here’s how Irish identity affects voting behavior while isolating everything else, income. Eliminating the effect of everything else, income, employment status, education, knowledge, which are different, knowledge and education, not the same thing, et cetera. And so this is how we study this. And because of this, there’ve been a lot of studies on this by people like Bryan Caplan, and Larry Bartells, Martin Gilens, let’s see, Scott Althaus and many others. We can say things like it turns out that high-information voters tend to have systematically different policy preferences than low-information voters. This is not explained by income, it’s not even explained by education, it’s not explained by race, or ethnicity, or religion, it’s just an information effect.

So this is a research method that political scientists and economists have used for a long time. And using this, you can do things like calculate well, what would the public have wanted if it were fully informed? We give a public a poll, we know how informed they are, what would they have said on this poll if they were fully informed? What would they have said if they were completely uninformed? How does that match up to what they actually want? So what if we took this research method and used it to actually create law?

So again, the idea is on election day, you come in, everyone gets to vote. Your cat can vote. Your kids can vote, there’s no reason not to let them. Let everybody vote. When you vote, though, you do three things. You tell us who you are, we get all of your demographic information. And there’s important interesting questions about how you collect this. Does it get assigned by the government? We want to let people just say at the point of voting, because they might strategically lie? That’s a complication that needs to be figured out, but that’s part of it. So we get your demographic information. We give you a quiz of say 40 questions about basic political information. Who’s your Senator? Who’s your representative? Who’s your member of parliament? What party is your member of parliament part of? What’s the unemployment rate? What’s the price of a bus ticket across town? It could be all sorts of things.

Then we find out what it is you want. How did you vote? And then this information is anonymized, it’s made public. And you calculate, what would the public have supported if it had gotten a perfect score on that quiz, a demographically identical public had have gotten a perfect score on this quiz? And you do that instead.

Now, I think the biggest complication here, which people are right to pick at, is well, who decides what goes on the quiz? It’s one thing when you have economists and political scientists and others who are just trying to do research, and even then there’s a question of what they put in. And we often are using things like the American National Election Studies and that sort of stuff. But they’re doing it to try to figure out the truth, they’re not trying to influence power. Well, once you have a quiz, then people have a incentive to game that quiz, because you might want to change the questions in a way that benefits your party at the expense of others. So my suggestion, and this is all very provisional, I really favor experimenting with this on a small scale, studying it more. Then if it works, trying it. I don’t favor let’s do this right now. We don’t know enough to do it right now.

Henning Stein:
Your proposal would be to inform the public about a potential outcome, or would it be to override the results?

I mean, at the very least, I think we should use it for information purposes, because if we say to someone, “Look, we gave you a quiz. We gave the public a quiz on basic political knowledge. The average score in that quiz was equivalent to chance. It was a multiple choice question. We gave you four answers, you got a 25% on this quiz.” That’s a pretty good estimate about how the public tends to do, the same as chance. We don’t know if they’re ignorant or if they’re guessing. So you got 25% of them right, but had you got 100% of them right, you would have voted for that person rather than that person. I think that would be good. But I also think it is worth trying actually just using this as the way we decide policy, that would be even better, I think.

In part, because if this were the way we were deciding policy, it would change the behavior of parties and candidates. When candidates are running for office, when parties are putting forward candidates or putting forward platforms, they are responding to the voting environment. This is one reason why Republicans don’t campaign in New York and California for the presidency. Because even though there are more Trump voters in California than there are in Texas, but because California is so big and has so many Democratic voters, it will always go to the Democrat, except I guess for Reagan in ’84, so it doesn’t make sense to even talk to those people. You don’t even try to appeal to them. You appeal to people in swing states.

And granted, in other countries with different systems, you’re going to get different but similar kinds of effects. The way you campaign, the platforms you push are based upon the voting system that you have. If you had an enlightened preference voting system, then what would happen is you’d have better campaigning, better quality candidates, better quality platforms, because that’s what it would take to win in that system. So they would purposely strategically modify their behavior to be higher quality in order to be selected by it.

Again, I mean the main puzzle is who chooses what goes on the test. And this may sound puzzling, but what I think we should try is actually having democracy choose to write the test. You randomly select 500 people, 500 citizens at random, almost the way you do jury duty in the US, and I think in the UK as well. I think you have randomly selected juries when you have a jury, is that right? Okay. So you randomly select people for the system. You pay them. They come in, they spend a weekend or a week together. And you tell them, “You are tasked with identifying an operational definition of an informed citizen. You give us 40 questions that you think an informed person ought to know. They can be any kind of question you want, you pick it. You’re going to deliberate together and decide that, write down those questions. And this is what we’re going to use to judge what counts as an informed vote. This is the one thing we’re going to use to weigh votes in order to generate the outcome.”

And this may sound perplexing because I’ve just argued at great length that citizens are completely uninformed, but it’s actually not that puzzling when you look at it more. In fact, when you go and ask voters, “What do you think you ought to know to be informed?” They say more or less the same thing I say. They say, “You should know the unemployment rate. You should maybe know the price of some commodities. You should know about recent changes in the economy. You should know the crime rate. You should know some facts about the social sciences. You should know who your representatives are. You should know the constitution of your country and some facts about the way the government operates and who has what power.”

Voters in a sense, they know what they ought to know in the abstract, they just don’t know it. So the typical voter’s like, “Of course you should know what’s in the first amendment.” “Great. What’s in the first amendment?” Well, they don’t know. “Of course you should know who your congressperson is.” “Cool. Who’s your congressperson?” “I don’t know.” “What party are they a part of?” “I’m not sure.” So it’s kind of voters are almost in a position like a student who did badly in chemistry class. They’re aware that like this sort of stuff… “I’m taking an organic chemistry class, so I need to know what a ketone is and how it reacts with all aldehydes or something, but I don’t actually know what they do. I just know I’m supposed to know that.”

So I think democracies are competent to tell us what counts as information, even if they don’t have that information. And this would be a way of avoiding having political parties and others manipulate that test for their own benefit. At least preventing them from doing too much of that.

Henning Stein:
Right. So one problem, as you mentioned, the voters knowledgeable and unknowledgeable voters. I understand that. And my question, are there also other measures to sort of encourage the superior decision-making at the highest level, if you will? I mean, you mentioned the quiz as one tool, if you will, but do we have to radically rethink also the people who run for office? Do they even should have party affiliation, or should they be having a neutral technocratic sort of agenda?

Jason Brennan:
I think there’ll be a lot to be said for not having parties. It’s something that hasn’t been studied at great length, so it’s largely speculative. What would happen if there weren’t parties? I’m worried though that there’s not really a way of avoiding that. That one way or another people are going to, even if there isn’t like a formal legal system, there’s going to be something where people affiliate themselves in a group, and people are going to use that as a proxy for deciding whom to vote for and why they vote for them. And then you’re going to have things like certain groups are associated with certain types of people, and then you’ll get that kind of fandom behavior. What you tend to see in the US when you… We have a lot of data on this. Is something like Boston Irish Catholics. I’m using them because that’s my group. The Boston Irish Catholics vote Democrat, regardless of whether they agree with the Democratic platform or know what the Democrat platform is, because there’s kind of a social benefit to it, in the same way that there’s a social benefit from being a Celtics, Red Sox and Patriots, or Bruins fan.

Southern evangelicals now tend to vote Republican, they used to vote Democrat. And again, there’s a kind of social benefit. I think you’re going to see that even if there aren’t formal political parties. So nevertheless, the thing is when you change the voting system, you change the number of political parties, the strength of the parties and the composition and also what the parties do and how they campaign. So some of the problems that arise as a result of party politics are themselves endogenous to the voting system. If you fix the voting system, you’ll reduce many of those problems.

Henning Stein:
Right. And do you actually need to elect a government, or can that be appointed and then run by KPIs? Basically, let’s say low-unemployment, high-growth, a functioning health system. I mean, these are not rocket science ideas, right? They are everywhere the same. Most thinking is there a way of you elect basically the government system around the government, but the government itself is basically almost like a company.

Jason Brennan:
Yeah. I mean, in a way that’s kind of what you have in Singapore. People like to say that Singapore is not democratic. I think it’s significantly more democratic than people give it credit for. But Singapore, I’m not sure how much we can explore their model elsewhere, but they do have a system which heavily rewards technocratic expertise and actual competence. And the result is that the technocrats who’ve run the government, and I’m not saying everything they do is good, they have problems with civil liberties and so on, but they took a country that was really poor and made it into this fabulous, incredibly rich place in a very short period of time, with a wonderful standard of living.

I don’t know if you can really scale that up to a large country like the US, or the UK, or Germany. I think there might be puzzles about the optimal size of government, and the city state is a sort of unique type of government in a way, where because you have so many people congregated in a small area and they’re all living roughly the same lifestyle, it might be easier to be technocratic in administering that than others. The other puzzle that you have though, and this is why representative government is important, is bureaucracies have agendas of their own. And there’s a strong tendency for bureaucracies to go against the common good and to promote the good of the bureaucracy instead.

I mean, the economist William Niskanen, said that, “The best way to model bureaucracy is a budget maximizer.” In the same way that you model firms in a competitive market is trying to maximize their profit, you should model bureaucracies as fundamentally concerned to maximize their discretionary budget, rather than to promote the common good. And there’s a lot of evidence that that’s true, and that’s how they behave. So if you have… In a way, yeah, if you could just appoint well-motivated bureaucrats to run everything, that would be wonderful. But the problem is they’re real people. And once you create this agency, they’re going to have a strong incentive to use that agency for their own benefit.

You work in college, I work in a college, we see this in universities. Individual units within the university system largely are trying to promote their financial and their status interests at the expense of the university, and at the expense of the students. That happens in government too. So we have to have some sort of genuine oversight, even though there are certain cases where it’s important to eliminate political control. One reason why most country’s central bank is not controlled democratically is because I think the leaders know, the empirics are clear, that democratic control of central banks is disastrous. You need it to be in government.

Henning Stein:
So I agree the political system shouldn’t be seen as immune from innovation and disruption. And always surprises me, we see all the corporations always on the edge of new innovation, and rethinking their own processes, but you don’t see that so much in the political system. So epistocracy almost seems to be a huge step change, right? So how likely do you think that is coming into action, and will be introduced by someone or a party, maybe?

Jason Brennan:
Yeah. Great question. And I’m glad you brought up corporations, because there’s a dynamic that applies to corporations, which very weakly or barely applies to governments, and this is why change is so slow. Corporations, it’s understood if you’ve ever read Ronald Coase, a corporation, a firm is in a way, the internal part of a firm is an immune to the market. It’s the elimination of market mechanisms and the creation of a hierarchy. What does that mean? Because a corporation could simply, instead of having a full-time accountant, whenever someone needs some accounting service, they could just hire someone in the market, but they don’t. They have internal employees, where there’s a command structure.

So you get all these bureaucratic problems, but nevertheless, corporations are still influenced by a market. They’re competing with others. They’re forced to be profitable, to avoid externalities, do these other sorts of things, or they will lose their business. In a contrast, governments are monopolies. the US government has a monopoly of control rights in the territory that we call the United States. The UK government has a monopoly of control rights in the territory that we call the UK. They don’t really compete very much. They do some competition, because there’s some ability for people to move between places, but usually it’s only very privileged people. There’s more ability for companies and financial capital to move. So there is some competition in the terms of foot voting, to try to make your place attractive for investment and for immigration and so on, but it’s not that much, they’re monopolies. They have all the problems of monopolies. And thus, they’re slow to change. They face much less competitive pressure. And so there’s much more internal rent seeking and capture.

What it really takes, I think, for massive institutional change, historically speaking, is crisis. You have a massive crisis. And when there is a crisis, people look around and say, “We’re having a problem. What do we do? This isn’t working. What do we pick instead?” And then they look for something that’s in the air, that seems better. So in a way, epistocracy, might we see it? I think I’m doing, by writing this book, a lot of what people were doing in the 1700s in Europe. You have philosophers and the equivalent to political scientists, the equivalent of economists, we don’t really have those terms back then, writing about why more or less democratic representative government, I mean, they just called it republicanism, would be superior to genuine absolute monarchies and dictatorial monarchies. Constitutional monarchy is still really kind of… it’s really a democracy, right?

So they’re defending that before you see it in place. And then as monarchies have these series of crises in the early 1800s, you see a tendency and then later, even to the 1900s, you see a tendency for them to look around, and these ideas, from economists and philosophers and others, have seeped into the public, have been spread to the public, who endorses them even though they don’t know the source of these ideas, they’ve just kind of inherited them. And then when these crises occur, they look to liberal constitutional republics, and liberal constitutional democracies as a solution to the problems of monarchy.

In a way, I’m trying to make an investment in the future, where if democracies have these sorts of crises and people are looking for an alternative, maybe 30-50 years from now, this will be something that’s seeped into the consciousness and people will be open to it. Is it working? I don’t know. The word epistocracy gets a lot of mentions now compared to in the past. And I do notice that politicians in the US are throwing the word hooligan around the way that I use it. So even in Congress the other day, one of the senators used hooligan. I’m like, “He said it. There you go. He doesn’t know who I am, but somehow he heard about that.” So maybe that’s what I’m up to, but I don’t think we’re going to see it in the next three years.

Henning Stein:
Right. Well, Jason, many thanks again. This is a conversation I wanted to have for a long time, ever since I read your book, so thank you for making this happen. And I hope our viewers have found it as fascinating as I have. And for anyone who hasn’t read it, I strongly recommend Against Democracy, even if you don’t agree with the message, I think it will make you think, and isn’t that a vital element of what democracy is ultimately about? So see you again soon.

Jason Brennan:
Yes. Thanks so much.