How to Predict the Presidency

In our previous episode, we had a wide-ranging conversation about presidential history, and presidential power, with the University of Chicago legal scholar Eric Posner.

Eric POSNER: The Founders could not possibly have imagined that the president would become as powerful as he has.  

And we wondered: is the U.S. presidency turning into something like a dictatorship?

POSNER: Yes, I think that is happening. Although, you know, “dictatorship” is such a freighted term.

But that conversation with Posner was recorded in 2016, a couple months before Donald Trump was elected president. Trump is of course now running again, against Vice President Kamala Harris. So, today on Freakonomics Radio, we go back to Eric Posner to talk about what’s happened over the past eight years, and what the future may bring:

POSNER: I think it’s actually pretty hard to be a dictator. You have to be kind of smart. You have to be tough. You have to be brave. 

Also: have you lost faith in election polls? If so, how would you feel about an election betting market:

Koleman STRUMPF: A market doesn’t delay information. A market doesn’t spin numbers. A market just gives you numbers. 

This is our election special — not what you’re likely to hear elsewhere — and it starts now.

*      *      *

The University of Chicago law professor Eric Posner has written more than a dozen books, on topics ranging from antitrust regulation to human rights to the U.S. Constitution. As I mentioned earlier, we interviewed him eight years ago — eight long years ago — about the evolution of presidential power.

POSNER: I listened to the interview this morning, and I will say that I have changed my views about the presidency to some extent. Because I do think Trump was a shock, and not the sort of person I expected to become a president at the time that I did a lot of my earlier writing on presidential powers, which tended to be more optimistic about the powerful presidency. 

Stephen DUBNER: What surprised you about your assessment of Trump and the presidency back then? 

POSNER: I think I was a little glib about the rise of presidential power. You could think that the risk of a too-strong president is dictatorship. But another risk of a too-strong president is just really bad governance and chaos. And I don’t think I was expecting that. 

DUBNER: So if I didn’t know anything about you, and I heard your quick musings here on presidential power, I might think, “Oh well, Posner sounds like he is aligned with the Democrats pretty strongly.” Persuade me that that’s not the case. Or if it is the case, persuade me that your research and writing is as objective as it can be for a legal scholar. 

POSNER: Trump is really sui generis. When we talked about him eight years ago, it was during the campaign, and I think — 

DUBNER: And we should remind people that it was during a campaign in which he was not expected to win by the vast majority of people. 

POSNER: Right, so we talked about him in a kind of casual way. And I didn’t expect him to win the presidency. I do think he was a bad president. Trump had certain goals, most of which he didn’t achieve. And then Biden has been a very different president. He had a number of goals which he achieved by using the instruments of power at his disposal. So I think in that sense, he’s been very competent. Biden’s particular policy choices, I don’t have as strong views about whether the Inflation Reduction Act was a good idea or the stimulus bill was. As I said to you eight years ago when you asked me about Obamacare, these are complex areas, and I hesitate to come down hard in favor of one perspective or the other. 

DUBNER: The last time we spoke, you said that the U.S. presidency had been turned into something of a dictatorship, although you didn’t like that word. You said “presidential primacy” was more to your liking. So I’d like to hear your current views. 

POSNER: There are two ways to look at this. One, just in terms of the empirical reality of the president’s role in our political system, and the other is whether it’s good or bad. Let me start with the first. I’m going to avoid the term “dictator.” You know, people think of Hitler. I don’t think that’s a helpful lens to look at Donald Trump, or any American president. I hardly need to say that Trump tested the limits of presidential power. What’s ironic is that Trump, for all of his bluster, he wasn’t able to use these tools very well and accomplished very little. A lot of it was just his temperament. If you read memoirs written by his former aides, this is a guy who couldn’t focus, was easily flattered and swayed, couldn’t control his subordinates. So it’s not surprising that he wasn’t able to maintain a strong and consistent vision. When he had a Republican House and Senate, he was able to obtain a tax cut. That seems to have been about the extent of his accomplishments. During the pandemic, he didn’t really do a whole lot. But I think it’s fair to say that by authorizing Operation Warp Speed, which resulted in the vaccine, he accomplished something of value. But if you take something else, you know — one of his policies was to reduce illegal immigration into the United States, and his main instrument for doing that was to build a wall along the border. I think a more competent president could have taken stronger actions. Probably —

DUBNER: Something perhaps less concrete? Something — 

POSNER: So to speak. 

DUBNER: Something that a construction executive might not have come up with, in other words. A policy, in fact. 

POSNER: As I understand it, the reason why Trump hit upon the wall as his major goal was that it just came to mind, and it worked well with the crowds, and so he repeated it, and eventually he probably felt he had to do it, whether it made sense or not. But nobody ever thought it was a good way to stop illegal immigration. Illegal immigration is one of these incredibly complicated problems that one can address in a variety of ways. And I don’t think he had the patience and discipline to figure that stuff out and put it into effect. I think he had power that he could have used to accomplish more that he wanted to achieve, but simply lacked — not just the temperament but experience. He was new to the government. He didn’t know how it works. And certainly he faced a lot of resistance in the bureaucracy. Biden has been a more successful president in terms of achieving his policy goals, but has been less reliant on the inherent powers of the presidency to accomplish them. 

DUBNER: And is that because Biden had more of a connection and facility with the existing infrastructure in D.C.? 

POSNER: I think that’s part of the answer. A lot of this has to do with Trump’s character and temperament, and —

DUBNER: Meaning Biden is a little bit more, let’s say, respectful of norms than Trump? 

POSNER: Well, Biden, even in his advanced age, has a longer attention span, more self-discipline, shows greater loyalty to his subordinates, appoints higher-quality people. He’s been a much more careful, and, I suppose, more conventional type of president. And that has helped him. 

DUBNER: Toward the end of Donald Trump’s first term — maybe only term, but for now, first term — you published a book called The Demagogue’s Playbook: The Battle for American Democracy From the Founders to Trump. So let’s talk about that a bit. Can we start by defining some terms? How do you define a demagogue, first of all? 

POSNER: A demagogue is a politician who tries to obtain and keep power by dividing the population, usually by choosing an enemy. The enemy could be foreigners, it could be a minority group in the country, or it could be — in the case of a left-wing demagogue — the rich, or the people with property. 

DUBNER: When you look around the world today, who are some of your favorite practicing demagogues? 

POSNER: Like everything else, it’s a spectrum. Trump was a true demagogue. Bolsonaro was a demagogue.

DUBNER: Orban?

POSNER: Orban is a demagogue.

DUBNER: Putin?

POSNER: I think applying the term to an autocrat is a little odd. I tend to think of demagoguery as a problem for democracies. 

DUBNER: And what are the circumstances within a democracy that tend to produce an appetite for, and/or an opportunity for, a demagogue?

POSNER: It’s a Greek word, and it was a term used by the ancient Greeks, typically by people who were suspicious of democracy. A lot of these Greek city-states had democracies or basic political systems that resembled democracies. It was very common for a person, usually a person from the upper class — and this person would like power but he’s not really part of the elites who are in power, and he doesn’t have any means for getting the elites on his side. So what the demagogue does is he appeals to the people. And he appeals to the people by basically making up stuff, often lying, propounding conspiracy theories just like today, riling people up, and trying to persuade them that the people in power are conspiring to harm the people. Now, when does this arise? I think it can always arise, but it’s more likely to arise when people are miserable. So it could be during an economic downturn, or it could be during a war, or some kind of natural disaster. But once it gets started, it can just occur in cycles. And this is really what the Greeks were worried about, was that you might have a demagogue who successfully comes to power, becomes an autocrat, and controls the people — and then somebody else tries to get power back by becoming a demagogue, and getting the people on his side. And this leads to civil war. Civil war is a terrible thing. And this is why a lot of Greek thinkers were quite skeptical of democracy. They thought it tended to collapse in this way.

And the Founders were very concerned about demagoguery. They knew all about ancient Greece and ancient Rome. And so the Constitution was designed in large part to limit the power of the people so that demagogues wouldn’t be able to achieve power. An early demagogue was Andrew Jackson, who actually introduced the party system. But the party system, after Jackson, became somewhat hierarchical, in the sense that each party was controlled by professional politicians. What Jackson really did was transfer power from the national elites to local elites. But the local elites were hierarchical superiors to ordinary people. And so they made sure that the type of people who were candidates for elections were professional politicians, or at least a respectable former general or something like that. In the 20th century, though, this way that the parties were organized increasingly was in tension with people’s sense of what a democracy requires. I think this was happening in part because in the national government, the bureaucracy was growing, and you had this distant-seeming administrative state that was controlled by these parties that ordinary people couldn’t really influence except maybe once every few years or so. One of the solutions to this problem was to introduce primaries where ordinary people would have a more direct impact on the choice of the leader of their party, who would then be pitted against the leader of the other party in the presidential election. I do think American democracy is different from what the Founders thought they were creating. The Founders, they didn’t use the term democracy. They thought of democracy as chaotic and horrible. And if they could look into the future, they might have said, “Yeah, Trump, that’s democracy, and we don’t want that to happen.”  

DUBNER: Let me run past you another characterization of Trump from your book, The Demagogue’s Playbook. “We need to see Trump,” you write, “as a political monstrosity who should be repudiated by the body politic so that politicians who eye the presidency in the future will be deterred from using Trump’s ascendance as a model.” Now, it does appear that roughly half of U.S. voters don’t see him as a political monstrosity. So what’s your explanation for that? 

POSNER: I think he’s a monstrosity in a kind of constitutional sense. The Founders are quite explicit that the checks and balances weren’t going to be sufficient unless virtuous people became office-holders. They were quite worried about this. Virtuous and competent people. And in that sense, they were real elitists. Whatever you think of the various presidents we’ve had — some of them were truly bad, but most of them were basically competent people who became president because they rose through the ranks and earned the trust of people, and so on and so forth. Trump is a monstrosity compared to that baseline. He persuaded a lot of people that the mysterious, powerful elites in government and elsewhere are arrayed against him. In some ways, he’s acted like previous demagogues all the way back to ancient Greece. But, like the old demagogues,  he wasn’t good at governance. And, it’s a shame that people support him. But this is always what happens in this setting. They support him. They say, “Okay, he lies, and he’s awful in many ways, but at least he’s not as bad as the people on the other side.”

DUBNER: Republicans do seem to dislike Kamala Harris about as much as the Democrats dislike Donald Trump. And a big part of the argument is that her ascendance to the presidential ticket was undemocratic. She didn’t win any of the primaries. So what is your view of that, in light of constitutional history especially. Does that make you uncomfortable? 

POSNER: No, I think the small-d democratic primaries have perhaps not worked out the way one hoped.

DUBNER: What do you mean by that? 

POSNER: Well, because the primary system led to people like Trump. Whereas before primaries were so heavily relied on, it was much more likely that you would get a seasoned professional politician or an accomplished general like Eisenhower. So, I think the primary system probably is to the advantage of demagogues. Look, our system is not a pure democracy. No system is. There are lots of ways in which people’s votes are constrained. We have the whole electoral college system. People are inattentive to politics most of the time, so elected officials have a lot of authority. The way that Kamala Harris obtained the Democratic candidacy is pretty small beans relative to all the rest of this stuff.  

Coming up: what would a second Trump term look like?

*      *      *

The legal scholar Eric Posner argues that Donald Trump didn’t accomplish much during his first term as president. Does he see Trump accomplishing more if he were to win a second term?

POSNER: I don’t think he’s going to productively navigate the system if he has a second term. I really don’t. I think the people around him, they’re trying as hard as they can to prepare a body of people he can appoint who will enforce his vision. So, you know about Project 2025, produced by the Heritage Foundation. There’s this 900-page manual that’s supposed to guide Trump, so that he achieves what these guys think his objectives are. 

DUBNER: Even though this project is supported by and drawn up by many former Trump allies, and colleagues, he has claimed to know nothing of it, and to not support it at all. Do you know anything further about that? 

POSNER: He said he never read it. And I really believe that. I mean, Trump is not a big reader. But really, no one could read this document. I’ve read parts of it. It’s just — it’s boring and it goes into all kinds of minutiae. It has this weird paranoid tone. It’s just an awful document — not really because it’s trying to establish a Trumpian dictatorship. It actually is in many ways conventional Republican thinking going back decades. In some ways, it reflects Trump’s particular concerns about illegal immigration and tariffs and so forth. But remember, Trump doesn’t like the idea that other people are telling him what to do. And now, I suspect he’ll feel constrained not to rely on Project 2025 if he’s elected because that’ll make him look weak and like he’s being manipulated. Anyway, it’s just dumb document. So, he’ll just start from scratch.

DUBNER: Let’s imagine for a minute that Trump is elected again. Let’s take three major areas. Let’s talk about the judiciary, federal agencies, and let’s talk about dealings with foreign counterparts. Can you walk me through your expectations in those areas? 

POSNER: Okay, with the proviso that Trump is an inherently unpredictable person. 

DUBNER: Sure.

POSNER: Let’s start with the judiciary. So you might think, one of his biggest accomplishments was appointing judges who are on the right and then the Supreme Court got rid of Roe v. Wade and so forth. But Trump doesn’t like these judges because his appointees on the Supreme Court ruled against him in his January 6th lawsuits. This seems to be how Trump thinks. So what went wrong? What went wrong was that these judges were affiliated with the Federalist Society. So maybe he shouldn’t appoint people who are affiliated with the Federalist Society. And then the question is, well, who does he appoint? Legal conservatives realized a long time ago that if they wanted to be judges, they should affiliate themselves with the Federalist Society. So that means there’s a very shallow talent pool outside of the Federalist Society. And Trump might end up appointing a bunch of incompetent judges who may or may not be able to achieve his goals. 

DUBNER: Okay, how about federal agencies? 

POSNER: This was a huge problem for Trump, and this was extremely predictable. He becomes president in 2017. There’s this huge bureaucracy. These people are supposed to do what he wants them to do. He issues orders. Nobody pays attention to them, or they slow-walk his projects. So what do you do about this? Will you make sure that the bureaucracy is somehow replaced with Trumpian loyalists? Again, there’s this problem of the shallow talent pool. There’s also the problem of getting rid of civil servants. So there is this theory that Trump’s supporters have advanced that he can reclassify a lot of these people, and fire them. He actually tried to do this at the end of his first term. It didn’t work. So I don’t think he’s going to be able to convert the bureaucracy into a usable and loyal bureaucracy in his next term. 

DUBNER: And then foreign counterparts.

POSNER: Foreign counterparts. 

DUBNER: Many of whom have expressed almost publicly — I mean, it’s leaked — that they really saw him as a joker. Some of them he forgives and forgets, some of them he doesn’t. So how do you see that playing out? 

POSNER: I think he’ll treat them the way he treated them during his first term. If you read the memoirs and the journalistic histories, it sounds like maybe he had rapport with some of them, but that he was easily manipulated. He’s a vain person who was swayed by charming foreigners. He might be a little bit more sophisticated now than he was. But he seems to be so impulsive, and so hard to control by his aides, and so unwilling to let them guide him, even though they’re the people who really know what’s going on, I just imagine it’ll be another chaotic term.

DUBNER: And how much does that matter? You and I first spoke years and years ago for an episode called something like, “How Much Does the President Actually Matter?” The underlying argument was that most Americans tend to attribute too much weight to the U.S. presidency. In the case of foreign affairs, however, there are some unilateral powers, and the figurehead status alone is pretty significant. So, if Trump were to be reelected and were to not, let’s say, put the best face forward, how much do you think it matters? 

POSNER: I think it matters a lot. And maybe to clarify a bit what’s at stake — I agree with you that a lot of ordinary people think the president has more power than he does. That if there’s inflation or an economic downturn or a war in the Middle East, it’s the president’s fault. And often there’s just nothing that presidents can do about that. But I think a lot of, for example, commentators, professors, journalists think the president has less power than he really does. He has all of these ways of influencing policy through the bureaucracy that are often invisible. Now with respect to foreign affairs, I think if there’s a real crisis, it could be a big problem. The foreign policy machine is very big and kind of hard to control. But Trump really did have this effect, in the sense that he moved the country more toward isolationism. Now, interestingly, Biden has to some extent followed Trump’s lead, which makes you think that Trump was, in fact, reflecting something fundamental about how the public was thinking.

DUBNER: With China especially, Biden following the lead, right? 

POSNER: Yes, with China especially. Maybe Trump was, you know, ahead of the foreign policy establishment, I don’t know.

It’s easy to forget that not so long ago, there was a question of whether Donald Trump would even be allowed to run for a second term. Multiple lawsuits were filed, trying to keep Trump off the ballot, arguing that his support of the January 6th riot at the Capitol constituted an illegal insurrection. The U.S. Supreme Court ultimately ruled, in a unanimous decision, that Trump could not be removed from state ballots. There have been many other lawsuits filed against Trump. Manhattan District Attorney Alvin Bragg won a conviction against Trump, on charges of falsifying business records in order to conceal hush-money payments to a porn star. There are also outstanding federal charges, related to election interference and Trump’s mishandling of classified documents. I asked Eric Posner what he thinks about these attempts to stymie Trump legally.

POSNER: There’s always been this worry, that one way that a democracy will collapse is that the people in power will use the legal system to harass their political opponents. There’s a lot of reason to think that’s a real danger. There’s a ton of historical examples of that. In the United States, there’s been relatively little of that kind of behavior. I think that’s been a good thing. It’s definitely better for the people in power to try to avoid using the legal system against their opponents, even if their opponents are maybe breaking the law a little bit. Now, if their opponents start murdering people — by all means, the police should arrest them, they should be tried. The Justice Department has always had this view that you’re not going to try to prosecute political opponents, especially at sensitive times, like during a campaign, unless you absolutely have to. So, among the various indictments of Trump, I do think the New York indictment, the Alvin Bragg indictment, was definitely the weakest. This was basically a very minor kind of fraud that Trump committed in New York, and —

DUBNER: And because it was connected to a federal election, that was the loophole, yes?

POSNER: Well, it wasn’t even clear it was connected to the federal election. They also said it was connected to violation of the tax laws. They also said it was connected to state election laws. And a lot of that stuff wasn’t really resolved because the jury doesn’t have to tell us the basis of its decision. I just think that all of these attempts to prosecute Trump have backfired terribly. There’s an interesting lesson here, which is that there’s a tendency — especially among people like me, law professors — to think that the reason why American democracy has lasted as long as it has is that every time we have a president, that president respects democratic norms. I think the causation might be a little backward. Presidents respect these norms because if they don’t, it backfires. People don’t like it when the government prosecutes its political opponents. And the government may be hurt because of public opinion. Or I think what we’ve really seen is, it’s hurt because judges are skeptical, they’re nervous, and there’s a lot more scrutiny of these trials. I think probably the lesson that’s going to be drawn from this was that it’s not smart, let alone consistent with whatever your political theory is, to go after your political opponents using your legal powers. 

DUBNER: Is it even less smart knowing that all the way at the top of the chain, you’ve got a U.S. Supreme Court whose key voters have been appointed by the person you’re prosecuting?

POSNER: I wouldn’t put it in quite that narrow a sense. I think the Supreme Court, maybe more than any other body in our political system, is concerned about this kind of political retaliation. And I think they’re concerned for self-interested reasons, but not in the way that you mean. When the judiciary is involved in these types of trials that are politically tinged, the judiciary is damaged just as much as the political actors who are responsible for the prosecution on one side or the other. The lower court judges and the appellate court judges, they’re all understandably nervous. They want to be extremely careful. And the Supreme Court above all wants to make sure that these prosecutions don’t happen unless they absolutely have to. I mean, the judges who ordered the execution of Charles I were hunted down and executed after the Restoration. 

DUBNER: So in your view, Eric, as a legal and constitutional scholar, should Trump have been forbidden by the courts from running again? 

POSNER: No. As bad as Trump has been, it would be worse if the judiciary were to intervene and remove from the ballot the leader of one of the parties based on this theory of the Constitution that it has some narrow legal attractiveness, but is just a totally unrealistic thing to expect the courts to do. I think the Supreme Court was wise in not disqualifying Trump.

In July of 2024, the Supreme Court ruled, in a six-to-three decision, that former presidents have absolute immunity for actions related to the core powers of their office, and that there is at least a presumption of immunity for official acts. Here’s what the three dissenting justices — the Democratic appointees — had to say: “The relationship between the President and the people he serves has shifted irrevocably. In every use of official power, the President is now a king above the law.” I asked Eric Posner what he thought about the immunity decision, and that dissenting view.

POSNER:  The immunity decision was not as outrageous as people say. Lots of government officials have immunity. And this idea of the president having immunity with respect to his core powers, I don’t think that comes as much of a surprise. But it may have a political effect. It might encourage Trump to do bad things. And the paradox is the judiciary, the Republican majority on the Supreme Court, is kind of making the government weaker, but the president stronger within the government.

DUBNER: In December of 2023, when Trump was set to run against Joe Biden —  this was several months before Biden ended up dropping out — you published a piece called “A Trump Dictatorship Won’t Happen.” I guess this was presupposing that Trump might win a second election. You wrote that: “Although Donald Trump was many things, most of them bad, he was not a fascist when he was president, and he would not become a dictator if elected again.” Okay, walk me through that. Why not?

POSNER: He won’t become a dictator, I think there are a number of reasons. First of all, I think it’s actually pretty hard to be a dictator. You have to be kind of smart — shrewd, at least. You have to be tough. You have to be brave.

DUBNER: You’re saying Trump is none of those — even brave, yeah? 

POSNER: From what I know about him, it’s just hard to imagine him having this ambition to be a dictator. I know he wants power, and he wants to hold on to power. But I think he does it in an ad hoc way rather than the kind of shrewd, planned way that real dictators do to obtain power. And then the other thing is just that, I do think that the other institutions — the press, Congress, the courts — they’re not going to let him be a dictator. 

DUBNER: You write further, “The power of constitutional and bureaucratic hurdles, combined with a dearth of sympathetic right-wing radicals, ensure that anarchy is more likely than tyranny.” You’re chuckling at anarchy, but I mean, anarchy is not very good either, is it? 

POSNER: No, definitely. Look, that was not a defense of Trump’s candidacy. But people have been complaining that the president is going to become a dictator, and they mean in the Hitler sense, not in the sense of having the ability to order certain types of environmental regulations. They’ve been saying that for so long, they’ve been accusing so many presidents of this, or presidential candidates. And I just think it’s become a substitute for a more careful diagnosis of the problems that we face. I think the real problem with Trump is that he makes bad decisions, and he’s corrupt, and he appoints corrupt people. And he doesn’t know how to manage people. And so, yes, anarchy might result, because the presidency is such an important role. But, you know, dictatorship is the other end of the spectrum. It’s extreme order. And we don’t recall extreme order from Trump’s term. It was the opposite. 

DUBNER: I know some people feel that we are approaching the end of democracy in the U.S. My sense is that you feel we’re perhaps approaching the end of American-style democracy. Is that accurate? 

POSNER: No, I don’t think we’re approaching an end to any kind of democracy. I think things will just sputter on as they have. I mean, what’s really unique about Trump was just how uninterested he was in understanding policy, leading the country in a competent way — or maybe he was unable — and the fact that people elected him anyway, that I just don’t know how to fit into my understanding.

DUBNER: So, if we were looking back on this period of time from 20 or 50 years hence, you think this would look more like a rocky patch than the beginning of the end? 

POSNER: If you talk about, let’s say, the last eight to 10 years, I’d say it looks like a combination of maybe the 1930s and the 1960s, and not as bad as the 1930s or as chaotic as the 1960s. 

DUBNER: So why is there the sense that it’s so much worse?

POSNER: People always think the political system is about to collapse. Look, we have to think about maybe Trump’s successors. If Vance or Hawley or Ted Cruz or Rubio or any of those guys eventually become president, they’re not going to be like Trump. Whether we like their goals or not, they’re just going to look like other presidents — Bush or Reagan or Hoover or Calvin Coolidge, whoever. I just don’t think that they look at Trump and say to themselves, “I want to be like this guy.” They say to themselves, “I want to be loved, and I want to be president, but I don’t want to be like this guy.” Trump has not provided a recipe for other people to follow. And then Biden over the last four years, in many ways, he’s a bit of a throwback. Most of what he’s done has been through legislation. He’s done a few aggressive things, like the student-debt cancellation, which was struck down by the Supreme Court. But he seems quite continuous with Obama and, for that matter, Reagan and both Bushes. I mean just a modern president who uses both legislation and executive power to accomplish his goals, and then loses power at the midterm because people get annoyed with him. Biden could very well have lost the election if he hadn’t agreed to step down. So, it seems like a return to normalcy except, of course, that Trump could be reelected. And, if he is, I think we’ll have another four years that are a lot like Trump’s previous term, but are not going to spell the end of democracy. If  all of our presidents are like him, I think eventually we’d be in big trouble. But, you know, a huge institution like the U.S. government — if you read about the history of the Ottoman Empire or the Roman Empire, for that matter, they had a lot of really terrible leaders for a long time. And, you know, they would last a few more centuries. I think it takes more than a bad president to destroy an empire. 

Coming up: will Donald Trump be elected again? We look for some answers in the betting markets.

*      *      *

So, who’s going to win the upcoming presidential election: Kamala Harris or Donald Trump? As we’ve all come to realize, political polling is an imperfect science. It often overpromises and underdelivers, which leaves a lot of people frustrated. So, is there perhaps a better tool to predict election outcomes?

STRUMPF: Whenever you can find elections, you can find betting. 

That is Koleman Strumpf.

STRUMPF: I’m an economics professor at Wake Forest University. 

Strumpf does research on topics like illegal file-sharing and tax evasion — as well as what economists call prediction markets. Which are, essentially, betting markets. Like, betting on elections.

STRUMPF: The work that I’ve done on this subject is with a tremendous economic historian named Paul Rhode. And I was very interested in the modern version of these markets. This would be about 1999. I said, “Look, they’re this new thing that came around with the internet, and it’s the wave of the future.” And Paul said, “No, I was just reading the Brooklyn Eagle in 1896. And there was a very large market of people betting on elections in New York City.” That led to a four-, five-, six-year odyssey of learning about these markets and their history.

It turns out the history goes back well before 1896.

STRUMPF:   In the 16th century, you can find people betting on who gets elected as a pope. Now, the Catholic Church was not exactly a very big fan of this — if you bet on cardinal selections of popes, you could get excommunicated. In the United States, you can find not so organized, but versions of betting going back to George Washington. These markets existed basically in any city in the United States. The biggest markets existed in New York, outside the stock exchange. Almost all the attention focused on who would get elected to president, governor, even mayor of New York. And these markets were really popular. In this period, there were no scientific polls. You did not have a New York Times-Siena poll. These things grew and grew in popularity. Then starting in the 1930s, they started to disappear. The newspapers were never comfortable talking about these markets. It was both morally and legally in a gray zone.

In 1935, the statistician George Gallup founded the American Institute of Public Opinion — which, a couple decades later, morphed into the Gallup polling organization. By using modern survey-sampling techniques, Gallup made it safe for media outlets to rely on polling data instead of gamblers. The betting markets went underground — but in the U.S., they started coming back in the late 1980s, thanks to the internet. Things were now a bit less carefree than in the old days.

STRUMPF: If you want to run one of these markets and have people invest real money in it, it’s considered a futures market. And in the United States, the Commodities Futures Trading Commission has jurisdiction over these markets.

But the C.F.T.C.’s guidance on these markets, according to Koleman Strumpf, is …

STRUMPF: … is extremely opaque and unclear. The C.F.T.C. has actually never formally recognized any political prediction market. But they’ve allowed two of these markets — the Iowa Electronic Market and a site called PredictIt, which is located in New York and Washington, D.C. — they’ve allowed them to run these markets under what’s called a no-action letter, which is essentially saying, “We’re not going to prosecute you for betting or participating on the site, but it’s not a fully recognized exchange.”

Okay, let’s begin with the Iowa Electronic Market.

STRUMPF: This was started in 1988. It’s run out of the business school at the University of Iowa. And it was always intended as an academic exercise. The people who run the site were mainly interested in testing theories about how markets worked. It was very small-scale. So there’s, I think, a limit of $500 that you can put in every market that they have. And then partly because the stakes were so small, and for reasons I can’t fully understand, people’s attention moved elsewhere. 

“Elsewhere” included the U.K., where gambling was more mainstream generally. Sports-gambling sites like Betfair carried bets for elections in the U.K. and the U.S. Then came a market called InTrade.

STRUMPF: InTrade was an Irish site. They did not really go down the route of trying to get approval from U.S. regulators, despite the fact that almost everybody on the site was American. This was the biggest show in town from around 2004 to 2012. The C.F.T.C. was very, very upset with them. The C.E.O. of InTrade basically couldn’t come to the United States for fear of getting arrested. InTrade collapsed around 2013 for old-fashioned reasons, misconduct by the people running the company. After InTrade disappeared, PredictIt came online. That started to really catch fire between 2016 and 2020. 

PredictIt came out of Victoria University of Wellington, in New Zealand. In order to operate in the U.S., it accepted limitations that would satisfy the American regulators.

STRUMPF: PredictIt, as part of the C.F.T.C. no-action letter, had to say, “We’re going to have limited stakes.” So you could invest up to $850 in one of these markets. If you’re a financial person, that’s not even pocket change.

But even that limit didn’t satisfy the C.F.T.C., which in 2022 withdrew its no-action letter and tried to ban PredictIt. That standoff has been in court ever since. But new players continue to emerge, despite — or maybe because of — the C.F.T.C.’s “opaque and uncertain” guidelines. Two of the most prominent markets right now are Kalshi and Polymarket. 

STRUMPF: Polymarket is a crypto-based site. They are set up, at least in principle, to not allow any American bettors on their site. But if you quickly Google “Polymarket,” you’ll find many people explaining that there are ways to send money to the site. So Polymarket has had its day in the sun in the last several months. They’re the biggest of any of these markets. Their biggest market, which is on who gets elected president, has taken about a billion dollars a bet.  

A billion dollars, at least, in a cryptocurrency called USD Coin, which is pegged to the U.S. dollar. And then there’s a market named Kalshi, which is an Arabic word meaning “everything.”

STRUMPF: Kalshi is a site that allows people to bet on events — weather outcomes, movies, things like that. The one thing that they really want to have markets on are elections. So they, starting about two years back, went to the C.F.T.C. and said, “Look, we are relatively experienced running these event contracts. We’re regulated by you guys. We want to run an election market.” The C.F.T.C. basically ignored their request and let it sit, and Kalshi then turned around and sued the C.F.T.C. and said, “We need a decision.” This was sitting in courts. And then a few weeks back, the judge made her initial decision, and ruled in Kalshi’s favor. This happened on a Thursday at noon. By one or two o’clock in the afternoon, Kalshi had set these markets up, and these were markets on which party would control Congress. They had a lot of money flowing into this. By the next day, the C.F.T.C. appealed, and we’re kind of in this gray zone again.

But that gray zone has shifted even since we spoke with Koleman Strumpf: a panel of three judges for the D.C. Circuit Court of Appeals voted unanimously to lift the ban while the C.F.T.C.’s appeal is ongoing, and Kalshi re-opened their election markets. So what is Kalshi’s pitch?

STRUMPF: Their biggest pitch is saying, “Look, we’re going to have very deep liquid markets,” which is just financial jargon for allowing people to put a lot of money into these markets. The reason why this could potentially be a good thing is, there are lots of people who have what we would call political risk or political uncertainty. So Imagine you’re an investor in green-energy companies. Well, based on which party wins the presidency, your financial future is going to be very different. So maybe you really wanted something that was associated with the Democrats winning the election, you would actually bet on the Republicans to win in this market. This is what a financial person would call hedging.

You can imagine that some people might not appreciate the idea of using electoral outcomes for financial hedging — or they might not appreciate the idea of betting on elections, period. One such person is Jeff Merkley, a Democratic senator from Oregon; he recently introduced the “Ban Gambling on Elections Act.” Here’s what Merkley said in a statement: “When big bets are cast on elections and dark money can smear candidates, you have the perfect combination of factors that can undermine trust in our democracy.” The C.F.T.C. has the same concern.

STRUMPF: Their concern is that this would lead to both manipulation of the markets and challenges to election integrity. In terms of manipulation of a market, what they’re concerned about is that somebody tries to move prices in the market for reasons that aren’t related to fundamental information. They’re just trying to trick a bunch of people into, for example, saying that Vice President Harris is a much stronger candidate than we think she is, and to financially profit from it. So that’s the first thing. The second thing, that is probably the more fundamental concern, is that if we have these markets, the C.F.T.C. claims, people will now have incentives to either try to change election outcomes, or, a related point is that if people watch these markets — just regular folks who aren’t even participating in these markets, and they see that there’s something going on in these markets — it might change how they vote. The head of the C.F.T.C. put all this up into a succinct phrase, says he doesn’t want to “become an election cop.”

You could, of course, argue that election polls can also influence how people vote. In any case, here’s how Strumpf sees the C.F.T.C.’s position:

STRUMPF: The C.F.T.C. does not have it in their powers to get rid of these markets. If the C.F.T.C. tomorrow said, “Look, we will not allow Kashi to have a market, we’re going to shut down PredictIt” — okay, does that mean election betting disappears tomorrow? Absolutely not. Exactly the same way that prior to 2018, if you went to almost any city in the United States and you wanted to make a sports bet, even if it wasn’t legal, you wouldn’t have too much trouble finding somebody to take your bet. The analogy that I think most people would understand would be Prohibition. Even if you totally oppose alcohol — you’re a teetotaler, you have religious reasons, whatever — having it legal and regulated as a country, we’ve decided, is a much better approach from a social perspective. A truism of economics is, when there is a tremendous demand for an activity, supply will rise to meet it. And when it is a supply that’s arising in an unregulated Wild West environment, bad things happen. 

This is one reason Strumpf thinks it might be wise to have legal, regulated election markets. There’s another reason too — even for all regular people who would never place a bet.

STRUMPF: For regular people, the amount of information you can get out of these markets is vast. It’s the best. 

Consider one research paper Strumpf wrote with the economic historian Paul Rhode. It’s called “Historical Presidential Betting Markets,” and it contains this tantalizing conclusion: “We show that the market did a remarkable job forecasting elections in an era before scientific polling. In only one case did the candidate clearly favored in the betting a month before Election Day lose, and even state-specific forecasts were quite accurate.” Strumpf also argues that betting markets can give us a better view of reality than most media coverage.

STRUMPF: I’ll give an example, which is in North Carolina, where I’m located right now, but it’s a national story. We have a very contentious governor’s race going on in the state. And in the recent period, the Republican candidate for governor, there was a big news story on CNN that said that the candidate had written some very offensive things on a not-very-nice website. If we don’t have these markets I have to rely on news media to think about, is this a big story? Even if you’re not in North Carolina, you’d say, “Wow, this is going to really hurt a Republican candidate in North Carolina, and that probably will have a spillover effect on whether Donald Trump will win North Carolina in the Electoral College. And this could change who gets elected president. So this is a gigantic story.” Okay, that’s what The New York Times says.  

Okay, and what do the betting markets say?

STRUMPF: A bunch of people who put hard money down collectively said both those two things are incorrect. First, it turned out this candidate had very little chance of getting elected governor before all this stuff came out. The second thing is, Trump’s election chances in these markets barely moved. So I could understand as a citizen what’s the effect of this big news story. I could go to a news site and have their spin on things, which sometimes is right and sometimes is wrong. Or I could go to one of these markets that cuts right to the chase and tells me the   information that I’d like to know. 

But what about the traditional election polls we’ve been relying on for decades — don’t they provide this information? Nate Silver, for instance, has gained fame by aggregating a variety of polls, weighting them according to a variety of criteria, and then creating a probabilistic forecast. So what can these markets do that Silver can’t?

STRUMPF: One of the things that he cannot do is, he can’t generate a real-time forecast. Like, okay, there’s a news story that comes out at one in the morning. Nate’s going to have to go collect a bunch of poll numbers. That usually takes a day for a poll to happen, and then run through his statistical model. And if this is a Monday, he’ll tell me Wednesday. And that’s sometimes not so bad. But these markets tell me Monday. One minute after the event happened, I can understand what’s exactly going on. The other thing to point out is that if you go look on these sites, we can have markets on anything under the sun. Nate Silver is one person. He can give us forecasts on the big stuff — who gets elected president, who wins Pennsylvania in the Electoral College. But suppose I’m interested in what’s going on with the mayor in New York City. 

In case you haven’t heard, New York City mayor Eric Adams is under federal indictment for a bunch of fraud and bribery charges.

STRUMPF: I don’t think Nate has a model for that. And if he has a model for that, he doesn’t have one for every city you could ever think of. He doesn’t scale very well. He can do one thing. He can do 10 things. Maybe he can do 100 things. But there’s thousands of these prediction markets on anything you’d ever be interested in. 

As it turns out, Nate Silver recently joined the advisory board of Polymarket, the biggest prediction market out there. So what happens if sites like Polymarket and Kalshi grow and grow, and if you really can make a bet on “anything you’d ever be interested in,” like Koleman Strumpf says? Wouldn’t that create even stronger incentives for political bribery and fraud?

STRUMPF: My claim is that those incentives all exist right now. The alarmist views of things that the C.F.T.C. would say, I don’t see a basis for this in history, in our modern experience, in anything. As somebody who has studied centuries of data from these markets, I can’t give you one piece of evidence. We’ve had a lot of very contentious things in our political system in the last 10 years. That had nothing to do with political prediction markets. I think the burden is on the people who make this claim to give an example, to give hard evidence, that this thing is going to happen. We just don’t see examples of this.

So Strumpf isn’t alarmed. But he does think there’s one industry that ought to be.

STRUMPF: I’m pretty interested in politics, but what I’m definitely not interested in is watching cable shows of people opining about what’s going on. CNN, Fox News is not in any broad sense different from ESPN. They want you to watch their show. So if they know something, they could tell you that. Or they could string it out and tell you in 30 minutes. That’s the benign part. The less-benign part is you might even think that they’re going to spin actual data or information in a way that’s consistent with their worldview. A market doesn’t delay information. A market doesn’t spin numbers. A market just gives you numbers. That’s a better way of learning information. The people who are probably most threatened by these markets are the media sites, because the media sites right now are intermediary to learning about polls, to learning about candidates, what they think about things. Newspapers and TV shows do great analysis. But if at the end of the day, I just want numbers, these markets cut through that and give me a direct line into what’s going on. That’s what I want. That’s what a lot of people want.  I think the big question is: Will people understand what these markets are saying if they become more prominent? I would say first, that you probably need some level of education, explaining what these markets are. But let’s just turn that same spotlight in the other direction. I actually would argue that most people don’t understand what polls are telling us. If I look at a poll and it says right now that the expected national voting numbers are, say, 50-50 — well, what does that mean? It doesn’t tell us very much about who’s going to get elected president, because, A, that’s based on a state-by-state contest, and B, it’s  just a guess of the total vote. We’re pretty sure that the Democrats are going to get more votes than the Republicans, but the Republicans could still win the Electoral College. Now, if you asked a typical reader of a poll “What do you take away from this number?,” you’d say, “Oh, this poll means that everything’s tied right now.” And I don’t think that’s the actual lesson.

So what do the betting markets say about the election between Trump and Harris? According to Polymarket, the race is essentially a dead heat. And the betting volume is rising fast — including some betting that could be the kind of manipulation that the C.F.T.C. is worried about.

STRUMPF: Because this is such a contentious election, I expect to see many billions of dollars bet on these markets.

Feel free to keep an eye on the Polymarket betting as the election approaches — also feel free to let us know what you thought of this episode; our email is radio@freakonomics.com. Thanks to Koleman Strumpf and Eric Posner for a pair of excellent conversations — and thanks to you, as always, for listening.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Theo Jacobs. Our staff also includes Alina Kulman, Augusta Chapman, Dalvin Aboagye, Eleanor Osborne, Ellen Frankman, Elsa Hernandez, Gabriel Roth, Greg Rippin, Jasmin Klinger, Jeremy Johnston, Jon Schnaars, Lyric Bowditch, Morgan Levey, Neal Carruth, Rebecca Lee Douglas, Sarah Lilley, and Zack Lapinski. Our theme song is “Mr. Fortune,” by the Hitchhikers; our composer is Luis Guerra.