This transcript was generated using AI automation and may contain minor formatting or transcription errors. Please refer to the original audio to verify specific quotes or context.

The Economics of It: Ryan Bourne with Kalshi CEO Tarek Mansour

Host: Ryan Bourne, R. Evan Scharf Chair for the Public Understanding of Economics, Cato Institute
Guest: Tarek Mansour, Co-Founder and CEO, Kalshi

Ryan Bourne: I’m Ryan Bourne, and this is The Economics of It. Today’s episode is the economics of Kalshi, with Kalshi’s CEO, Tarek Mansour. Kalshi is a prediction market letting people make bets on events like: Will the Federal Reserve raise interest rates? How much rain will fall next month? Will your sports team make the playoffs? The evolving bet prices on the answers to these questions give us a constantly changing indication of how probable traders, staking their own money, think certain outcomes are.

These prediction markets have grown rapidly. In July last year, Kalshi and Polymarket, the two big players, recorded around $2 billion in monthly trading volume. This July, that figure was $53 billion, more than 25 times as much.

For economists like me, the attraction of prediction markets is easy to see. Putting money behind a forecast encourages people to take being wrong seriously. Reams of studies now show they do as well or better than expert forecasters in the implied prediction of outcomes. And these markets enable firms to effectively hedge against certain risks to their businesses, whether that be heavy rainfall or certain public policy decisions.

But not everyone shares the enthusiasm for exchanges like Kalshi. Some see them as fuelling troublesome gambling, others as venues for insider trading. Many US states see them as competition for their jurisdiction over sports gambling and the revenues it brings. So are these platforms just gambling in financial clothing? Or do they have a more important social role? Can we trust their prices? And who has the right to regulate them, the federal government or the states?

Tarek Mansour is Kalshi’s co-founder and CEO, and he joins me today. So, Tarek, welcome to The Economics of It.

Tarek Mansour: Thanks for having me. I’m excited.

Ryan Bourne: So somebody opening your app, Kalshi, for the first time might look and see an opportunity to put a stake on a football game or something like that. That seems very familiar to many people who’ve perhaps staked some money on sportsbooks before. When somebody says, like, New York Attorney General Letitia James, that you’re just gambling, what’s the response? What’s different about prediction markets to traditional forms of gambling?

Tarek Mansour: You know, I mean, it’s very similar to what— maybe I’ll start with this, actually. I recently found this court opinion. It reads, essentially, “We’re clearly of opinion that dealing in these types of markets is void by the common law. For among other reasons, it is contrary to public policy. It is not only contrary to public policy, but it is a crime, a crime against the state, a crime against the general welfare and happiness of the people, a crime against religion and morality, and a crime against all legitimate trade and business. The species of gambling has become emphatically and preemptively the national sin. It’s proportional to an extent. It is immeasurable.” And they go on and on about how horrendous this activity is.

What’s interesting about what I just read is that it’s an opinion from 1888. And it’s not about sports prediction markets or prediction markets. It’s about grain futures. It’s an opinion that at the time was calling grain futures gambling, and they’re calling them sin. And these farmers are betting on the price of grain. At the time, society was used to thinking of these things as gambling, and why? Because you had the bucket shops, which were what the modern-day casinos or sportsbooks are like, where farmers would go in and bet on the price of grain. But there was a structural difference. There was a very important structural difference between the bucket shops and the company that this lawsuit was about, which is the Chicago Board of Trade, which is now CME, the largest derivatives market on the planet, the largest financial market on the planet, really.

And that structural difference is: Are you betting against a house, where the house sets the odds, and it sets them in such a way that if you win too much, if you’re a card counter, you get kicked out? Over time, if you lose, they make sure you keep losing by giving you promos and incentives. It’s a model that’s structurally solely about entertainment, where you’re transferring wealth from people to a casino. And that’s very different from an open marketplace, an open exchange, where the marketplace is a neutral place. The Chicago Board of Trade is neutral; it matches buyers and sellers that may have a disagreement of opinion. They’re settling a disagreement of opinion via trade, right? One way is to do it via debate. The other is you trade. If you think the price should be higher, you can buy. If you think the price should be lower, you sell. And the platform itself is neutral. It’s not making more or less money from the winners or losers.

And there’s a purpose for this. It’s price discovery. You’re creating a public good, which is that as the market essentially negotiates, as people are buying and selling these markets, you’re figuring out in real time what something is worth. At the time, it was: What is the price of grain in the fall? Then it was the price of metals. What is the price of a stock? In the case of prediction markets, it’s: What is the price of a future event that people care about or relate to?

When I think about Kalshi and prediction markets, it really is a financial market, just broader. One way to think about the stock market is as a specific instance of a prediction market. And so I think there’s a little bit of a misunderstanding. The purpose of a free market is aggregating knowledge that’s dispersed and surfacing it into light. Why do people look at our political forecasts? Because they’re accurate. And why are they accurate? Because there’s a lot of smart people out there in the world, and they research in their labs, they run their own polls, they do all sorts of things to acquire knowledge. And then they get rewarded for that knowledge by trading. Why is society rewarding them? Because they’re making us smart about the future. They’re helping us do this price discovery mechanism.

Those people are not allowed in a gambling business model. A gambling business model is not about acquiring knowledge or price discovery or anything like that. It’s, again, about entertainment, because the house loses out if people become too knowledgeable and keep winning on their platforms. They limit them, right? It’s a fundamentally different model. Now, obviously there’s risk. The expectation comes with risk, and that’s true for options, it’s true for futures, it’s true for prediction markets. But I think it’s a little bit of a misunderstanding of what these markets are.

Ryan Bourne: Let’s try and make this concrete for perhaps an audience that hasn’t had experience on your platform. Suppose there’s a contract on there asking whether the Fed will raise interest rates by half a percentage point at the next meeting, right? At the moment, that’s trading at 40 cents. So the market price is implying that there’s a 40% chance of that happening. If somebody buys that and that event comes to light, then they’ll win a dollar; they get a dollar return. So who are they actually buying from? And how does Kalshi make money from that?

Tarek Mansour: Yeah. So if you’re buying a market, it’s like you said, each event contract is a share, right? It’s a bit like an option. Historically, an option pays out if the stock is above a certain price or below a certain price, if you have an option on the stock. In this case, this option, really we call it a swap legally, pays out based on whether an event happens or not. It pays out $1 if the event happens and zero if not. You can buy one option or one share. You can buy a hundred of them. You can buy a thousand of them, however many you want.

Now, if you’re saying you want to buy a hundred shares at 40 cents, it’s a bit like you’re implying that you think there’s a 40% chance of the event happening. And so you basically are putting up $40, because it’s a hundred shares times 40 cents. And then somebody on the other side is selling you those shares. The way we really think about it is: if you’re buying Yes at 40 cents, someone on the other side is selling you the Yes at 40 cents, and it’s the same as them buying a No at 60 cents. So you’re buying these 100 shares from somebody else, and that could be anyone, anyone in the open market. When you guys match, agree on a price, you have a trade. And then whoever wins ends up getting the dollar, effectively.

Now, what’s interesting is most people don’t actually hold it to maturity. You might buy at a 40% chance, but then a series of events happens where the market price moves to 70 cents, and then you sell it for a 30-cent profit. You don’t wait till maturity. So inflation comes in higher than expected. Well, before inflation, if the market moves, if the expectation around inflation goes up, right? It’s a little bit like right now with the midterms. The odds of Republicans and Democrats winning the Senate are moving over time. People buy low and sell high often, and they’re basically trading. It’s an active market. I mean, that’s what they do.

And the way we make money: we just talked about $40 and $60, so there’s a hundred dollars of volume that just moved. We take around 70 basis points, so 0.7%. So we would be making 70 cents. We basically take close to half and half from each of the parties. So it’s just a transaction fee on each exchange. It’s like the New York Stock Exchange; it’s the exact same thing. And we’re indifferent to who wins or loses. It’s exactly how the New York Stock Exchange works.

Ryan Bourne: And how does that change— you kind of alluded to this in your first answer. How does that change your incentives for getting people to the platform in comparison to, say, traditional bookmakers?

Tarek Mansour: What we want is more transparency and more efficiency. Markets work because we want competition. Somebody says, “Hey, I’m willing to sell you these 100 shares at, let’s call it, 50 cents.” You say, “I’m willing only to buy them at 40 cents.” Then somebody else comes out and competes with the first person and says, “Well, I’ll do it at 48.” And you might say, “Okay, I’ll move to 42.” Then someone else comes in and says, “I’ll come down to 45.” That competitive thing that I just described is how markets usually work. It’s an order book. People are saying, “I will buy at a certain price,” and people are saying, “I will sell you at a certain price,” and they’re all competing against each other in some sort of open auction. It’s a bit like an auction that makes the prices more efficient over time.

So foundationally, how do prediction markets, or really any financial markets, work? You want a very large diversity of participants, which makes the prices more accurate over time and the markets more efficient. Competition is what we want: more and more participants, more volume, more liquidity gets those forecasts to be more accurate over time. And that’s really the key foundational thing of our business. The reason people trust our markets, when they look at it and ask, “Is it predictive of the future? If I want to figure out who’s going to win a certain election,” and so on and so forth, is because they work. And the only way for them to work is that they need a lot of volume, and they need a lot of volume from informed traders that are coming out and surfacing new knowledge and insight into the market.

So I’m always working out ways to attract a very large diversity of participants, setting up a uniform set of rules that are fair and equal across all my participants, because it’s a competition. I want everybody to have access to the same rights and obligations, and then figuring out how to always promote more and more sophistication. People are acquiring more and more knowledge over time. That incentive just doesn’t exist in a sportsbook. A sportsbook or a casino has one simple incentive, which is: I’m going to maximize the losses of my customers, because the losses of my customers are equal to my profits. That’s it. It’s very simple. And if somebody is too smart, I’m going to ban them. The card counter always gets kicked out of the casino, because they will take money from me, as a sportsbook, as the house. So that’s the different structure that leads to different incentives.

And so what do you do as a business, inevitably, as a sportsbook? You’re going to ban the winners, and then you’re going to keep sending bonuses and all sorts of ways to attract the losers to come back to your product. And that’s some of the stuff that’s created addiction issues; all these issues just come from a gambling business model. Whereas in an exchange, you’re going to promote fair, equitable rules that promote price discovery and liquidity. That’s really how you grow as an exchange and get more and more volume over time. So very different structures, very different incentives, which leads to different needs for regulatory structures around them. One you need nationwide, because you need a very large diversity of participants trading in an open, liquid exchange. And the other one you need localized rules for each state, to make sure the book is not tilting the balance too much against its own customers, that it’s not doing too many predatory things to its customers.

Ryan Bourne: So let’s talk about some of the social benefits of prediction markets. Economists are pretty open to them for some of the reasons that you’ve alluded to, right? We do see the prices that you generate as aggregating the dispersed knowledge that we have. It’s a very Hayekian, Friedrich Hayek kind of principle, about the wisdom of crowds bringing together the prices. And one consequence of that is that we imagine, because people have financial stakes in what they’re betting on, what they’re putting their money on, that people care a lot about their forecasts. And so as a result, the prices, or the forecasts, will tend to be more accurate on average than perhaps a panel of experts. I’ve seen this recently with the Federal Reserve. There was a paper that came out quite recently that showed that the implied predictions about the direction of inflation on platforms like Kalshi were far more accurate than Wall Street, the Bloomberg consensus of economists. Are there any other examples, actually empirical examples, of where your implied predictions have been much more accurate than perhaps experts or other information-gathering entities?

Tarek Mansour: There are many, many examples. There are a lot more papers being written about this. And I love going back to Hayek, because when I decided to start the company, I actually read a lot of Hayek. And I think Hayek is actually the founder of prediction markets. At the time, he didn’t really call them prediction markets; he called them information markets. And he talked about how market prices are a very good way to surface information, because markets incentivize people to acquire or seek knowledge and then use that knowledge to trade, and they get rewarded if they’re right, if their knowledge is high quality, if that piece of information they have is high quality. And they get punished if their piece of information is bad. And markets do a very good job at averaging all of that out.

Before getting to examples, I think there are two foundations, two reasons why that works. One is skin in the game, right? There’s a very easy way to think about that. Why is skin in the game good? It’s because it’s a very clean incentive towards truth. Talk is cheap. Talk is very cheap. I sometimes think about prediction markets as a tax on bullshit. If you write something on Twitter that’s completely false, what is the punishment for that, right? The example I like to think about is: it’s Thanksgiving and you’re having dinner, and you always have a relative that’s so certain about everything. They’re so dogmatic about every single view that they have. And then the minute you tell them, “Hey, how about you put $50 on this?” they get way more careful. There’s something that happens. “Oh, wait. Now I will win if I’m right, and I will get punished if I’m wrong.” They start saying, “Well, it’s conditional. It depends on this happening. It depends,” right? Exactly.

So there’s the skin-in-the-game piece, and then the second piece, which I think is really the foundation of prediction markets, is the wisdom of the crowds. And there’s something very nuanced about this, but it’s so interesting. The Federal Reserve put out a paper, the one that you mentioned, called “Kalshi and the Rise of Macro Markets.” They compared the Wall Street forecasts for economic indicators against Kalshi, which is the Main Street forecast, like average people, and Main Street outperformed Wall Street. So it says that Main Street actually is more accurate, and it’s real time, and it’s distributional. So it’s very useful to have.

And it’s so interesting, because when I was deciding to start the company, I had read Tetlock, and he had done some analysis—

Ryan Bourne: Philip Tetlock, yeah.

Tarek Mansour: He was an MIT professor, and I went to MIT. He had done a bunch of analysis on superforecasters, like who are these people that are very good at predicting the future? He did a bunch of competitions. For example, he pitted a bunch of geopolitical experts, elites, the big talking heads that have been in the field for decades, against a bunch of basically random people that like reading the news, and had them predict geopolitical events. And guess which group outperformed? The random people. And there’s something really powerful about that, because it tells us that actually domain expertise is not very relevant when it comes to figuring out what lies ahead. What’s relevant is the skill of forecasting, which is self-calibration. It’s thinking critically about the world, being in the business of seeking knowledge, and being in the business of actually caring more about being right than sounding right, or about ideology, right? And I think oftentimes the experts are a bit too ideological.

Ryan Bourne: They become convinced of their own theories, right?

Tarek Mansour: Exactly. In some way, the markets reduce the ideological burden that people usually have, because you have to be right. When I did my master’s at MIT, the joke in research circles was always that working in Wall Street and working in academia is mostly the same thing: you have to do the research, except that on Wall Street you have to be right, because you have to make money. But really, the thing that’s exciting about this is that we now have a hive mind, a critical mass of people that we can ask any question, and they will surface a very large diversity of information. It goes back to diversity of people being extremely critical for this market, because everyone may have different opinions and insights and bits and pieces of information, and you have to average out as much as possible. You cannot localize to just DC. If we just ran a prediction market where, let’s call it, decision makers in DC will weigh in, it would be less accurate than if you bring in someone in California, and someone in Texas, and someone in Hawaii. It is presupposing that that’s the only way prediction markets work.

We saw there’s a paper recently on the weather. I forgot which university compared our weather forecasts to the blended average, the max of all the weather forecasts across the nation, and our weather forecasts were 10% more accurate. It’s pretty amazing. How is that possible? Well, it’s possible because these users are doing all sorts of new things to figure out how to improve; the incentive for profit pushes them to push their boundaries, to ask how we could model the weather and how we could think about it more critically, which increases our knowledge about weather patterns and our understanding of weather, and so on and so forth. This is happening in politics, economics.

Ryan Bourne: I think that leads to a broader social benefit that often doesn’t get talked about much, which is that other people have access to what your markets are putting out in terms of the prices, even if they’re not directly trading. There are many people that can look at your market prices and use them to inform the decisions that they make about what the weather is going to be, or who’s going to win a particular election, or whether a public policy is going to go through. So that’s a kind of social spillover benefit that is beyond the traders themselves.

Tarek Mansour: That’s the spillover. So let’s go back to price discovery. When you look at how the commodities markets got regulated and got started, the foundation of price discovery is figuring out what something is worth, right? At the time, it was farmers, right? It was farmers thinking, “How much should I invest in my crops right now? I’m not sure how much I could sell in the fall, because it’s going to take me six months to harvest, and then I have no idea what the price is going to be, because prices are moving all the time.” And so the futures market on commodities, on grain and others, started because farmers wanted to have more predictability. They wanted to figure out, “Okay, this is the forecast from the futures market for what it’s going to be in September, so that will help me inform my decisions today.” And then that has gotten broader, right? We move into metals and oils, and then it moves into the banking sector, and we started having commodities or derivatives markets about a much wider set of things that are important to price.

What prediction markets do is they generalize that concept. They take it from all these different things and add more and more different types of markets and questions that are worthy of being forecast, being price discovered. And when you look at Kalshi today, 75 to 80% of our customers don’t trade. They’re active customers; they use it as a news feed. They use it to get informed by all sorts of things. What is the price of diesel going to be? Is it going to rain tomorrow or not? Is my home team going to win or not? One of the biggest sources of predictions we see is around elections and sports. It’s very simple things. Think about it the same way as the farmers making business decisions about what they should do. Bars in New York are basically deciding how much to stock up on inventory before the game, because if the home team wins, they know they usually sell three or four times more than when they lose. So that has a very immediate economic impact on how much they should stock up on inventory, because you never want to be at a bar where you don’t have enough beer when the team wins. Who would ever go back to that bar?

But think about that at a much broader scale. When you think about weather, when you think about bills passing or not, this idea of pricing risk and pricing economic questions more effectively is in some ways crucial to our resource allocation and decision making over time. We’re not going to be able to make good decisions if we don’t have a good gauge of where the future is going to land on things that matter to us. And I think prediction markets do an incredible job of fitting that function. That’s why people use them. Right now in politics, for example, most campaigns essentially use our markets religiously. Why? Because they are more accurate, and they are real time, and they’re very, very useful.

Ryan Bourne: So I’ve read previous interviews that you’ve done where you’ve talked about the origin story of the company, and you were an equity derivatives intern, and you were thinking, well, there are all these financial products out there, but actually they’re very imperfect for people trying to protect themselves against certain risks to their business, or for them as consumers. And I was really struck recently that even in what people think of as pure entertainment markets, like sports, for example, there are a lot of businesses out there using your platform to hedge against risk. One of the ones that I came across recently: Rizzo’s, a bar near Wrigley Field in Chicago, actually took out a hedge on your market. Their income is highly dependent on whether the Cubs go on a good playoff run. So they came and put a stake on because they were worried about the downside, that the Cubs wouldn’t do very well. They put a stake on the Cubs playing fewer than two home playoff matches. They put a small stake on that, and that helped mitigate some of the revenue lost from the Cubs not going on a good run. But this stuff is going on in a whole range of different markets and applications, whether it’s the stuff that you’ve talked about for particular businesses, or you’re a tourist site or something and you want to know whether your customers are going to come through the door. And presumably that will only proliferate as this technology develops.

Tarek Mansour: For sure. Because what you just described is the very definition of hedging. It’s exactly how hedging was defined for any commodity futures, for the very existence of the Chicago Board of Trade and the Chicago Mercantile Exchange, the CME. It’s this idea that you have some risk that is tied to something you don’t really control, and you buy a position that offsets some of that risk. Some people want to offset the whole risk; some of them want to offset some of the risk. What Kalshi is doing is essentially pricing a broader set of risks. It’s not just about what the price of grain is or what the price of oil is going to be. It’s about the price of all these different risks that we’re exposed to, and then you can buy offsetting positions. Some people will buy to speculate; they want to increase their exposure to an event. But a lot of people basically do it to reduce their exposure to an event, because they have calculated some exposure they’re going to be hit by if an event goes one way or the other.

And I think our function, when I was at Goldman, I thought about it like this: well, yeah, if you’re a big corporation, a Fortune 500, you can pick up the phone and call my boss’s boss at Goldman so that we can structure a product for you. But if you’re an average business or an individual, you don’t have access to any of these products, right? And I think this idea that you can actually shield yourself against some of the volatility that you don’t exactly control is very, very important for businesses and individuals, because what businesses need most is predictability. Businesses with very unpredictable cash flows usually get valued less than businesses with predictable cash flows. Why? Because if you have predictability, you know how to invest, you can raise more money, you can prepare for the future better. And so that’s what we’re bringing at the core: we’re bringing that notion of risk hedging to a much larger number of businesses and individuals.

And I love the sports market, because people are really trying to create a bifurcation between sports and economics, as if economics and weather markets are fine but sports is not fine. But I always find it interesting, because based on the exact same structure, there are people speculating on economics, or on weather, or on politics, or on financial market prices, and they’re speculating on sports. That exists across the board, and there are people hedging across all of them as well. The structure that I just described, on how these markets work, is the same. The structure doesn’t change. The topic could change; it could be sports, weather, climate, whatever. But the structure is not changing. It’s the exact same structure. And to me, this idea that sports doesn’t have economic consequence, or elections don’t have economic consequence, is ludicrous, right? When the FIFA World Cup was happening, advertisers were talking to us a lot about looking at the odds of the US making it, plus progressing. Because if you have to buy the advertising slots ahead of time, and if the US loses, the viewership basically goes down, and the dollar value of your advertising slot essentially gets halved immediately, purely based on that loss.

Ryan Bourne: You can make those arguments for marketers backing certain players.

Tarek Mansour: The one I love, I always like to go to the sports bar, because it’s something that’s so understandable and tangible to people. People get it, right? Sports bars do a lot of hedging on Kalshi against their home team losing, for the same exact reason. They make $2,000 on a night if the team wins, and if the team loses they make $800 on sales, but the cost is usually the same: they stock up on inventory, and they do $1,200. So in one night, if they win, they are up $800, and if they lose, they’re down $400. So what do they do? They take an offsetting position. They buy No on their home team winning. So they short their home team, so that if they lose, the revenue loss in their business is offset, at least partially, with the hedge, the insurance. It helps smooth their income, right? They usually smooth it out perfectly. They end up being plus $300 no matter what, and that’s better for them than the volatility. They have less exposure, and they can invest more decisively, which is a great outcome. And that’s why hedging is so important.

Ryan Bourne: So I think you’ve said some really important points. As a libertarian, I don’t object to people betting just for fun or for entertainment. But you’re embroiled in a legal battle at the moment here in DC, because states have traditionally regulated sports gambling as they’ve defined it, yet you are regulated by the federal government, because, as you say, you’re a federally designated exchange under the CFTC, the Commodity Futures Trading Commission. Now, this is quite a highly technical case, and I know some courts have ruled in favor of the states, other courts have ruled in favor of prediction markets, but recently 39 states and the District of Columbia urged the Supreme Court to take up New Jersey’s fight with you guys. So just outline to us: what is the strongest argument that the federal derivatives law, that the federal regulator, should really preempt these gambling laws? We talked about some of the economics of it, but in law, what do you think is the strongest argument for this?

Tarek Mansour: I mean, look, I believe in free markets, and if you think about it, it goes back to the Illinois opinion, which was a very interesting one. The TLDR is: these markets are swaps, it’s the same structure as anything else, it promotes price discovery in the same way, it’s structured as a financial exchange, there’s no difference whatsoever. Some of them just happen to be fun. And nothing presupposes in financial markets that, you know, some people do— do you believe that people trade zero-day expiry options for fun on Robinhood or on any of the other platforms?

Ryan Bourne: Do I believe that? I mean, certainly some people do, yeah.

Tarek Mansour: Does that mean we should regulate it as gambling? It’s the exact same thing. What is the line? There’s no limiting principle, there’s no sound basis in the law. And the interesting thing, so I go back to history: when the Chicago Board of Trade gets enacted, close to the same number of states was actually suing the Chicago Board of Trade, saying this is illegal gambling.

Ryan Bourne: Yeah. Now, I’ve read some New York Times articles from the period, yeah. They say that the Chicago Board of Trade is the biggest gambling house in the US.

Tarek Mansour: Exactly. So the notion that sports prediction markets are not like this, the idea that derivatives or financial markets have never been considered to be gambling, or that there’s some clear line in our heads, is not true, right? All these markets have been considered gambling at certain points. So how do you differentiate? What is the structural thing? The interesting thing about the Commodity Exchange Act, our law, the statute under which the CFTC operates and commodities futures operate: when you look at the history, it was enacted for the very purpose of preempting state gaming laws and state insurance laws. So when states come and say, “The CEA cannot possibly be preempting state gaming laws,” it’s a bit historical, because the very purpose of the statute was to preempt state gaming laws. Why? Because pretty much most of the things that were traded on financial exchanges, and are still traded on financial exchanges, have at some points been considered and viewed as gambling, because of the act of speculation.

So the line is very clear. Again, it says: Are you in an open marketplace that has a public good, where you’re promoting price discovery, and it’s a competitive order book? Or are you betting against the house? So there’s bucket shops and casinos, bucket shops from 100 years ago, casinos and sportsbooks from today, versus the Chicago Board of Trade, Nasdaq, Kalshi. That structural difference is the key piece. And here you could be speculating, you could be hedging, and the public good is the price discovery that comes out of these markets, and the prices are set by the market. That’s the key thing: supply and demand, open, free supply and demand is the thing that’s setting the prices.

And sure, some people now are claiming that they want to carve out sports because they think that’s a big market, but there’s no principled way to carve it out, and the law is actually structured in a way to help these markets evolve as the economy evolves. And sports now is a massive economy. I mean, private equity put $70 billion into sports in the last few years. The NIL bill just passed the Senate. This notion that there’s big dollars and big money in sports, and these athletes being paid these big dollars, and cities— you know, the next one in New York, Mamdani put out a press release about how this had a $200 million economic impact on the city. I think it’s a bit of an underestimate, actually. That’s real, that’s real economic, tangible impact. And so as you think about this industry becoming an asset class, what do you do with asset classes? You create financial products that are layered on top of that. And sure, some people are doing it for fun, but that doesn’t negate the fact that they’re financial instruments and they’re swaps. And nothing in the law, actually— the law is very clear. It says that if a swap relates to gaming, which you could argue sports is a game, the CFTC has the authority to decide what to do with it. And it’s very simple, it’s written in no uncertain terms, basically.

Ryan Bourne: Yeah. I mean, all financial markets have a combination of speculators and hedgers, as you kind of alluded to, otherwise it wouldn’t work, right? So sports is obviously a big part of the trading volume. If you look at some statistics, I think one month it was about 60% of your trading volume. I’ve seen other figures for Polymarket, one of your competitors, 70, 80% at certain times. So this isn’t a legal argument about an obscure product. I just wonder why you’d argue that this is so important to your overall offering. Is it because of some of the social benefits we talked about earlier? Is it because you have people coming to your markets through sports, and then they stake things in other areas? Is it just that it helps finance the cost of the broader infrastructure? Is it all of the above?

Tarek Mansour: I go back to thinking about this notion that sports has no economic consequence—

Ryan Bourne: I mean, you don’t have to convince me.

Tarek Mansour: But this is my answer, right? I’m reasoning through it in a principled way. Do you believe weather markets have economic consequence?

Ryan Bourne: Sure, yeah.

Tarek Mansour: Sure. So we should have them traded; we should have derivatives on those. Weather markets we have. Sports outcomes: do they have economic consequence?

Ryan Bourne: Of course, yeah. Of course.

Tarek Mansour: So we’ve discussed that we should have derivative markets on sporting events. I’m just trying to— so my point is that those markets, in and of themselves, are valuable, equally as much as a bunch of other markets are valuable. And when I look at how much hedging is happening on Kalshi every week, close to 60 or 70% of the hedges that get placed are on sports. The very purpose of these markets, risk mitigation, is on sports. So it should show us that, yes, maybe speculation is 60 or 70% of the volume, and that’s going down over time, but the hedging is as well. So there’s, again, there’s no structural difference between those two, except that it’s big business, and people have historically thought about it as, “Oh, there’s DraftKings and FanDuel and their sports betting.” So there’s a bit of that analogy that’s being created, that superficial analogy that’s being created in people’s heads. But then, yeah, I agree with you on the two and three. Number one, you need a critical mass of customers, and sports are a very good way to educate customers about how this works. You’re seeing it now with the growth of sports: we have 30 million customers in America, and that helps us actually promote liquidity. Now, if I want to make a market on, like, where will AI evolve, or how much job displacement is going to happen from AI in the next 10 years, I can do it in a way I couldn’t a year or two ago, because I didn’t have enough customers, I didn’t have enough liquidity. With these markets, liquidity begets liquidity. You need participants in a bunch of different things, and that begets participation in a bunch of different things. So I don’t think prediction markets really exist if you start carving out one or the other.

And then there’s a question of why it should be state versus federal. Why are financial exchanges regulated by the federal government rather than the states? It’s a very simple thing, again: if you’re creating a competition and you need a very large pool of customers, you want uniform rules across those customers. The easiest way I can really think about it is, imagine we broke down the New York Stock Exchange state by state. If you’re in New York, you can only buy Apple stock from New Yorkers. What would that do? It would bifurcate liquidity, fragment the whole market. Liquidity would get fragmented. You would have different prices for Apple stock in different places, and prices would get worse, because now you have less competition, you have fewer counterparties for the thing you want. It’s an extremely inefficient and anti-market structure. What you want in markets is free and open competition with a uniform set of rules. And so these markets cannot exist if you fragment them and bifurcate them.

Ryan Bourne: Yeah, no, I mean, you don’t have to convince me.

Tarek Mansour: No, no, I know. Unfortunately, you know, there’s a lot of people, for whatever reason—

Ryan Bourne: Well, I think there’s a coalition. The reason is very simple: there’s a coalition here, right? There are people who are just morally against gambling, and they think sports gambling is something that younger people or whatever are more likely to engage in, so it gets them hooked or whatever on staking money. And then there are the state governments, who are trying to protect their own regulatory power and, in some cases, their tax revenues as well. So it’s like a bootleggers-and-Baptists coalition, as we describe it in economics, like the two groups that were in favor of alcohol prohibition: the bootleggers, who actually profit from it, and the Baptists, who are opposed to people drinking alcohol. That’s the kind of dynamic.

Tarek Mansour: I don’t think— I think you missed the biggest bucket of them all. I think it’s the casinos, and the—

Ryan Bourne: Well, they’re the bootleggers in this, yeah.

Tarek Mansour: Yeah. In the gap, where, you know, you mentioned— so look, I think for the first bucket, it’s not a choice of whether there’s going to be sports betting or not. If the sound principled position is people saying, “Hey, let’s ban sports betting altogether in America,” we can have that conversation. But the idea that sports prediction markets should not exist but sports betting apps should exist, I think you would agree, is ludicrous, right? Because sports betting, we know, their incentive structure is to promote addiction and promote all the issues that have been happening to young men. And I do think the young men issue is real, and that’s why I’m pro-regulation. I think we should establish the right guardrails, and I think prediction markets are a much healthier model, because I don’t have nearly anywhere near the incentive of promoting addiction. The addictive-type customer is very profitable for the sportsbook; for me, they’re losing money to somebody else. So I could curb that activity and be more willing to curb that activity, and we do, than the sportsbook.

So the point here is, it’s not a choice. I always talk about the morality. Look, no matter what you think, I think these markets are great. I think personally people trading on sports, it’s statistical. It’s like Moneyball. Especially our core customers: 80% of volume comes from, like, Jonah Hill in Moneyball. They’re doing the stats and the math, and they do it on the weather, and they do it on politics and economics and sports, and often they’re the same customers. They’re very sophisticated. In The Big Short, the two kids that had these monitors in the basement and they’re doing that: that’s the core customers. Those customers would never be allowed in a sportsbook, right? So I personally think these markets are amazing in that they make us understand sports, understand politics better. Again, it goes back to price discovery.

Now, if you think sports betting, or sports trading, even, because of the speculation on sports, is bad, to me I always think: you cannot think that and think that retail speculation in zero-day expiry options is okay, because we clearly see a lot of addiction in those markets. And you certainly cannot think that prediction markets sports trading is bad and decide that the right policy solution is to give a monopoly to the sportsbooks, which have shown themselves to be—

Ryan Bourne: So there’s one economic study, admittedly of Polymarket, which you don’t have to answer for, that found that the top 1% of profitable users captured about 77% of the profits. So it’s unfair to put you on the spot for Polymarket’s numbers, but do you have any sense of what proportion of your users are profitable, and what those traders tend to look like?

Tarek Mansour: It’s interesting to me, that question is really interesting to me. Let me maybe reverse the question a little bit. The critics love to use that stat, and you’ve seen it a lot, posted on Twitter by the casinos and their coalitions all the time. But what do you think it should be?

Ryan Bourne: I have no idea.

Tarek Mansour: Let’s assume it’s 50/50: 50% of people win, 50% of people lose. I would argue that would be weird and quite bad. It would mean there’s no skill in this game; it’s pure chance. There’s no real price discovery. It is a skill-based competition, like any form of financial trading. And maybe another way I’ll think about it: what percentage of hedge funds outperform the S&P every year?

Ryan Bourne: Very low.

Tarek Mansour: Twenty to 30%. So 20 to 30% are the winners on Wall Street, in that competition. That is Wall Street, that is generally zero-sum, because you’re benchmarking against the S&P, and 70 to 80% are, let’s call it, net losers, right? And in financial markets, on Kalshi, we’re like 65/35. It’s interesting, right? If it were 50/50, we’d have a problem. It means there’s no skill, there’s no point in that market, because you don’t have people that out-compete other people and do better over time. And so I don’t know about this 1%. I don’t know how that statistic was arrived at. I don’t think it’s that tilted, frankly. But the point I’m trying to make is, whatever number you’re going to find, prediction markets are actually going to be a bit less tilted than Wall Street. Why? When you ask our core customers, “Why do you prefer trading on Kalshi over Wall Street?” it’s because a lot of people feel like when they trade zero-day expiry options, Citadel has way more asymmetric information than them, so they cannot actually do a good job. Whereas on Kalshi, who’s our best inflation forecaster? There’s an article about him on Bloomberg a day or two ago, and it’s a retired movie theater veteran. I mean, it’s amazing. This person never traded in financial markets, but loves reading about inflation and the economy. And you see this, like, our best guy is actually a person that lives in Kansas. A different person, but there was another article about this different person that trades inflation. But my point is, it’s a bit of an equalizer. It’s a bit like we’re broadening the set of things so that they relate to a much larger set of people. Some people are political junkies; they read about politics, they like to read about candidates, and they are outperforming the big funds more so than they can in traditional financial markets. So I think that’s the right frame. Again, if it was 50/50, it would have a real problem; there would be no skill in this.

Ryan Bourne: Yeah, no, I think the context in which people use the statistics is trying to play on people’s fears about insider trading and things like that. So, you know, there have been some stories, like we had the Bad Bunny dancers who were trading on what songs, you know, what other acts would be part of the halftime show.

Tarek Mansour: I don’t know if that’s true, by the way.

Ryan Bourne: There were stories about that.

Tarek Mansour: I don’t think there were stories.

Ryan Bourne: I’m just saying there are a lot of different stories.

Tarek Mansour: I mean, you know, there was a story on Friday about potential insider trades on the White House secretary, and they were claiming that the $19 trades were insider trades. It’s like, I can’t comment on any ongoing investigation, but you can make the conclusion of— I mean, the point is, look, I think the percentage is the same as in traditional financial markets, and I don’t really think insider trading is influencing these at all. I just think it’s an attack line to make it sound like, “Oh, people are losing money, so it must be bad.” But that’s just a misunderstanding of a competition. I always say, what percentage of teams win the NBA Finals every year?

Ryan Bourne: Well, it’s one team, you know.

Tarek Mansour: Yeah, yeah, but my point is, I don’t know what percentage that is. The point is it’s a highly competitive competition. I mean, that’s the point of a competition: the people that do better jobs, that do more research, that pay money to get different data sources to really lock in, will do better than people that do less research and don’t do as good a job. That’s the beauty of a market. You need that diversity of participation.

Ryan Bourne: You mentioned insider trading. I mean, I think Kalshi’s position has been very clear.

Tarek Mansour: We are very anti-insider trading. Because, you know, I always think that if there is corruption in the market, it’s very anti-free market.

Ryan Bourne: So do you mean information that people are contracted not to use because it’s illegal, or do you mean something beyond that, which is, if they’re party to certain information which is non-public—

Tarek Mansour: Yes. And they shouldn’t use it. So here’s the line, and it’s the same line that is drawn in the stock market: what information can you use and what can’t you? How do you draw the line? It’s the exact same line, and we uphold the same line. We use the same systems to surveil it, because usually when people come in with insider trading, the patterns are always suspect; they’re weird. Is the information material non-public information? And what does that mean? It’s information that you’ve acquired by virtue of your unique position, some sort of privilege or special access, because you work in government, or because you’re part of an executive team. Your job, your role, gives you that specific privilege and that specific information that you’re not supposed to disclose. So with that specific information that you have, can you go on CNBC and publicly say it? If the answer is yes, that’s not insider information. But if the answer is no, that’s insider information, and you should not trade on it, because trading is a form of disclosure, right? If you know that Tesla deliveries are going to be higher than expected, you should not buy the markets on the Tesla deliveries, because if it’s trading at 10% and you buy it to move it to 20%, that’s a way to disclose that piece of information, right? But that’s a line, right?

Ryan Bourne: Yeah, I understand your position. But from the purpose of these being information delivery mechanisms, though, doesn’t that get us closer to the true market price? Isn’t there an argument—

Tarek Mansour: It’s a great question. I mean, Robin Hanson, who’s been a pioneer of prediction markets, argues that about insider trading. I could argue— I go to the stock market as a specific instance of a prediction market, so that same logic should in theory apply to the stock market. Wouldn’t the price of Apple be more efficient, more correct, if you let the insiders trade on the price of Apple? In theory, yes. But what happens actually, in practice? In practice, if you let the insiders trade, people stop trading. And your incentive is to have more people trading on the platform. You want a healthy, fair— you need to figure out a set of rules that are fair. If people are trading on Apple stock, on the price of Apple, but they never know if someone on the other side could have inside information that they could never have access to, no matter how hard they work, no matter what research they did, they will never trade. They will just not participate in the market, because the market is inherently unfair; it’s tilted.

Ryan Bourne: But you acknowledge there’s a bit of a trade-off.

Tarek Mansour: Certainly, there’s a bit of a trade-off, and that’s where regulation is a delicate line. I think actually the securities laws do a great job at this, and the Commodity Exchange Act, over time, will, I think, also do a pretty good job at this. It’s a healthy balance. But again, it’s: could you have access to that information without having that special, privileged position? If the answer is yes, then sure. It’s like when, what is it called, Two Sigma used to get satellite images of the cars parked in the parking lot outside of, I forget, Target or Walmart, to forecast the sales. That’s totally fine. That’s not inside information, because they did their own research and figured out a smart way to estimate what the sales are going to be and correlate them with the number of cars. That’s amazing. They didn’t call someone who works at Walmart and say, “Hey, can you read to me what the next quarterly earnings is?” which would have been totally insider information. And there’s a clear line there that’s supposed to exist so that the markets are fair. And I think markets cannot be fair if you let insiders trade.

Ryan Bourne: And, I mean, to be fair to you guys, none of this stuff is unique to prediction markets, right? So where I’m from in the UK, somebody was hauled in by the authorities because he made a bet on when the general election in the UK would be, because he had that inside information as somebody working in Number 10 Downing Street. And he did that through an ordinary betting site. This wasn’t through a prediction market.

Tarek Mansour: Well, it exists in the stock market too. I mean, insider trading happens in the stock market all the time, right? So it’s not a unique problem. I think what’s unique about prediction markets is that it’s more transparent, because all of our trading is transparent, so people see it way faster, way more effectively. So I don’t think the percentage of trades that are insider trading on prediction markets is higher than in the stock market. I don’t think so at all. Actually, I think significantly lower. I just think that we see it more often because it’s so clear and transparent.

Ryan Bourne: So we’ve, of course, got the midterms coming up in just over a week’s time. How excited are you about the kind of products that have developed, and the applications of prediction markets, to those midterms?

Tarek Mansour: It’s a really exciting time, because even though the 2024 election was when prediction markets went mainstream, I feel like this is the first true prediction market election. The reason is, you’re seeing now it’s become commonplace for candidates to talk about their predictions, and pretty much most, if not all, campaigns are using prediction markets to inform their decisions and really run their campaigns, because people are trusting it. It’s becoming more accurate, and so on and so forth. And I’m really excited because this year we actually launched— you know, prediction markets are yes/​no questions: Who’s going to win the Senate? Who’s going to win a specific seat in the House? Who’s going to win the presidency? It’s a yes or no question, or people are predicting whether that specific discrete event is going to happen or not.

Now, not all questions are discrete, and this is also another one of those powers of prediction markets over time, or really markets in general. We’re starting to move towards questions that are a little bit more vague in nature and don’t have a yes or no answer at a specific time. One question I’ve been very excited about is: Are we leaning more Republican or Democrat over time? What is the balance of power? What is the projected balance of power over time? And so we’ve created an index, a little bit like the S&P, but for politics. Instead of the index averaging the prices of a number of different stocks, which is a measure of how the markets and business are doing in America, it’s averaging the predictions from all our prediction markets on the presidency for ’28, on the midterms, for the House, the Senate, the Supreme Court, and so on and so forth, to show us what people are projecting as the balance of power, and how the balance of power is going to shift. It moves between, like, plus 50 Republican to plus 50 Democrat, and it’s this ever-moving metric that shows us how the world is tilting one way or the other as you progress.

What’s beautiful about that is, because it’s a little bit like the S&P, when a piece of bad news happens, like the Fed raises interest rates, you can immediately see what the impact is on business, because you have an aggregate that shows you this move. And that’s the same thing now in politics. When a policy gets enacted, when a big bill passes in Congress, now you have a metric that you can go to and see immediately: Okay, was that actually positive for Republicans, which are currently in power? Is that positive for Democrats? What’s exciting about that is it’s going to push us to have a better understanding of the inner workings of our politics over time: what people care about, what the policies are that move things one way or the other. And so I expect a lot of big swings for KPOW, the Kalshi power index, over the next few weeks.

Ryan Bourne: Well, Tarek, that’s all we’ve got time for today. Of course, here at Cato, we believe in the personal liberty to stake your money as you see fit and have access to these new financial applications. But I think prediction markets are so much more than that. For us, they’re a great teaching tool about the importance of market prices, how market prices arise, and how they are, to quote Alex Tabarrok, “a signal wrapped in an incentive.” Not only does the price tell us something, but it informs us about how we should plan our lives from that probability. So it’s a really exciting industry, and we really appreciate you coming in today to talk to us. So thank you again.

Tarek Mansour: Thanks for having us.

Ryan Bourne: So that’s all for today’s episode of The Economics of It, a production of the Cato Institute, a think tank dedicated to advancing individual liberty, limited government, free markets, and peace. Don’t forget to subscribe wherever you get your podcasts, and if you like what you’ve heard, please give us a decent rating on whichever platform you found us on. I’m Ryan Bourne. Until next Thursday, goodbye.