Thanks. Nice article. At InsightPrediction.com, due to fear of the CFTC, we had to block US IP addresses. Makes it hard to grow and uneconomical as a business. Most prediction market users are in the US. You'd need the ability to handle credit card deposits, and to trade on sports and elections to be profitable. And free range to do policy-relevant markets in real time. Basically, all the advantages Intrade had. Without all those features it's hard.
Will be mentioning this in my newsletter tomorrow at Nobull.blog. I've created a sort of template for messaging your senators / reps if you all would like to use.
Dear Mr./Mrs ____:
Unfortunately, due to a forthcoming issue with the CFTC, the usage of prediction markets in the United States will be suspended. Prediction markets, in my opinion, can be a really useful tool for Americans to not only engage in but also utilize to assess things like international events, financial markets, etc. I'd love it if you could look through the legal justifications for the CFTC's decision to outlaw these companies since I don't think prediction markets—which are highly accurate—are "gaming" in the same sense as slot machines and other traditional casino games. Visit my friend Richard's blog about prediction markets at https://richardhanania.substack.com/p/how-to-legalize-prediction-markets to learn more about them and why they are important to me.
On an unrelated note I don't understand why I'm the only person who seems to like their own comments. I mean, I usually agree with what I've said so why not like them. I figure that's why reddit seems to do that by default.
There’s a clear technical issue here that you don’t mention but which may represent a great compromise that gets the foot in the door: financial regulators have a strong preference for securities with continuous payouts, and a strong bias against securities with discontinuous or “binary” payouts, for good reasons related to manipulability and fraud. Predictions such as vote share (or electoral vote share denominated in pennies) are continuous enough and then options on them would also qualify. You could have a market on when an official would be fired/resign rather than “would he still be in his job on day X”, which would pay more the sooner he left, and expired worthless in 2 years, etc. or you could bet on climate or energy variables (maybe not crime rates for some crimes though because of the war/terrorism/assassination provisions).
From some friends in the industry, I'm under the impression that prediction markets don't look that appealing to professional finance, and this is a bigger problem than the regulatory ones. (To be clear, would still like them unregulated). See Zvi here: https://thezvi.wordpress.com/2018/07/26/prediction-markets-when-do-they-work/
It doesn't matter what finance people think under conditions where prediction markets are banned. The entire calculus changes when they're legal, and up and operating themselves. We'd soon have more data than one can imagine on their effectiveness, and all kinds of opportunities will be created.
Yes, my sense is that many of those concerns are specific to the case where the law keeps the markets under capitalized and insufficiently liquid so there is no incentive for big players to build quantative models to turn predictably irrational behavior into profit.
I have no idea, but those sites suck. Maybe non-Americans just can't do business or tech well. PredictIt is actually good, and would be a lot more liquid if it were allowed to be.
As much as I'd love to see prediction markets made legal that law review arguing they are protected by the first amendment is a load of crap (re: 1st...the analysis that they aren't gambling is fine).
For instance, it suggests the bc court rulings have held that code can be speech or that certain regulations limiting financial advice impinge on the 1st amendment that we should therefore regard the market price signal itself as speech. And further seems to suggest that because there is some expressive conduct involved in buying a contract in a prediction market it is therefore protected under the 1st amendment.
That obviously proves too much as all the same arguments apply to bank robbery. Robbing a bank has expressive content (it might express a belief that the system is corrupt or at the very least that I should have more money) and you could equally well alledge that the rate at which bank robberies occur conveys a message about the overall degree of dissatisfaction with our system or greed (not a very useful message but that's not the point...once we are weighing costs/benefits we are likely outside clear 1st amendment protection).
Ultimately, if the 1st amendment allows gambling to be rendered illegal it doesn't protect prediction markets. If you want to argue that the current rules against prediction markets are illegal far better to either focus on the text of the law delegating authority or on whether the rule against them meets the rational basis test.
The primary issue that concerns me about prediction markets (while still in favor of legalization) is the incentive they create to keep data and analysis private.
If I'm Nate Silver, a pollster or even an academic researcher currently it's in my interest to publish, share my reasoning and to some extent and use that to increase my reputation. With a highly liquid and well-capitalized prediction market that trades off against an incentive to instead keep methodological advances (eg the details of my prediction model) secret to gain a trading advantage.
True, you might counter that the market itself then reveals that information in the form of a price signal. However, that ignores the fact that it's often more powerful to aggregate data/methodology than to consider them separately.
I'm not saying I'm convinced they would be harmful. I'm still broadly in favor of legalization here. I'm just saying it's a reason to temper optimism about their benefits.
You have an incentive to use secret information to establish a position, but then to release that information, because doing so brings the market closer to what you think is the accurate value and therefore allows you to cash out without waiting for market resolution, which has obvious benefits in terms of time value of money and being able to avoid the risk of losing everything to a black swan when you were right about the actual probabilities.
Of course there's still an incentive to keep methodology secret, but how much you can do that is limited by your incentive to persuade people that your conclusions are accurate, which can be done better if you release more information about your methods.
More specifically, if going the legislative route, someone really needs to get members (ideally the chair and ranking member) of the House and Senate Agriculture Committees on board. On the bright side the Farm Bill is up again this year… on the downside you need to hurry.
Edit: yes, that’s right the Agriculture Committees… and this isn’t exactly their normal world so it’s a difficult issue to get folks there excited. The way “secret Congress” works it really is crucial that all 4 of the leading members of these Committees is on board, otherwise it’s likely not going anywhere.
Kalshi is a legal prediction market, but they are quite limited in what they can offer, and appear to be losing money at the moment. If they are allowed to have elections, it would likely push them into profitability.
It would also be worthwhile to note that Kalshi is almost certainly going to be lobbying against legalization, since that would eliminate their monopoly, which they no doubt spent a lot of money to obtain.
There's no guarantee of that. They still have to fight the CFTC to have election markets, and would probably benefit from being able to operate more freely. Either way, it doesn't change what the right path is.
You're just differently cynical. Given they are losing money and still have a bunch of overhead and regulatory compliance burden but are still limited in how much they can allow an individual investor to trade I suspect they might have an interest in legalization.
If things stay illegal they'll be at best a small company and at worst out of buisness. If all the restrictions are removed there is a very real chance they become a 100 billion dollar company or at least a 10 billion dollar one.
I wonder if there’s a possible compromise that allows only accredited investors or institutions to invest. This might not be ideal but might be easier to sell to legislators or regulators.
If you don’t have a following but do have expendable income, you can also trade money for influence by paying people with followings to share this article if they wouldn’t otherwise.
A doubt I have about prediction markets is that it seems they can create a situation where we never take any moonshots, due to their low odds of success now having a Seal of Truth from prediction markets. That seems like it could breed stagnation. Contrarian cranks often move things forward (e.g. Linus Torvalds), but then again, listening to the libertarian intuition, maybe we would all be better off with a largely inert government, a government that doesn't dream big or make bold moves? It's an interesting question.
Great article and very timely. (Another route is for illegal prediction markets to make enough money that they can simply lobby Congress to change the law.) I'll devote some pro bono time to the issue this year. Cheers.
Do you think that would be successful or would it instead prompt even bigger players in the financial industry to see them as potential threats and to lobby against them?
If I'm a dominant player in brokering consumer trading in traditional securities I might very well see the rise of a new form of short term investment as a potential threat to my business model. Especially given the fact that prediction markets are usually structured in a way that lets consumers invest directly without needing a broker to serve as middleman.
True there are opportunities for expansion into a new market, eg, offering the ability for ppl to take leveraged positions or use their stock as collateral but I don't think it's that unreasonable to guess that the appetite for short term trading is relatively fixed and that the more interest shown in prediction markets the more they represent a potential avenue for competitors to challenge my dominant market position.
I think the better strategy is, rather than creating a market outside the law and trying to get it declared legal, is to get ppl in the financial industry onboard with potential projects, eg, get Goldman onboard with a potential prediction market so they're confident that their lobbying will result in profit for them rather than turning an existing illegal market into a legal competitor.
Thanks. Nice article. At InsightPrediction.com, due to fear of the CFTC, we had to block US IP addresses. Makes it hard to grow and uneconomical as a business. Most prediction market users are in the US. You'd need the ability to handle credit card deposits, and to trade on sports and elections to be profitable. And free range to do policy-relevant markets in real time. Basically, all the advantages Intrade had. Without all those features it's hard.
Will be mentioning this in my newsletter tomorrow at Nobull.blog. I've created a sort of template for messaging your senators / reps if you all would like to use.
Dear Mr./Mrs ____:
Unfortunately, due to a forthcoming issue with the CFTC, the usage of prediction markets in the United States will be suspended. Prediction markets, in my opinion, can be a really useful tool for Americans to not only engage in but also utilize to assess things like international events, financial markets, etc. I'd love it if you could look through the legal justifications for the CFTC's decision to outlaw these companies since I don't think prediction markets—which are highly accurate—are "gaming" in the same sense as slot machines and other traditional casino games. Visit my friend Richard's blog about prediction markets at https://richardhanania.substack.com/p/how-to-legalize-prediction-markets to learn more about them and why they are important to me.
I sincerely appreciate your time.
On an unrelated note I don't understand why I'm the only person who seems to like their own comments. I mean, I usually agree with what I've said so why not like them. I figure that's why reddit seems to do that by default.
There’s a clear technical issue here that you don’t mention but which may represent a great compromise that gets the foot in the door: financial regulators have a strong preference for securities with continuous payouts, and a strong bias against securities with discontinuous or “binary” payouts, for good reasons related to manipulability and fraud. Predictions such as vote share (or electoral vote share denominated in pennies) are continuous enough and then options on them would also qualify. You could have a market on when an official would be fired/resign rather than “would he still be in his job on day X”, which would pay more the sooner he left, and expired worthless in 2 years, etc. or you could bet on climate or energy variables (maybe not crime rates for some crimes though because of the war/terrorism/assassination provisions).
From some friends in the industry, I'm under the impression that prediction markets don't look that appealing to professional finance, and this is a bigger problem than the regulatory ones. (To be clear, would still like them unregulated). See Zvi here: https://thezvi.wordpress.com/2018/07/26/prediction-markets-when-do-they-work/
It doesn't matter what finance people think under conditions where prediction markets are banned. The entire calculus changes when they're legal, and up and operating themselves. We'd soon have more data than one can imagine on their effectiveness, and all kinds of opportunities will be created.
Yes, my sense is that many of those concerns are specific to the case where the law keeps the markets under capitalized and insufficiently liquid so there is no incentive for big players to build quantative models to turn predictably irrational behavior into profit.
Why doesn't Smarkets or BetFair do more volume?
I have no idea, but those sites suck. Maybe non-Americans just can't do business or tech well. PredictIt is actually good, and would be a lot more liquid if it were allowed to be.
As much as I'd love to see prediction markets made legal that law review arguing they are protected by the first amendment is a load of crap (re: 1st...the analysis that they aren't gambling is fine).
For instance, it suggests the bc court rulings have held that code can be speech or that certain regulations limiting financial advice impinge on the 1st amendment that we should therefore regard the market price signal itself as speech. And further seems to suggest that because there is some expressive conduct involved in buying a contract in a prediction market it is therefore protected under the 1st amendment.
That obviously proves too much as all the same arguments apply to bank robbery. Robbing a bank has expressive content (it might express a belief that the system is corrupt or at the very least that I should have more money) and you could equally well alledge that the rate at which bank robberies occur conveys a message about the overall degree of dissatisfaction with our system or greed (not a very useful message but that's not the point...once we are weighing costs/benefits we are likely outside clear 1st amendment protection).
Ultimately, if the 1st amendment allows gambling to be rendered illegal it doesn't protect prediction markets. If you want to argue that the current rules against prediction markets are illegal far better to either focus on the text of the law delegating authority or on whether the rule against them meets the rational basis test.
The primary issue that concerns me about prediction markets (while still in favor of legalization) is the incentive they create to keep data and analysis private.
If I'm Nate Silver, a pollster or even an academic researcher currently it's in my interest to publish, share my reasoning and to some extent and use that to increase my reputation. With a highly liquid and well-capitalized prediction market that trades off against an incentive to instead keep methodological advances (eg the details of my prediction model) secret to gain a trading advantage.
True, you might counter that the market itself then reveals that information in the form of a price signal. However, that ignores the fact that it's often more powerful to aggregate data/methodology than to consider them separately.
I'm not saying I'm convinced they would be harmful. I'm still broadly in favor of legalization here. I'm just saying it's a reason to temper optimism about their benefits.
You have an incentive to use secret information to establish a position, but then to release that information, because doing so brings the market closer to what you think is the accurate value and therefore allows you to cash out without waiting for market resolution, which has obvious benefits in terms of time value of money and being able to avoid the risk of losing everything to a black swan when you were right about the actual probabilities.
Of course there's still an incentive to keep methodology secret, but how much you can do that is limited by your incentive to persuade people that your conclusions are accurate, which can be done better if you release more information about your methods.
More specifically, if going the legislative route, someone really needs to get members (ideally the chair and ranking member) of the House and Senate Agriculture Committees on board. On the bright side the Farm Bill is up again this year… on the downside you need to hurry.
Edit: yes, that’s right the Agriculture Committees… and this isn’t exactly their normal world so it’s a difficult issue to get folks there excited. The way “secret Congress” works it really is crucial that all 4 of the leading members of these Committees is on board, otherwise it’s likely not going anywhere.
Great article overall, but weird to keep claiming that prediction markets are illegal while ignoring that Kalshi is a prediction market that is legal.
Kalshi is a legal prediction market, but they are quite limited in what they can offer, and appear to be losing money at the moment. If they are allowed to have elections, it would likely push them into profitability.
Sure, but that’s different from the blanket statement that prediction markets are illegal.
It would also be worthwhile to note that Kalshi is almost certainly going to be lobbying against legalization, since that would eliminate their monopoly, which they no doubt spent a lot of money to obtain.
There's no guarantee of that. They still have to fight the CFTC to have election markets, and would probably benefit from being able to operate more freely. Either way, it doesn't change what the right path is.
Wow, I'm more cynical than Richard! I hope you're right about that and I agree that pushing for legalization is still the right thing to do.
You're just differently cynical. Given they are losing money and still have a bunch of overhead and regulatory compliance burden but are still limited in how much they can allow an individual investor to trade I suspect they might have an interest in legalization.
If things stay illegal they'll be at best a small company and at worst out of buisness. If all the restrictions are removed there is a very real chance they become a 100 billion dollar company or at least a 10 billion dollar one.
Surely Kalshi could make way more money as one of several players in a vastly larger market than they can as the only player in a tiny market.
Great post!
I wonder if there’s a possible compromise that allows only accredited investors or institutions to invest. This might not be ideal but might be easier to sell to legislators or regulators.
Most balls that should be rolling are idling motionless. This one deserved your push.
If you don’t have a following but do have expendable income, you can also trade money for influence by paying people with followings to share this article if they wouldn’t otherwise.
A doubt I have about prediction markets is that it seems they can create a situation where we never take any moonshots, due to their low odds of success now having a Seal of Truth from prediction markets. That seems like it could breed stagnation. Contrarian cranks often move things forward (e.g. Linus Torvalds), but then again, listening to the libertarian intuition, maybe we would all be better off with a largely inert government, a government that doesn't dream big or make bold moves? It's an interesting question.
Great article and very timely. (Another route is for illegal prediction markets to make enough money that they can simply lobby Congress to change the law.) I'll devote some pro bono time to the issue this year. Cheers.
Do you think that would be successful or would it instead prompt even bigger players in the financial industry to see them as potential threats and to lobby against them?
If I'm a dominant player in brokering consumer trading in traditional securities I might very well see the rise of a new form of short term investment as a potential threat to my business model. Especially given the fact that prediction markets are usually structured in a way that lets consumers invest directly without needing a broker to serve as middleman.
True there are opportunities for expansion into a new market, eg, offering the ability for ppl to take leveraged positions or use their stock as collateral but I don't think it's that unreasonable to guess that the appetite for short term trading is relatively fixed and that the more interest shown in prediction markets the more they represent a potential avenue for competitors to challenge my dominant market position.
I think the better strategy is, rather than creating a market outside the law and trying to get it declared legal, is to get ppl in the financial industry onboard with potential projects, eg, get Goldman onboard with a potential prediction market so they're confident that their lobbying will result in profit for them rather than turning an existing illegal market into a legal competitor.