Chris Hayes is a saint.
Hayes, the host of All In with Chris Hayes on MS NOW, was on The Late Show with Stephen Colbert a few weeks back. He talked, he chatted, he kibbitzed at the late afternoon taping.
Not long after the taping, a market appeared on Kalshi, specifically asking what Hayes would say during the interview.
It ended up being a market with over $1 million traded. To give some semblance of that number, a similar market appeared for President Donald Trump’s recent appearance on FOX News. That garnered about a third of the trading volume, so yeah, Hayes’ market clearly caught fire.
As well it should have, as what Hayes said was filmed in front of the entire Late Show staff, not to mention a few hundred people in the audience. They all had about seven or so hours to get their bets in on what Hayes would say. This was easy money, as Hayes already said what he was going to say.
But you know who didn’t get their bets in? Chris Hayes and his friends and family.
“And obviously I’m not going to go in and place a bet or tip my friends off to it. That would be unethical,” Hayes said in an Instagram post. “But a lot of people would. And for a lot of the bets that are on Kalshi it just seems like these are essentially opportunities for people with inside information to monetize their inside information.”
Well, if Hayes is a saint, then I’m a sinner, because let me tell you this: If I were able to find myself in possession of inside information — or, in this case, information — I would like to think I would bet every dime I have, along with every penny, nickel, and quarter, on the outcome. Sure things don’t come around too often in this life, and presented with one, I would like to think I’d go all-in.
Truth is, I’d probably put $500 or so on it because I am both weak and a low roller, but still, it doesn’t change the calculus: Presented with an angel and devil on my shoulder when it comes to prediction markets, I will listen to the devil and tell the angel, in no uncertain terms, to bugger off.
Let’s dig in
And here’s the thing: I’m not special. I’m not weak-willed. I’m just honest about it.
Research backs me up. Nina Mazar, along with On Amir and Dan Ariely, published a paper called The Dishonesty of Honest People in 2008. Their finding? People cheat. Not a lot, but consistently. Enough to pocket some extra cash while still being able to look in the mirror and see a good person looking back.
In one study, participants could overreport how many math problems they’d solved. Given the chance, they inflated by about 13% of what was possible. Not zero. Not the max. Just enough.
But here’s the rub, especially for prediction markets and really, all forms of online wagering: When the researchers had people earn tokens instead of cash, tokens they’d swap for money seconds later, cheating more than doubled. One step removed from actual dollars, and everybody’s inner grifter wakes up.
Think about Kalshi. You’re not handing cash to a bookie. You’re clicking buttons, watching numbers move. It’s clean. It’s techy. The money doesn’t feel like money.
Furthermore, even though betting on inside information would mean I’m taking money from another human … well, it doesn’t really matter, as long as I don’t know who that human is.
Gary Charness and Uri Gneezy, a couple of economists, ran a simple experiment and published their findings in an 2008 edition of the Journal of Economic Behavior & Organization They had people play a game where one person splits a pot of money and the other person just has to accept whatever they’re given. But in some rounds, they told the splitter the family name of the person on the other end.
That’s it. Just the last name. Not a photo, not a life story. Just: “The other person’s name is Schwartz” or whatever.
When people knew the name, they gave about 50% more.
Here’s where it gets interesting. Charness and Gneezy also ran a version where the other person could reject the offer, thus turning it into a strategic game, a game you’re trying to win. And in that version? Knowing the name made almost no difference. People played to win. As the researchers put it: “Strategic considerations crowd out impulses toward generosity.”
Once it’s a game, the nice-person stuff just evaporates.
Prediction markets are games. Anonymous games. You’re not screwing over Schwartz. You’re clicking buttons, beating the market, winning. The person on the other side of your trade isn’t a person. They’re … liquidity.
Chris Hayes is a name. He’s a face on TV, a voice on a podcast, a guy posting on Instagram. He has to look people in the eye. The rest of us? We’re faceless, placing bets on an app, playing a game we’re trying to win.
Now here’s where I’m supposed to say this is all terrible and somebody should fix it.
Except Robin Hanson, an economist at George Mason, has a different take. He’s been writing about prediction markets forever, and he thinks insider trading rules don’t make sense here. Those rules were built for stock markets, to protect regular investors from getting fleeced. But prediction markets aren’t about investing. They’re about getting to the truth.
You want insiders to play, Hanson argues, because when they do, the price gets smarter. The whole point is to get people who know stuff to put money on it.
And besides, as Hanson argues, only you get to decide which secrets to keep.
If it ain’t broke (?)
The Chris Hayes market wasn’t broken. It was working. People in that studio audience knew things. They could bet on it. The market would clearly move toward the truth.
So where does that leave Hayes, the saint, refusing to bet on himself while a million bucks sloshes around?
Maybe he’s right. Maybe there’s something gross about cashing in on what you did or didn’t say.
Or maybe he’s bringing a rulebook to a game that doesn’t use one.
At any rate, the devil on my shoulder isn’t whispering anymore. He’s just pointing at the screen. The angel has left the chat.
I should probably start going to television tapings is my main takeaway here.
