Why winning an auction often means overpaying

HostYou sometimes hear about these business school classes where the teacher puts a big glass jar full of pennies on the desk and tells the students they're going to auction it off. Everyone wants to win, they all start bidding, and eventually, one student walks away with the jar. But here is the thing. Almost every single time, that winner ends up with less money than they started with. They win the pennies, but they paid way more than the pennies are actually worth. Why does the person who wins the prize so often end up as the biggest loser in the room?

GuestThat's a classic trap called the winner's curse. It's a strange bit of math where winning a prize is actually the proof that you messed up. But we have to be careful, because it doesn't happen every time you bid on something. If you're at a garage sale and you see a vintage rug that you absolutely love for your living room, you can't really be cursed. That's what people call a private value auction. The rug is worth exactly what you're willing to pay for the joy of having it in your house. If you pay a hundred dollars and it makes you a hundred dollars worth of happy, you won.

HostSo the curse only shows up when we're not buying for the heart? When it's just about the cold, hard numbers?

GuestRight. It happens in what are called common value auctions. This is when an item has one real, set value that's hidden from everyone. Think of that jar of pennies, or the rights to drill for oil in a specific patch of ocean, or even the rights to show a sports season on TV. In those cases, the item has a real dollar value that will be the same for whoever wins it. But no one knows that number for sure while they're bidding. Everyone is just guessing. And because no one has all the facts, everyone's guess is going to be at least a little bit off.

HostBut if everyone is guessing, wouldn't some people be too high and some be too low? It seems like it would all even out in the end.

GuestThat's exactly the trap. Think about it like a big curve. If fifty people guess how many marbles are in a jar, most people will be pretty close to the true number. Their guesses will cluster in the middle. A few people will guess way too low, and a few people will guess way too high. But in an auction, the people who guessed low or even the people who guessed the right amount don't win. The winner is almost always the person at the very edge of the curve who made the biggest mistake. The more people you're competing against, the more likely it's that the winner is the one who made a one in a hundred error.

HostThat sounds like it would get better if more people were bidding, like you would get a more accurate price. But you're saying it actually gets worse?

GuestIt gets much worse. The more people you add to the room, the more chances there are for someone to make a massive overestimation. In a crowd of ten people, the highest guess might be a little bit over the real value. But in a crowd of a thousand people, the winner is likely the one person who's the most wildly wrong out of a thousand. The winning bid moves further and further away from the real value and deeper into the curse.

HostHas this actually happened in the real world, or is it just something that happens with pennies in a classroom?

GuestIt happens with billions of dollars on the line. Back in the nineteen seventies, some engineers at a big oil company called ARCO started looking at why their industry was struggling. Companies were spending huge amounts of money to win the rights to drill for oil in the Gulf of Mexico. They would win the bid, they would start drilling, and then they would realize they weren't making any money back. The engineers realized that the winner of an oil lease wasn't the company with the smartest scientists. It was just the company whose data was the most incorrectly positive. They were winning because their math was the most wrong.

HostThat has to change how you think about bidding. If winning is a sign that you probably overpaid, how do you ever come out ahead? It feels like you just shouldn't play the game at all.

GuestYou can still play, but you have to use a trick called bid shading. To be smart about it, you have to assume that if you win, it means every other person in the room thought the item was worth less than you did. You have to assume your own guess is too high. So, you purposefully lower your bid based on how many people are in the room. You aren't bidding what you think the item is worth. You're bidding what you think the person with the second best guess believes it's worth.

HostSo you're basically building in a safety net for your own excitement.

GuestYou have to. If you don't shade your bid, you're just waiting to be the outlier at the edge of the curve. The real winners are the ones who realize that being the highest bidder is usually just a very expensive way to be wrong.

HostThe next time someone wins that jar of pennies, I'll be thinking about those oil companies in the seventies and how that heavy jar is actually a very light trophy.