Monday, June 23, 2008

The Failures of Markets

Free markets are the best available system of allocating resources and we're all blessed to live in one and blah blah blah. I like the free market. I certainly like it a lot better than all of the alternatives. But I am driven to distraction by free market androids who insist that markets are perfect and that any result a free market produces is, by definition, a correct result.

You know the type. This is the guy who insists that the Worldcom CEO was absolutely worth $120 million a year--because if he wasn't then nobody would have been willing to pay him at that salary!

Instead, I would content that while free markets are great and all that (see above), they produce failures at a much higher rate than most Americans would probably suspect. These matters normally hinge on something subjective, but today we have an example that seems pretty cut and dry, courtesy of Santino.

Betting markets are a pretty pure distillation of a free market. But Sonny notices that at least one casino in Vegas was (earlier this year) giving Tiger Woods 4-1 odds on winning exactly three majors--but 3-1 odds on winning the Grand Slam. That's right: The betting public thought that there was a better chance of Tiger winning all four majors (something which has never been done) than there was of him winning three of the four.

I'm not sure there's a way to dress up this result as anything other than a market reaching an objectively irrational result.


Anonymous said...

If betting is a business and you're trying to get people to bet on a really long shot that will probably not lead to a pay-out than you might be interested in manipulating the ods to make that long shot at least somewhat attracting for those who don't usually bet, but might be somewhat knowledgable to try.

Chris Mayhew said...

Stephen Colbert has a great line about "Global warning is real because Al Gore's movie made money. The market has decided". I've always thought that was a hilarious parody of exactly the sort of thing you're talking about.

Anonymous said...

This is a fascinating example, and I'm always glad to hear people thinking through examples of market failure, but I'm not sure that it shows quite that strong a result, i.e. an objectively irrational outcome.

If the payoff for winning all four majors were less than the payoff for at least three majors, you'd have no rational reason to bet for the four major sweep. But for the "exactly three majors" outcome, there are certainly states of affairs in which winning all four events is more probably than winning exactly three. Supposing independence of the events, winning all four is more probable than winning exactly three when the probability of winning any one is greater than .8. Not knowing much about golf, maybe it's crazy to expect that Woods should be that heavily favored to win, but it's not conceptually incoherent like P(winning at least three) > P(winning all four) would be.


Anonymous said...

Tiger's winning percentage as a pro is hovering around 29%...with over 200 events that's a large enough sample size to begin to draw conclusions.

The concept of a perfectly efficient market is nonsense...but markets are by far the most efficient method of allocating scarce resources.

Anonymous said...

Tiger winning all four majors (something which has never been done)

Actually, Bobby Jones won all 4 majors in the same year (although he did it before the Masters, so there were a different Big 4). And as I'm sure you know, Tiger held all 4 championships at one time (although not all in the same calendar year).

I'd guess that there are two reasons why the lines are different:

1) When people place those type of prop bets, they're usually doing it on a lark. They don't actually expect to win, and in fact the point is not to win. It's like betting on the Eagles to win the Super Bowl. You don't actually think it's going to happen, but you do it because they're your favorite team. What you're buying is a rooting interest. And there's very little sense in rooting for Tiger in 3 majors, and then rooting against him in the 4th.

2) There's also the problem of incomplete information, which frequently leads to apparently irrational results. I would imagine that most casinos have odds on Tiger winning the Grand Slam. I would not expect all casinos to have odds on Tiger winning exactly 3 majors. So if I wanted to bet on Tiger, most people would probably assume that the prop bet to make is the Grand Slam.

3) Betting markets are inherently irrational. If they were rational, they'd be empty. It's not rational to place bets at all because the house always wins money.

-- MattM

Anonymous said...

Markets? Odds are manipulated by bookees like central planning committees. But they are much smarter. I don't know why, but there could be some very good, mathematical reasons for a grand slam having better odds - especially if you just paid out big on him winning 3 and think it would be inevitable simply due to tiger-momentum at that point.
Sounds to me like you just wanna pick on someone and the market forces got in your way. Let 'em rest - they've never been free.
There are always bookees.