« Why are gun owners so happy? | Main | Ain't My America »
Non-profit prediction markets
At Bet2give, an online electronic prediction market, anyone can “bet” real money on the outcome of these events but with a twist – the “winnings” go to a charity of the person’s choice.
Here is the full article, and here is Bet2give.com. One of the "problems" with prediction markets is that they are zero-sum investments and traders cannot on average hope to come out ahead. So why trade? If only people really "in the know" trade liquidity will be low. If too many "betting for fun" fools trade, the prices don't mean that much. You might get the right mix of informed and uninformed but who knows? So the idea of doing these in charitable form might make some sense.
Posted by Tyler Cowen on April 19, 2008 at 03:29 PM in Economics | Permalink
Comments
"So why trade?"
For hedging purposes.
Posted by: ryan at Apr 20, 2008 12:19:31 AM
Isn't it better to just give money to charity? In this way, you know they get it for sure - and here if you lose, money goes into the pockets of the owners...
Apart from that, you have all kinds quasi-agency problems - e.g. how do you know that the gambling money actually goes to charity? Can you believe a gambling company?
********
Remy Piwowarski
Exec Director
Economics International Open Forum
http://economicsinternational.blogspot.com/
House of Young Econ Wolves
Posted by: Remy Piwowarski at Apr 20, 2008 5:15:03 AM
Prediction Markets can be directly subsidized with a market maker, allowing all traders who provide info to improve the price to expect to profit. Also, the more fools the more informed traders should be attracted to profit from them, so the mix is endogenous.
Posted by: Robin Hanson at Apr 20, 2008 6:06:17 AM
A question I have.... what happens if prediction markets are actually taken seriously by policy makers? If serious decisions are made based upon prediction markets, it could be possible that the cost of manipulating the prediction market is less than the benefit some entity might receive through policy decisions based upon erroneous prediction data.
Maybe a little farfetched, but imagine the fed lowers interest rates based partially upon recession expectations on Intrade. In Intrade's current state, it would clearly be possible to bid up recession numbers for under $10k in losses while in financial markets, one could profit by orders of magnitude more than $10k from an interest rate move.
Does potential for manipulation mean that that the Fed, along with other policy makers, should ignore (in a credible fashion) prediction markets given their current state?
Posted by: Matt Gunn at Apr 20, 2008 6:30:00 AM
From what I have read, market manipulation is both extremely expensive, and ineffective over even the short term. Once the market figures out that there is an irrational buyer, the market profits heavily. Apparently, attempts at market manipulation can even boost overall accuracy.
Posted by: David Shor at Apr 20, 2008 7:16:32 AM
"One of the "problems" with prediction markets is that they are zero-sum investments and traders cannot on average hope to come out ahead. So why trade?"
Why do people trade options?
Posted by: Anthony at Apr 20, 2008 7:43:05 AM
So why trade? If only people really "in the know" trade liquidity will be low.
By this logic, no one would pick stocks to buy and sell for profit. Sometimes people think they have knowledge when they don't.
Posted by: farg at Apr 20, 2008 10:11:21 AM
Prediction Markets can be directly subsidized with a market maker
So what is in it for the market maker?
Posted by: ad at Apr 20, 2008 1:40:58 PM
The Fed will never look to Intrade to gain an insight on whether or not America is heading towards recession. Very far-fetched. The Fed shits on Intrade.
Posted by: mpkomara at Apr 20, 2008 3:25:39 PM
It seems likely to me that people will be more careless with their betting in charity gambling, since theoretically no matter who wins the money is going to a worthy cause.
Farq makes a point, but I the stock market has a lot of buy and hold investors. If the market were zero sum there would probably be a lot less money in it.
Posted by: George Weinberg at Apr 20, 2008 5:22:17 PM
Other than entertainment, the value of a prediction market is that it makes predictions. If people actually made decisions based upon the predictions, I just wonder if the predictions would become less accurate because additional incentives are being introduced beyond trying to maximize profit within the prediction market. (eg. a strategy might be lose $X trying to manipulate an illiquid prediction market but make $X + $Y taking advantage of decisions made using the erroneous prediction.)
Far fetched to be sure, but I don't think its implausible.
The moment you attach consequences to some measurement, it creates incentives to game the measurement.
Posted by: Matt Gunn at Apr 21, 2008 5:02:40 AM
"Prediction Markets can be directly subsidized with a market maker"
"So what is in it for the market maker?"
-Information
Posted by: Mason at Apr 21, 2008 9:39:18 AM
I'm convinced that prediction markets are only going to get better and better. See here for my rationale:
http://blog.kevinmarkham.com/2008/04/how-to-predict-future.html
Posted by: Kevin Markham at Apr 22, 2008 12:00:18 AM
Did the futures and options markets not arise for main street folks to hedge against volatile real-world conditions like how much wheat was going to come in during the harvest? Only later did these evolve to entertainment and gambling. As with stocks there is some intrinsic value other than gambling. Everyone can make money in stocks if they stopped trading. And some trading occurs because it's time to send the kids to college or whatever. With gambling it is win or lose.
Things that are broadly predicted tend to become either self-fulfilling prophesies or become less likely the more the prediction is believed. This makes long-range predictions on events that can be attenuated even dicier.
Maybe additional compensation should be given to those actually making bets, like a slice of the information that Mason refers to. The market maker knows more than just the price and volume. But, much of this information likely has little value to the market maker and might have great value to traders. Can't give them free room service like casinos give the high rollers.
The intrinsic value of the prediction market is the accuracy of the predictions, right? Who benefits from this and has incentive to compensate predictors? Maybe individual companies should sponsor prediction markets about facts related to their business. Then they'd figure out what they need to do to incent the predictors. Then again, they'd be able to influence the outcome. So, negative predictions might get a premium. Stock prices are predictors, but they aggregate too much information to be useful as specific short-term predictors.
Posted by: Andrew at Apr 22, 2008 5:40:01 AM
Maybe taxes on prediction market gains should be reduced and losses should get tax credits.
Another value of the prediction market is the aggregating of the information. This work is done by the predictors and the market. There are people who benefit from the transparency of this information, which justifies tax incentives.
Posted by: Andrew at Apr 22, 2008 6:14:22 AM
Mason, the information is available to those who do not make the market as well. And they do not have to pay a penny. So why would anyone step forward to make the market?
Perhaps they would if they were the only people who found the information valuble. But in that case the value to society of the information gained is unlikely to be great.
The market maker has to have a way to make money out of this, just like other information aggregators (the Times, CBS evening news etc).
Posted by: anon at Apr 22, 2008 7:44:33 AM
