Prediction markets (again)

Prediction markets are currently getting more and more attention, since they allow investors to hedge risks they couldn't do before.

This example is from mySA.com

HedgeStreet.com, an online futures exchange, has opened up a hurricane predictions market where people can wager on the financial damages caused by a specific hurricane — or by the whole hurricane season.

The California-based exchange, regulated by the Commodity Futures Trading Commission, launched the hurricane market two weeks ago. While it may seem like a morbid novelty aimed at gamblers or weather junkies, company officers tout the new market as much more.

They claim it offers hurricane-weary residents a way to hedge their risk, and could develop into a powerful tool for predicting storm damage.

...

The exchange sells small contracts in an effort to attract individual as well as institutional investors. Contracts cost something less than $100, depending on what the market bears. If you're right, they pay out $100. If you're wrong, you get nothing.

For example, if you think the damage from this year's hurricane season, as calculated by the Insurance Services Office, will top $10 billion, you can, as of close of business Friday, buy a contract on HedgeStreet for $32.80. If damage tops $10 billion, you get $100 back. If it doesn't, you get nothing.

but remember two things that are important in determining the efficiency of markets: they work best with lots of liquidity and neglible transaction costs:

Borghesi, however, isn't sold on the ability of the new market to allow people to hedge against hurricane risk.

"If you own a house and a hurricane is heading towards you, you would have to bet a lot of money to hedge your risk," he said.

...

Although futures markets can be good predictors, there are no guarantees, Strumpf said. A good way to judge such markets is by how close the asking price and the bidding price are. Within a dollar or two is best, Strumpf said.

By that measure, HedgeStreet has a ways to go with its hurricane market.

People haven't exactly flocked to it yet. Trades over the first couple weeks are measured in the dozens, and there's commonly a $5 or $10 difference, or even more, between a bidding price and an asking price.

finally mickslam has a good comment about prediction markets on marginal revolution

It is possible that this market doesn't have enough volatility on a daily basis to attract the ecosystem of traders required for a vibrant market. Or enough data points to attract traders. I think thats one of the problems with Robert Shillers take on things, that for many of the markets that might be good (ok very, very good) for the economy, it would be difficult, if not impossible to create the necessary critical mass for these markets that would make them viable tools for risk managers.

I am very encouraged that we are seeing these markets. I am skeptical that these markets can be fully used. For example, Jason Ruspini and I have talked about the problems inherent in GDP contracts for example, and his favorite idea, tax futures, has the problem that implementing them would very likely be prohibitied, for very good reasons.

It will be a long winding road. The confluence of quantitative risk management, widespread recognition that markets really do work and the ability to dream is creating opportunities that didn't exist just a few years ago.

Something else I am starting to think on again is the Nassim Taleb critique - "we don't know what the real risks are". The odds are pretty high that for some events and styles of markets, we will never know enough for people to feel informed enough to trade them. Trading these would not be trading, but gambling. Any cogent person looking at these markets will demand a wide spread to trade them - which will work against the the very point of having a market in the first place.