The entire finance industry does a fancy form of betting. Risk and reward are packaged up and bought and sold by people according to their preferred allotment, and once the outcome is known the spoils (or losses) are correspondingly shared.

Prediction markets can also be called "information markets" because they expand the class of things we can understand the risk characteristics of beyond the traditional instruments.

Lloyds of London began by issuing insurance on shipping vessels so that ship owners shared the risk of a wreck across all their fleets, reducing the devastation of losing a ship for each individual owner. Prediction markets let us do this for a host of other outcomes.

A prediction market is nothing more than a betting market, the innovation (attributable to Robin Hanson, I believe) is the idea of the binary future, which makes the probabilistic outcome work nicely with a traditional futures contract structure, and allows for a lot of nice intuitions based on price movement.

They provide a far more accurate insight into the likelihood of events than any other alternative. You can think of it as betting if you like, but it's betting with lots of positive externalities.

Consider a business that needs to reason about the probability of a law passing, severity of global warming, or any number of other difficult to predict events. Prediction markets would provide by far the most accurate insight.

It's wisdom of crowds + skin in the game.

>Consider a business that needs to reason about the probability of a law passing, severity of global warming, or any number of other difficult to predict events. Prediction markets would provide by far the most accurate insight.

Surely such prediction markets are a guarantee of corruption?

That or I just don't understand - if a market exists that takes bets on human actions such as passing laws, how long someone will live etc, the betters will have a very active interest in doing everything they can to make a large profit, of which corruption alone is the most obvious target.

I mean in a way you're right - it is wisdom of crowds, albeit crowds of people that "know" only because they're corrupted the final event.

They don't guarantee corruption, because if the corruption shows up on the prediction market it has occurred (or will occur) elsewhere already, and if large amounts of money are betting on a certain outcome, others will follow, and thus the payout won't be as large.

> Consider a business that needs to reason about the probability of a law passing, severity of global warming, or any number of other difficult to predict events

Like the probability of a smart contract being exploited via a loophole? ;)

This is OT now, but in addition to the corruptive effects of traders influencing results they are betting on mentioned by mootothemax, another possible negative externality is the conservative effect the prediction market would have on the traders. Eg. if you got a whole population to bet on the market, suddenly it is in the interest of the majority to uphold past predictions, thus making society resistant to change.

EDIT: Moved to the correct location in the discussion tree.

It's very frustrating for me to see "probability of a law passing" and "severity of global warming" being thrown in the same basket.

While the first lends itself naturally to be framed as a binary prediction market (and there are good reasons to believe that the market will produce reliable forecasts), "severity" doesn't lend itself to be framed in this way (in Taleb terms, it is a non-linear payoff from a complex domain, "4th quadrant"). Ignoring that there is a lot of ideology involved in defining the exact terms, in my mind it is just unreasonable to expect prediction markets (or any mechanism for that matter) to deliver reliable results in this case.

Even when not accepting the Taleb argument regarding our ability to predict, at least it should be acknowledged that it is not straight-forward to generalize from binary outcomes to open-ended ones.

Presumably he meant a specific prediction like "Greenwich will be permanently flooded by 2025" ?