Betting markets inside corporations

There is a new paper in the JPE on this, with encouraging results:

A Pari-Mutuel-Like Mechanism for Information Aggregation: A Field Test inside Intel
Benjamin Gillen, Charles Plott & Matthew Shum
Journal of Political Economy, August 2017, Pages 1075-1099

A new information aggregation mechanism (IAM), developed via laboratory experimental methods, is implemented inside Intel Corporation in a long-running field test. The IAM, incorporating features of pari-mutuel betting, is uniquely designed to collect and quantize as probability distributions dispersed, subjectively held information. IAM participants’ incentives support timely information revelation and the emergence of consensus beliefs over future outcomes. Empirical tests demonstrate the robustness of experimental results and the IAM’s practical usefulness in addressing real-world problems. The IAM’s predictive distributions forecasting sales are very accurate, especially for short horizons and direct sales channels, often proving more accurate than Intel’s internal forecast.

Hat tip goes to the excellent Kevin Lewis.  Here are earlier, ungated versions.


Hewlett-Packard did this as well to reveal, for example, whether a product release would occur on time. Good way for finding out information embedded within an organization. For example, an engineer's secretary and someone in shipping probably have information that can be captured on internal markets.

There are also bidding markets within the White House.

Staff in the White House are betting on how long Priebus stays on. What are the bets on how long Jeff Sessions remains in office.

Of course, there are the Trump impeachment betting markets:

Does anybody know the status of the peer-to-peer prediction market that Robin Hansen was associated with? From last year or the year before. If they can legalize 'gambling' de facto, for chess, via such an arrangement, it will be a boon.

The mission has 312 Mexican restaurants.

I really hate the betting part, as I feel it is something that is responsible for putting people in unwanted situations and that is really not good for anyone. I do Forex trading only and due to broker like OctaFX, I find it a lot easier with their complete setup from having over 70 instruments to pick from to mighty rebate program where I get 50% back on all trades even if it’s lost, so keeps me in really good shape.

Results do not look all that impressive, and the pari-mutuel mechanism leaves me with the impression of needless complexity. What was wrong with a standard prediction market?

My initial reaction is what about moral hazard, e.g the point-shaving scandals in college basketball? Someone places a bet on a negative outcome (or against a favorite) and then subtly works or simply shirks so that the negative outcome is more likely to happen (or the favorite loses).

I would imagine the idea is to give people an incentive large enough for information sharing but too small to motivate the harder work of gaming the outcome.

The IAM concept feels very bayesian to me, so I will add this paper to my summer reading list.

Comments for this post are closed