A prediction market for toxic assets

  • Authors:
  • Alan Holland

  • Affiliations:
  • Cork Constraint Computation Centre, Department of Computer Science, University College Cork, Cork, Ireland

  • Venue:
  • AICS'09 Proceedings of the 20th Irish conference on Artificial intelligence and cognitive science
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose the development of a prediction market to provide a form of collective intelligence for forecasting prices for "toxic assets" to be transferred from Irish banks to the National Asset Management Agency. Such a market allows participants to assume a stake in a security whose value is tied to a future event. We propose that securities are created whose value hinges on the transfer amount paid for loans from the agency to a bank. In essence, bets are accepted on whether the price is higher or lower than a quoted figure. The prices of securities indicate expected transfer costs for toxic assets. Prediction markets offer a proven means of aggregating distributed knowledge pertaining to estimates of uncertain quantities and are robust to strategic manipulation. We propose that a prediction market runs in parallel to a pricing procedure for individual assets conducted by the government agency. We advocate an approach whereby prices are chosen as a convex combination of the agency's internal estimate and that of the prediction market. We argue that this will substantially reduce the cognitive burden for the government agency and improve the accuracy, speed and scalability of pricing. This approach also offers a means of empowering both property experts and non-experts in a cost-effective and transparent manner.