Introducing information markets

Prediction Markets - a safe bet for forecastersThis article explains how information markets can assist policy developers by estimating predictions and making decisions. Economists have long been convinced by the power of markets to allocate resources efficiently. Markets are, so the theory goes, able to condense all available information about a particular product into one measure: the price. But what if the commodity you're trading is itself information? Information markets are designed to put the power of trade to work on information: predictions, decisions and estimates more generally.

The markets operate like a betting shop. Gamblers bid on the expected outcomes of an event - the 2008, 4th Quarter Employment rate for example - and the bookmaker pays out to those who judged the value successfully. Justin Wolfers' Political Prediction Markets: a Webcast for the Uninitiated provides an inspiring introduction to the topic. You can find a link to the webcast over at the Freakonomics blog.  To summarise, information markets can:

  • summarise all available information
  • make accurate predictions
  • respond rapidly to news
  • provide an incentive for information discovery
  • encourage truth revelation

Information markets are even robust to manipulation attempts because traders can profit from betting against the lies espoused by manipulators.

Potential application of information markets for government

Information markets could be used to provide answers to the questions of policy development. Information markets can provide predictions for policy research and make decisions to guide strategy and implementation.

Prediction Markets as a Research Intelligence tool

Forecasting is an obvious application. Information markets can summarise the implications of an evidence base in a meaningful way: indeed the only limit to the practical applicability of the results is the designer's imagination. With a large enough body of traders, the market can provide a continuously updated and accurate reflection of all available intelligence. A dynamically adjusted indication of labour market indicators for example. With the appropriate incentives, the market will induce traders to gather new evidence on the designer's behalf. Experts may be sufficiently motivated by the credibility that their company can earn from accurate prediction. Kudos points translate in contract awards. Ask five economists the same question, you will get six different answers, but if you ask them to bet on their answers you can only get one response. In fact, as Justin Wolfers explains above, the results from information markets are more accurate than expert surveys.

Decision Markets as a Policy Choice tool

What if we apply decision markets to the questions of policy development? In 2000, Robin Hanson wrote a manifesto describing a new form of government. In "futarchy," we would vote on values, but bet on beliefs. Elected representatives would formally define and manage an after-the-fact measurement of national welfare, while market speculators would say which policies they expect to raise national welfare. The basic rule of government would be: When a betting market clearly estimates that a proposed policy would increase expected national welfare, that proposal becomes law. Futarky is a strong governance model because of two attributes. Follow the links to understand why these attributes are so desirable:

  • Pursuit of welfare that includes both productivity and happiness - Nowadays the first is a given. If you aren't convinced that well-being is important, then you obviously haven't heard the evidence from Richard Layard's study into this "new science ". As he points out: So here we are as a society: no happier than fifty years ago. Yet every group in society is richer, and most are healthier... Through science, absolute material scarcity has been conquered in the West, and we need to think hard about what would now constitute progress.
  • Checks and balances - Watch the video below to see Paul Collier explain why this is such an important aspect of modern social democracy. Source: Paul Collier - 4 ways to improve the lives of the "bottom billion", TED Talks.

A health warning

While it is tempting to launch straight into information markets, the pragmatic policy-maker ought to first consider some of the caveats. Like any other market, information markets are subject to market failures. They will fail if there are too few participants - like a traditional commodity the market fails in the presence of monopoly power. Information markets must be designed carefully (particularly where a market-maker is used) so that the rules of trade do not bestow an unfair disadvantage that distorts the incentives for trade. Just as information markets can be distorted by robotic market-makers, human experts may have unfair access to relevant knowledge. Information asymmetries make it difficult for traders to judge the qualities of the different options facing them and may, in extreme cases, cause the markets to collapse completely. This is a critical issue when choosing who will participate and how they are engaged. Furthermore, information markets fail if nobody has any useful information! They can be used to illicit information from people who wouldn't ordinarily share their models or data.