The recent ‘Arab Spring’ and the subsequent realignment of much of the political sphere of the Middle East took many traditional commentators by surprise. Matthew Partridge argues that, in light of this, the ‘prediction markets’ of political bookmakers may be able to provide a significant contribution to intelligence gathering by adding greater accuracy and objectivity to forecasting.
One of the most striking things about the changes in the Middle East has been that they took almost everyone by surprise. Intelligence agencies, academics, news organisations and commentators were all wrong-footed. While it is heartening that the desire for freedom in the Arab world has surpassed expectations, this does present a practical problem for those wanting to get an accurate appraisal of how things are going to turn out. While Gaddafi’s regime is thankfully in its death throes, its has taken over six months, rather than the couple of weeks most analysts predicted. Similarly, Bashir Assad is still in power, and overseeing a brutal crackdown.
In recent years an increasing number of academics have begun to argue that “prediction markets”, their term for betting exchanges and bookmakers, have the potential to use the “wisdom of crowds” to add a degree of objectivity and accuracy to forecasting that more traditional methods of information-gathering lack. Justin Wolfers, assistant professor of economics at the Wharton School in Pennsylvania and a Research Fellow with the National Bureau of Economic Research, contends that “Across a range of contexts; predicting economic indicators, elections, box office takes, markets have typically out-performed alternative forecasting approaches”.
Wolfers also points out that “For many of us, there aren’t useful alternatives in terms of quantitative predictions about the Middle East”. Professor Leighton Vaughan Williams, editor of the Journal of Prediction Markets, and author of the recently published, “Prediction Markets: Theory and Applications” goes even further. In particular he suggests that they “can help construct the intelligence jigsaw” and “potentially offer a significant and currently under-exploited contribution to intelligence gathering”.
This is not solely a hypothetical question. The betting exchange Intrade.com is currently running markets on the survival of a range of Arab dictators. While gambling on the lives and liberty of millions is certainly controversial, Carl Wolfenden, Intrade’s Exchange Operations Manager, defiantly points out that “Intrade has been listing markets on major international events for a number of years now”. He claims that “the response from our members has been positive” and vows that “If we see the possibility of a new market, and we think it will be of interest to our customers, then we will definitely launch new markets. This applies not only for the Middle East, but for the rest of the world”.
Unsurprisingly, this approach has its critics. With the rapid gyrations in the value of sovereign debt suggesting that the financial markets can be heavily influenced by the “animal spirits” of fear and greed, it is unlikely that the less liquid experimental markets are more robust. Technology pundit Chris Masse of Midas Oracle feels that “the usefulness of the prediction markets is small”. In his view, their key weakness is that “the US traders on Intrade do not have exclusive insights. They just read the Washington Post’s foreign affair pages and a bit of CNN International”.
The intelligence and security community also has mixed views. John F McCreary, the Chief Analysis Officer for KGS Security, who had a senior role at the United States Defense Intelligence Agency, notes that the National Intelligence Council successfully trialed prediction markets, until political pressure ended the experiment. However, he believes that, since they “violate the scientific principles of auditability, replicability and irrefutability”, they lack credibility. Retired Naval Intelligence Commander Jennifer Dyer sceptically points out that, “anyone who has bet on sports would be wary of the effects of a large, very specifically motivated group of bettors”. According to Professor Anthony Glees of the University of Buckingham, “there must be a few analysts in the FCO and at GCHQ looking at these sorts of websites but I doubt if they spend much time on them”.
Both McCreary and Dyer emphasise that intelligence is not just about making predictions. According to McCreary “the judgment about what leader will fall next, is entertaining but less important than how can we keep a valuable ally in power”. He claims that under his supervision and direction, a group of over a hundred analysts were able to answer the latter type of questions to a very high degree of accuracy. Dyer contends that “as a rule, since World War Two, the US has not sought to predict and prepare for outcomes, so much as to shape them”.
Surprisingly, the most negative comments come from those who regularly bet on more mainstream political events, such as the US Presidential Election. Mike Smithson, editor of PoliticalBetting.com, is “very negative about these markets”. Not only is he “not sure who is betting on them”, but Smithson is also concerned about possible disputes since, “a big problem can be in defining how the bet will be settled”. In 2006, Intrade, then named Tradesports, controversially ruled that the absence of an official confirmation from the US Department of Defense meant that, for betting purposes, the widely reported North Korean missile tests did not take place.
Thus far the balance of evidence seems to be in favour of those sceptical of “prediction markets”. In Libya, the odds of Gaddafi being removed by the end of this year fluctuated wildly, and followed, rather than led, events. Less than a fortnight ago, it was essentially even money on whether Libya would be liberated from his rule. Even Professor Vaughan Williams concedes that there are limits to their use, stating that “I don’t see them replacing all other intelligence community structures”. As he puts it, “Replace James Bond? Surely not”.
Please read our comments policy before posting.