Wednesday, October 15, 2008

The (Surprisingly) Long History of Political Stock Markets

We have witnessed unprecedented betting volume on the Intrade political markets. Perhaps more interesting has been the sharp price swings. Some have attributed these changes to a “rogue trader” who is seeking to slant prices away from their fundamentals. This charge could prove a serious concern in the ongoing debate to legalize these markets in the United States.

What is often forgotten in these discussions is that betting on elections is hardly a new concept. Election betting has occurred in the US since at least the 19th Century, and even earlier in England and Italy. In the US betting became centralized in organized markets outside of Wall Street in the early 1900s, and newspapers reported the current prices and even individual wagers on their front pages (see images above).

My co-author Paul Rhode and I have documented the existence of these markets and have analyzed the time series of prices and bet totals.

These markets provide several lessons for the present experience. First, the historical markets had a very strong forecasting record. In the fifteen presidential elections between 1884 and 1940, the underdog in mid-October won only once (in the closely fought 1916 election). This suggests giving credence to the current markets, despite the occasional price gyrations.

Second, volatile prices were common in the months leading up to the elections in the historical markets. At the same time, they tended to rapidly move in favor of the favorite in the last few weeks before the election (likely reflecting the diminished chance of an October surprise which would change the leadership in the contest). So the recent strong shift towards Obama in the political markets is nothing new.

Third, they show that these markets do more than simply mimic polls. This must be true since scientific polls simply did not exist during the existence of the historical markets.

And finally, they help support the case for legalizing these markets. It is sometimes argued that political futures markets might be subject to manipulation by political partisans which in turn might influence election outcomes. Although large sums of money were at stake in the historical presidential betting markets, I are not aware of any evidence that the political process was seriously corrupted by the presence of a wagering market. Many current concerns about the appropriateness of political prediction markets are not well founded in the historical record.

Further reading:

Rhode and Strumpf (2004). "Historical Presidential Betting Markets." Journal of Economic Perspectives.

Rhode and Strumpf (2008). "Historical Political Futures Markets: An International Perspective." working paper.

Slate, "When Wall Street Bet on Elections." 26 September 2008.