Statistically Measuring 2016 Presidential Candidate Electability: Evidence from Prediction Markets

AEI Economics Working Paper 2016-04

Posted: 10 May 2016

See all articles by Jonathan Hartley

Jonathan Hartley

University of Pennsylvania - The Wharton School

Kevin A. Hassett

American Enterprise Institute (AEI)

Date Written: February 3, 2016

Abstract

We use prediction market data from Betfair, the world’s largest Internet betting exchange, to measure the electability of 2016 Presidential candidates using regressions that compare the general election contest and party nomination win probabilities for each candidate. A candidate who is more electable should see a higher response of the odds of becoming president to a given change in the odds of receiving a party’s nomination. Our regressions estimate this response for each major candidate, and these estimates constitute our measures of electability. The data indicate that there is a high degree of variability in the electability of candidates. We present a number of different model estimates, in order to explore the sensitivity of results to specific assumptions. Across specifications, we find that Jeb Bush and Marco Rubio are the most electable, while Chris Christie and John Kasich also have high electability scores. We also find that Hillary Clinton has the highest electability score in the Democratic field and that Bernie Sanders’ electability is sensitive to specification changes. We interpret Mrs. Clinton’s very high electability scores as suggesting that markets are pricing in a significant probability that Republicans will nominate a candidate who has little chance in the general election.

Keywords: Forecasting and Prediction Methods, Economic Models of Political Processes, Information and Market Efficiency

JEL Classification: C53, D72, G14

Suggested Citation

Hartley, Jonathan and Hassett, Kevin A., Statistically Measuring 2016 Presidential Candidate Electability: Evidence from Prediction Markets (February 3, 2016). AEI Economics Working Paper 2016-04. Available at SSRN: https://ssrn.com/abstract=2777233

Jonathan Hartley (Contact Author)

University of Pennsylvania - The Wharton School ( email )

3641 Locust Walk
Philadelphia, PA 19104-6365
United States

Kevin A. Hassett

American Enterprise Institute (AEI) ( email )

1150 17th Street, N.W.
Washington, DC 20036
United States
202.862.7157 (Phone)
202.862.7177 (Fax)

Register to save articles to
your library

Register

Paper statistics

Abstract Views
207
PlumX Metrics