James Cambell Visit: 2/25/11 Meeting

The discussion group was fortunate to have James Campbell, coauthor of The American Campaign and one of the nation's most experienced presidential election forecasters, stop by and share some insights on the 2012 cycle. Dr. Campbell was visiting UGA to give a talk on his current research, "The Economic Records of the Presidents: Party Differences and Inherited Economic Conditions” which reanalyzes (and in his words, debunks) Larry Bartels' highly acclaimed book, Unequal Democracy: The Political Economy of the New Gilded Age.

When asked to predict whether any of the candidates widely considered to be likely contenders for the Republican nomination (Romney, Huckabee, Barber, Gingrich, Palin, Bachmann, Paul, Huntsman, and now Romer) Campbell predicted an Obama victory but cautioned that changes in the electoral environment, particularly in Obama's approval ratings, are likely between now and 2012. Campbell said that every President who had a 45% approval rating or higher in the final months before the election has won reelection-so that approval rating is something to watch closely as the campaign season progresses.

Campbell also spent some of the meeting discussing the performance of forecasting models in the 2008 election. On average, the forecasting models for '08 were fairly accurate, especially considering that the forecasts are made months before the election (how many months varies with each forecaster) and do not account for campaign effects such as debates performances, gaffes, and of course in the case of 2008, an economic meltdown the likes of which had not been seen in this country since the Great Depression. For those of you unfamiliar with presidential forecasting models, it's important to point out that because of a very small N size (for 2012 models it will be N=17) forecasting models are constructed to be as parsimonious (simple) as possible. Thus, campaign events like those described above, as well as factors such as fundraising are not taken into account in the predictions.

Campbell also discussed why his model incorrectly predicted a 4% margin of victory for McCain. He argued that despite gaffes such as McCain’s ill-fated trip to Washington and temporary campaign suspension, the true reason McCain’s numbers dropped so precipitously in the last two months of the election was because his approval was tied to President Bush’s, whose own approval ratings plummeted from the low 40s to the mid-20s as he lost support from his base because of the economic crises. Thus, despite being an open seat contest focused on the Iraq War initially, 2008 became a classic retrospective election where the party in power was held responsible for poor economic performance.