The "Bread and Peace" Economic Forecasting Model for predicting Presidential elections utilizes two factors:
1) "The Bread" - i. e., the per capita real disposable personal income over the incumbent president’s term, and
2) "The Peace" - i. e., the cumulative U.S. military fatalities in overseas conflicts.
Well, just how good a predictor is the model? And how well has it correctly predicted the outcome of past Presidential elections?
The answer to this pithy question is provided by the above chart, showing the percent share of the 2-party Presidential vote gained by the incumbent party for past Presidential elections. Of course when this percentage falls below 50% it means that the incumbent will usually lose, since the opponent's share in this case will then be greater than 50%.
The result of applying the model to past Presidential elections dating back to 1952 is shown in the above figure. The model computes incumbent shares of the 2-party vote of less than 50%, correctly predicting incumbent party losers, for the election of 1980, 1952, 2008, 1960, 1968, 1976, and 2000, which were all elections in which the incumbent party actually did lose.
Contrariwise, the model computes incumbent shares of the 2-party vote of greater than 50% for the elections of 1996, 2004, 1988, 1956, 1984, 1972, and 1964, which were in fact elections in which the incumbent party did actually win. The names of the candidates in each of the above elections are shown in the chart, for referece.
Judging from these results, and noting that while the elections of 1960 and 2000 were both effectively a tie split at 50% share for each candidate, that in both these elections the incumbent party still lost. This suggests that an incumbent needs more than the bare 50% share of the 2 paarty vote in order to win.
So based on the above results, the "Bread and Peace" Economic Forecasting Model has correctly calculated winning percentage for elections in which the incumbent party did win, and correctly calculated losing percentages for elections in which the incumbent party did lose This would indicate a very good track record, and would make the predictions of the model regarding the current Presidential race something that should interest us. So what does the model say about the current race?
Applying the model to the current presidential race, the Bread and Peace model predicts that Romney will win, 52.5% to Obama’s 47.5%, which is close to a landslide for Romney. Note that this is in disagreement with recent polling results.
The model is a leading indicator, however, and we shall see in the next 97 days whether the polls catch up to the model, or vice versa.
Referenced URLs:
1) http://www.douglas-hibbs.com/HibbsArticles/HIBBS_OBAMA-REELECT-31July2012.pdf, and
2) http://www.aei-ideas.org/2012/08/economic-forecasting-model-predicts-obama-will-lose-in-near-landslide/
No comments:
Post a Comment