In my view, presidential elections are usually pretty easy to predict. You just look at how the economy is doing. When we’re at war or in a major foreign policy crisis, you factor that in. But an incumbent president has to be judged very harshly over his handling of the war or foreign crisis before he will lose in a good economy.
It turns out that there’s a predictive model that formulizes these intuitions. It’s called the Bread and Peace Model. Douglas Hibbs, a retired economics and politics professor, developed it. According to Hibbs, the model nicely explains nearly every post-World War II presidential elections. Only 1996 (when Clinton overperformed) and 2000 (when Gore underperformed) are problematic.
For the Peace side, Hibbs basically counts American fatalities, subject to a proviso discussed below. For the Bread side, he relies on growth of per capita real disposable personal income.
Under Hibbs’ model, President Obama will likely receive only 47.2 to 47.8 of the popular vote. In other words, Romney’s margin will approach the margin that Obama won by four years ago.
In my view, the “Peace” component of Hibbs’ model is working against Obama more strongly than it should. Hibbs stiplulates that he does not count deaths caused in a president’s first term by a war initiated in a previous administration. Thus, Hibbs doesn’t count American fatalities in Iraq against Obama. He does, however, count those in Afghanistan, since Obama implemented a new stepped-up campaign there.
I doubt, though, that Americans will hold Afghanistan fatalities against Obama. For one thing, he’s pulling out. For another, Mitt Romeny is slightly more hawkish on Afghanistan than the president.
I wouldn’t be surprised if the Bread side of Hibbs’ formula standing alone, correctly predicts this year’s race. If so, the election will be closer than Hibbs’ model predicts, but with Romney still winning the popular vote.
Via James Pethokoukis.