This is a calculation of the result of all presidential elections since 1952 based on two factors: 1) economic growth and 2) military fatalities. It’s creator, political scientist Douglas Hibbs, calls this the Bread and Peace model of presidential elections. As you can see, these two factors do a pretty darn good job of explaining the outcome of presidential elections since the end of WW2.
Based on this model, if the election were held a few months ago, Obama would have lost in a landslide. It’s not surprising: the economy is still not doing great.
Hibbs posts his predictions for 2012 here: http://www.douglas-hibbs.com/Election2012/2012Election-MainPage.htm. In essence, he predicts that Obama will likely lose the 2012 election unless the average per capita annual income growth rate between now and the election is at least +2%.
The way that political scientists see the outcome of presidential elections is pretty simple: campaigns, candidates, advertising, fundraising, and debates all matter, but not nearly as much as the condition of the economy when the election occurs. The economy sets the “boundaries” of the outcome within which the campaign, candidates, debates, etc. can all make marginal to moderate differences. But the economy determines the “upper” and “lower” limits of the ultimate outcome. In other words, even the best incumbent candidate will likely lose with a poor economy.
According to this model, the best thing (but not the only thing) that Obama can do to ensure re-election is to do everything possible to boost individual income growth rates for American citizens between now and election day. Strategically, the jobs bill he is currently pushing is a modest step in the right direction. Given that the personal income growth rate in August was -0.1%, though, it suggests that Obama has his work cut out for him over the next 13 months…