IE 11 is not supported. For an optimal experience visit our site on another browser.

Economy and incumbency favor Bush

President Bush's re-election is a virtual certainty if you believe a model developed by a  professor who tries to predict election winners based almost purely on economic factors. By Martin Wolk.
/ Source:

Economic forecasters like nothing better than a clear-cut result, and that is exactly what Yale economist Ray Fair offers in his 2004 presidential vote equation – a near-certain prediction that President Bush will win re-election.

Fair is well-known in the academic community for a free Web site that offers a wealth of data about the U.S. and other economies, including a well-regarded model for predicting future growth and inflation. Among his many peripheral areas of interest, Fair has been working for 25 years to refine a model that attempts to predict the outcome of presidential elections based almost purely on economic factors.

Fair cautions that he is not a pundit, and his model does not have a perfect track record – its variables completely fail to account for Bill Clinton’s six-point popular vote victory over the first President Bush in 1992. Richard DeKaser, chief economist for National City Bank in Cleveland, figures that Fair’s model accurately identifies the winner in 18 of the past 22 presidential elections, or 82 percent.

But Fair is less interested in the winner than in his model’s ability to approximate the actual result of the popular vote. Over the past 10 elections, the model has accomplished that task within 2.6 percentage points on average, and in the past two elections the formula has been accurate to within about 1.2 percentage points.

Based on Fair’s relatively conservative economic forecast-- a 2.4 percent GDP growth rate -- Bush is likely to win 58 percent of the 2004 popular vote not including any third-party votes – a landslide over whoever wins the Democratic nomination.

Even if you plug your own variables into Fair’s model it is almost impossible to come up with an economic scenario that results in a Bush defeat. For example if the economy, as measured by real per capita income, fails to grow at all over the next year, and inflation shoots through the roof, Bush still would get more than 54 percent of the popular vote according to the equation. (Fair figures his standard error at 2.37 points, so any result of less than 53 percent is considered too close to call.)

Head start for incumbent
One reason for Bush’s strong standing is that he gets a substantial head start by virtue of the incumbency effect as calculated by Fair. All members of the incumbent party are presumed to derive some benefit of incumbency, but none more than a Republican whose party has been in office for only a single term. Based on Fair’s analysis of elections since 1916, Democrats always have a slight disadvantage, and voters tend to tire of the incumbent party the longer it has occupied the White House.

Importantly, Fair’s equation does not account for what economists call “exogenous” events, occurrences that by their very definition are impossible to model or predict. Fair’s model includes a variable only for the declared wars of World War I and World War II which were presumed to help the incumbent party. A model created by economist Douglas Hibbs hints at how an economist might try to calculate the impact of the current Iraq war on next year’s election outcome. Hibbs’ model attempts to account for the Korean and Vietnam wars by factoring in the number of U.S. soldiers killed in action, which are presumed to be a negative factor for  the party initiating the commitment of U.S. forces – the Democrats in both cases.

Fair stresses that his model is mainly a teaching tool meant to interest students in his specialty of econometrics, the application of mathematics and sophisticated computing to economics. He applies the same technique to marathon running times, college grades, the quality of French wine and extramarital affairs in his book “Predicting Presidential Elections and Other Things.”

In any case, the results he predicts are consistent with the view of many analysts who say the economic fundamentals are steadily lining up to support Bush’s re-election – probably just as his advisers had hoped when they devised the tax package that gave the economy a huge boost beginning in the middle of this year.

Retail sales, for example, jumped 0.9 percent in November, according to government figures reported this week. The results virtually assured that retailers will have their best holiday season in at least four years, analysts said.

The only truly troubling piece of the economy continues to be the labor situation. Only 300,000 jobs have been added to the economy over the past three months, although that is a welcome improvement from earlier in the year when the economy was shedding nearly 100,000 jobs a month.

DeKaser, of National City, said November’s employment figures were weakened by a grocery strike in California, but other figures indicate the job market is improving substantially. He pointed to an increase in average overtime hours, a lengthening workweek and a decline in new claims for unemployment benefits.

“It’s not going to be the economy that causes (Bush)  to lose the election,” said Mark Zandi, chief economist at, a forecasting firm. “The only thing that could derail his election is some imponderable event, which most likely would come from overseas.”

Still, Zandi said it was “premature” to break out his firm’s 2004 election model. Although that model failed miserably to predict the 2000 election result, it is unique among presidential election models in that it looks at economic conditions on a state-by-state basis and attempts to predict the outcome of the electoral college vote.

“And we all know now how important the electoral college is,” he said.