The Economy & You #57: Can an Economic Model Pick a President?
Today is Election Day where millions of people stand in line to select the leader of our nation. Throughout the campaign season we have seen numerous polls and even more political ads. Still, from an economic perspective, is there a model or equation that can predict who will win the presidential election? Actually there are a few. One of the best known was created by Professor Ray Fair from Yale University.
According to Professor Fair’s model, growth and inflation are two important factors that influence voting behavior. In terms of growth, Fair measures how much the economy has grown on a per-person basis in the first 9 months of an election year, and how many quarters has the per-person growth rate exceeded 3.2% over the last 15 quarters leading up to the election. Fair does not use data like the unemployment rate or job market data as they are not as predictive.
In terms of inflation, Fair postulated that Americans prefer that prices remain stable in that prices neither rise nor fall too quickly. Therefore, he the average rate of price change for the 15 quarters leading up to the election.
Included in the Fair model are political factors such as incumbency and the number of terms the incumbent party has held the White House. A sitting President has an advantage when running for re-election. Still, the longer one party has held the presidency, the lower the chances of continuing in office.
The important question is how accurate is the Fair Model? In 24 elections from 1916 to 2008, his model predicted the winner 21 times and was incorrect three times. The three exceptions were Kennedy over Nixon in 1960, Clinton over Bush in 1992, and Bush over Gore in 2000. It can be argued that there were circumstances that contributed to the wrong prediction. In 1960, the Kennedy victory was extremely small. In 1992, Ross Perot put up a strong third party candidacy. In 2000, Gore won the popular vote will Bush prevailed with the electoral college.
The interesting part of this model is that the predictions do not take into account the candidates themselves. Regardless of the personalities of the candidates, the strength or weakness in the economy appear to be a more accurate way of predicting presidential elections. One could argue that political consultant James Carville was right. It is “it’s the economy stupid!”
So . . . who will win? According to the Fair model, it is too close to call. The model results indicate that Mitt Romney will win with 51% of the vote to President Obama’s 49%. Still, the model has a standard error of 2.5%, so we will have to wait and see.
Professor Fair also has an economic model that predicts Congressional House elections. More information about Professor Fair’s Vote Prediction Models can be found here:
- Economic Model Predicts Narrow Romney Victory (blogs.wsj.com)
- Obama or Romney? How Accurate Are Economic Models? (pbs.org)
- Election forecast model predicts Romney wins 51-49 (aei-ideas.org)
- Election Forecast Models Clouded by Economy’s Middling Growth – Bloomberg (bloomberg.com)
- This Respected Election Forecasting Model Says Romney Wins (businessinsider.com)
- Electing a President in a Microtargeted World (blogs.hbr.org)