Sunday, March 28, 2004

Predicting presidential elections 

I had my eye on Yale economist Ray Fair's new book Predicting Presidential Elections and Other Things for my next Amazon splurge, until I stumbled on this post by Daniel Munz. Munz, a student enrolled in Fair's class at Yale, describes Fair's lecture on his apparently famous "Presidential Vote Equation." I had to see for myself what this equation was. Here we go. Are you ready?
The equation to predict the 2004 election is
VOTE = 55.57 + .691*GROWTH - .775*INFLATION + .837*GOODNEWS.
So you can see for yourself how simplistic this equation is, let me explain the terms involved:

VOTE is the proportion of the two-party popular vote going to Bush;

GROWTH is the growth rate in real per capita GDP for the first three quarters of 2004;

INFLATION is the average annual inflation rate during Bush's term;

GOODNEWS is the number of quarters during Bush's term where the real per capita growth in GDP exceeded 3.2%.

Okay. We start with a base term of 55.57% of the vote for Bush. The GOODNEWS term is always positive, adding even more votes for Bush. Go try it yourself. The site lets you plug in whatever numbers you want for growth, inflation, etc. The only way Bush can ever fall below 50% in this model is if inflation reaches double-digit levels or there's a huge (10%) drop in GDP.
Back to the Odd Hours main page
© 2004 Odd Hours
Reproduction permitted provided Odd Hours or the author of the quoted post is credited.