Legal Commentary

Fair's model focuses on a small set of issues: whether an incumbent is running, how long the incumbent party has been in office, the rate of inflation and the growth of the economy. His model figures in the presence or absence of war but defines it so it rarely matters, in fact it makes a difference (as a dummy variable) only for the elections 1920, 1944, and 1948. (To me, at least, this is a puzzling feature of the model). It's worth noting that in 2004 Fair predicted a significant victory for George W. Bush, on the order of 57 percent of the vote. Although Bush won, he only received 51 percent. Fair's model did not account for the nature of Bush's campaign, which tried to appeal largely to a conservative base, or the importance of the Iraq war.

As John Cassidy writes, there are several other econometric models, and they offer differing results. Most of them agree, however, that if the country experiences an economic slowdown during the election year, the incumbent party (that is, the Republicans) will suffer for it. However, a terror attack or a national security crisis might turn things around for the Republicans, although it might be a double edged sword, depending on how the current President handles the crisis. It might suggest that only the Republicans can be trusted to keep the country safe (which is likely to be a theme of the 2008 campaign), or on the contrary, that the Republicans have proved themselves incompetent in foreign affairs and so it is time to give the other party a chance.

Full post as published by Balkinization on February 13, 2008 (boomark / email).