Tuesday, June 15, 2004

It Is the Economy -- And a Few Other Things

Predicting presidential elections is fairly easy. To begin with, the incumbent or the nominee of the incumbent party wins most of the time -- 60 percent of the time from 1920 through 2000. (I've picked 1920 as the starting point for this analysis because it marks the birth of the modern, post-Teddy Roosevelt, more-or-less laissez-faire Republican Party.)

Why do incumbents or the incumbent party's nominee tend to win? Because the private economy grows most of the time. (My measure of growth in the private economy is the change in real GDP per capita, net of government spending.) Naive voters (that's most of them) tend blame or credit the current president with the state of the economy. Such blame or credit is seldom merited.

With three exceptions, which I'll come to, the incumbent or the incumbent party's nominee lost the election when the private economy was shrinking in election year or was about where it had been four years earlier. Such was the case in 1920 (Cox lost to Harding), 1932 (Hoover lost to FDR), 1952 (Stevenson lost to Eisenhower), 1960 (Nixon lost to JFK), 1976 (Ford lost to Carter), 1980 Carterr lost to Reagan), and 1992 (Bush I lost to Clinton). The more attractive personalities of Eisenhower (vs. Stevenson), JFK (vs. Nixon), and Reagan (vs. Carter) may also help to explain their victories. Clinton's personality may have helped him beat Bush I, but he was helped greatly by Perot, who probably siphoned far more votes from Bush I than he did from Clinton.

The only three election outcomes that violate the rule of growth are those of 1944, 1968, and 1980. The private economy was growing in 1944, but it had shrunk since 1940. However, it had shrunk because of the massive diversion of resources to a popular war. People understood that and stuck with Roosevelt, albeit by a smaller margin than in 1932, 1936, and 1940.

In 1968 and 2000, by contrast, incumbent vice presidents (Humphrey and Gore) failed to win election in their own right, despite strong economic growth.

The 1968 election was, as much as anything else, a referendum on the war in Vietnam and the cultural war in America. Humphrey was the target of the electorate's rage for the failure of the war in Vietnam and for the rise of the counter-culture. Nixon's victory over Humphrey would have been even larger had Wallace not captured a large part of the counter-counter-culture vote.

The 2000 election was Gore's to lose, and he lost it, but barely. Yes, the U.S. Supreme Court made the right decision, albeit for the wrong reason. The Florida Supreme Court tried, selectively, to give some voters a second chance after they had failed to cast proper ballots, in contravention of the rules adopted by the Florida legislature and therefore in contravention of the U.S. Constitution.

The strong desire of conservative voters to punish Gore for Clinton's sins was probably offset by the strong desire of liberal voters to vindicate Clinton by electing Gore. Gore may have been hurt by the contrast between his own rather scolding, school-marmish personality and Clinton's sunny personality, but Bush certainly wasn't (and isn't) an Eisenhower, JFK, or Reagan. Gore simply ran a bad campaign; he zigged left and zagged right, always transparently pandering to one interest group or another. He ultimately lost the election because many people on the far left suspected his bona fides and cast their votes for Nader. And there went Florida. Gore was the Titanic of presidential candidates -- seemingly robust but fatally vulnerable.

What will happen in 2004? Bush stands a good chance if the economy continues to rebound strongly until election day. But he will probably lose if the economy stutters or if voters see Iraq as a second Vietnam. A major terrorist attack in the United States could cut either way. It will be close.