By Christopher Farrell
It's Kerry vs. Bush in the contest for the Oval Office. The election is eight months away, and the campaign is universally expected to be long, nasty, and brutish. The electorate is sharply divided, as well as deeply engaged, if the unusually high participation rate in the Democratic primaries is any indication. George W. Bush and John Kerry are essentially tied in most political polls, with a slight edge to Kerry.
Most economists, however, distrust political polls, which have become a campaign-season staple ever since George Gallup demonstrated the superiority of scientific polling as a forecasting method over the straw polls that were common in the 1930s. (The track record of organized wagers on Presidential elections was remarkably accurate in the last half of the 19th and early 20th century, but by the late 1930s these betting pools had fallen into disfavor.)
CURTAINS FOR KERRY?
When it comes to predicting political outcomes for most economists, the intellectual mentor seems to be Karl Marx and his perspective that "the mode of production determines the social, political, and spiritual life processes in general." Or, as President's Clinton's campaign manager James Carville famously quipped, "It's the economy, stupid." In other words, economics determines other realms of society. The verdict of Presidential forecasting models largely based on economic variables? By that measure, Kerry doesn't have a chance.
Yale University economist Ray Fair, an academic with wide-ranging interests, has designed the best-known Presidential election formula. Among the variables he has developed over the past quarter-century are incumbency (the power of a sitting President is formidable), the number of quarters inflation-adjusted growth in gross domestic product per capita exceeds 3.2%, and the inflation rate. The Fair model predicts that Bush easily triumphs in November.
Economy.com, a leading economic consulting firm, also tries to capture the pocketbook issues that would lead an electorate to favor or reject the incumbent. Income growth and inflation play a central role in its forecast, which includes a state-by-state breakdown. Economy.com last ran the model in early February, and Bush won by a landslide, with 488 electoral votes and 57.9% of the popular vote.
WHEN INCUMBENTS LOSE.
Not everyone agrees. An intriguingly simple analysis by economist Martin Mauro of Merrill Lynch suggests that the election may be much closer than more sophisticated models calculate. For most people, the economy isn't about growth rates in GDP, productivity, international trade, and other statistical measures. The measure of economic satisfaction is synonymous with jobs.
So, Mauro looked at the change in the unemployment rate in the four quarters up until the vote during the past 13 Presidential elections. The unemployment rate rose only three times since the 1952 election year -- in 1960, 1980, and 1992 -- and in each case the incumbent party lost. Comedian Billy Crystal captured this dynamic on Feb. 29 as he looked back to 13 years ago when he first hosted the televised Oscar awards ceremony. "Things were so different then," he said. "You know how different it was? Bush was President, the economy was tanking, and we'd just finished a war with Iraq."
Of course, forecasting is a hazardous business. Economic-based prediction formulas are far from infallible. For instance, Fair didn't predict Bill Clinton's victory over George Bush the First. The course of events in Iraq is a wild card, too, and it may be that international affairs -- and the threat of terrorism -- play a bigger role than typical in this election. So, follow the contest for the Presidency through the lens of the economy, just not too closely.
Farrell is contributing economics editor for BusinessWeek. His Sound Money radio commentaries are broadcast over Minnesota Public Radio on Saturdays in nearly 200 markets nationwide. Follow his weekly Sound Money column, only on BusinessWeek Online
Edited by Beth Belton