With the Presidential election less than a year away, a growing number of economists are dabbling in polimetrics, the arcane science of calculating how economic factors influence voter behavior. True believers assert that it is not a President's public image or statements that determine whether he (or his party) remains in the White House, but the state of the economy in the period before the election.
Not surprisingly, the current polimetric omens are hardly propitious for a Bush victory. Economists at Merrill Lynch & Co., for example, note that during the first three years of the Bush Administration, real aftertax income grew less than 1% a year. "Except for the demobilization period after World War II," they observe, "that is the worst record since the Hoover Administration." In the entire postwar period, no President or his party has ever been reelected when real disposable income grew less than 3.8% in the 12 months prior to the election.
Economist Stephen S. Roach of Morgan Stanley & Co. believes Presidential elections are heavily influenced by recent changes in income tax rates (which, of course, affect the growth rate of disposable income). A rise in the average rate between Presidential elections, he says, "is almost always followed by a change of party in the White House."
Roach calculates that the current rate--based on federal, state, and local income taxes--is more than one-third of a percentage point higher than it was in late 1988 (chart). Further, he notes that rising state and local rates and changes in the federal tax code, such as phasing out the personal exemptions at upper-income levels, imply that the tax take will move higher over the next year.
The catch to these auguries of doom for Bush, of course, is that the economic outlook may well brighten significantly by next November. Indeed, research by Ray C. Fair of Yale University indicates that it is the real growth rate of per-capita gross national product in the nine months before a Presidential election that determines the result.
Fair's econometric model predicts the percentage tally of votes in all Presidential elections since 1916 with an average margin of error of only three percentage points. It indicates that Bush should win handily if real GNP rises by at least 3.5% next year, but would face a close election if economic growth is close to zero. "The biggest threat to Bush would be a protracted double-dip in 1992," says Fair.
All of which suggests that the Republicans will opt for fiscal stimulus if the economy doesn't revive by yearend. And if they do, they're unlikely to get much resistance from Congress, because history suggests that flagging economic conditions also raise the risk of defeat for all incumbent legislators, be they Democrats or Republicans.