It's Not the Economy, Stupid
In the final days of this fall's campaign, Republicans have turned to an unexpected issue: the economy. President Bush touted the nation's prosperity last week, insisting that "a strong economy is going to help our candidates."
And why not? The Dow is soaring. Unemployment is low. Inflation is tame. Gas prices are falling. And the overall economy has been growing steadily. If Americans practice what political scientists call "retrospective voting" (captured by President Ronald Reagan's famous question: "Are you better off today than you were four years ago?"), then one would think that incumbent politicians should be cruising to victory.
There's just one problem: Despite the sunny talk and favorable numbers, voters aren't happy with the economy. Though they've become somewhat more positive in recent weeks, they remain strikingly dissatisfied -- and favor Democrats by wide margins on economic issues.
Many observers have attributed the disconnect to the war in Iraq. Unhappiness over the conflict, they argue, is coloring voters' economic perceptions. Yet a similar disconnect played out in the early 1990s, when voters also felt much more negative about economic conditions than the numbers would have suggested. Then, as now, conventional economic indicators didn't seem to capture voters' fundamental anxieties -- the sense that their jobs, health care, pensions and family finances were ever more at risk.
In this climate, Republicans bragging about the economy may appear not as saviors, but as out of touch -- much as the administration appears on the other dominant campaign issue, the war in Iraq. Just as events in Iraq undermine GOP reassurances about the conflict, so anxious financial discussions in American homes help explain why Republicans are gaining so little from today's economic numbers -- numbers that once would have seemed like tickets to victory.
In the past, the link between economic conditions and election results seemed simple. Economist Ray Fair's landmark 1978 paper "The Effect of Economic Events on Votes for President" in the Review of Economics and Statistics showed that a simple forecast based on election-year economic numbers (primarily inflation and economic growth) did a remarkably good job of predicting the result of presidential elections in the 20th century.
But models such as Fair's are becoming increasingly less reliable, according to Dartmouth political scientist Joseph Bafumi in a recent paper, "The Stubborn American Voter." For example, Fair's approach overshot President Bush's margin of victory in 2004 by 6.5 percentage points. And the last president to lose an election when the raw economic numbers predicted otherwise in Fair's equation was George H.W. Bush, running for reelection in 1992.
Some analysts have described current voter angst as a hangover of economic success. "Americans have developed perfectionist standards," economics columnist Robert J. Samuelson has argued. "We expect total prosperity and are disappointed by anything less." And conservative pundit George Will recently decried the nation's "economic hypochondria" -- an entitlement mentality characterized by a low threshold for economic pain.
But the problem isn't the public -- it's the standard statistics used to judge the economy. Inflation, unemployment and economic growth all capture economic performance at a particular moment or period. Yet a growing body of theory and evidence suggests that to understand public perceptions, one should look at the security and stability of family finances over time. With that perspective, the grounds for unease suddenly look much clearer.
Consider the evidence of rising income inequality in the United States. In a path-breaking recent paper, "The Evolution of Top Incomes: A Historical and International Perspective," Thomas Piketty of Écoles Normales Supérieure in Paris and Emmanuel Saez of the University of California at Berkeley have shown that the share of national income held by the richest 1 percent of Americans -- stable at about 32 percent throughout the middle decades of the 20th century -- began to rise sharply in the late 1970s and by 2002 had surpassed 40 percent. In the past few years, most income gains have gone to people at the very top of the income ladder, with middle-class Americans seeing only a small boost in their economic standing.
Yet there's another reason for middle-class dissatisfaction. Many assume that growing income inequality means that rich people are becoming steadily richer. But virtually all income statistics are based on annual snapshots of Americans' finances, so they cannot tell us whether rich people stay rich -- or whether poor people stay poor. In other words, these statistics tell us about inequality, but not about mobility -- either up the income ladder or down it.
This is a major oversight, because there's good reason to think that our economic lives are more unstable than they used to be. Bankruptcy, for instance, is much more common today than it was just 25 years ago, and research by Elizabeth Warren of Harvard Law School -- presented in a 2003 law review article, "Financial Collapse and Class Status" -- shows that many of those who file for bankruptcy were once squarely middle class. Princeton economist Henry Farber, in his article "What Do We Know About Job Loss in the United States?" has found that the likelihood that a worker will lose a job over a three-year period has been rising -- and is now about as high as it was in the early 1980s, which saw the worst economic downturn since the Great Depression.