The latest news on consumer confidence is depressing. On Tuesday, the Conference Board reported that its widely followed Consumer Confidence Index tumbled in October to 79.4, a nine-year low. The 14.3-point drop from September was nearly five times greater than economists expected.
Every recession in the past 40 years has been preceded by drops in consumer confidence. (Although not every drop in consumer confidence has presaged a recession.) In an economy where consumer spending accounts for 73 percent of all activity, couldn't four straight monthly declines in consumer confidence lead inexorably to recession?
The Bush administration certainly hopes not. Glenn Hubbard, chairman of the Council of Economic Advisers, brushed aside the numbers, implying that they reeked of mumbo-jumbo psychology rather than scientific economics. Hubbard was confident that his tough-minded colleagues would focus more on hard data than soft sentiment. "Economists will be watching what consumers do more than what they say," he told the Wall Street Journal. "Confidence has declined since May, but household demand has been strong. With rising incomes, consumer spending should grow reasonably well during the next year of the recovery."
It's tough to take Hubbard's word for it. The groups that compile the two rival consumer confidence indices—at the University of Michigan and the New York-based Conference Board (read Slate's " Explainer" about them)—and other disinterested economists have found distinct correlations between the way people feel and the way the economy actually performs.
While their methodologies vary, both indices survey Americans about financial conditions, recent economic activity, and future plans, as well as assessments of the current and future economic situation. The idea of the surveys is commonsensical: When people are optimistic, there's a good chance they'll buy more. When people are pessimistic, they'll postpone buying a new suit.
Each index has two components: current conditions and expectations. The current condition number is a mirror, a real-time reflection of how people feel about the economy. Since consumer confidence figures are published before data on sales and unemployment, we know how people think the economy is doing in November weeks before we'll actually know how the economy performs that month.
By contrast, the expectations measure is a crystal ball. "Our sentiment index is a leading indicator," says Richard T. Curtin, director of Surveys of Consumers at the University of Michigan.
In a 1998 article titled "Does Consumer Confidence Forecast Household Expenditure?" Jason Bram and Sydney Ludvigson, economists at the Federal Reserve Bank of New York, set out to determine whether consumer sentiment actually provides economically meaningful information about future consumer spending—apart from the data contained in other economic indicators. In other words, do we spend more when we feel good about the economy than when we feel worse, even if actual economic conditions are the same. The answer is a qualified yes. The expectations data seems to correlate favorably with actual performance in areas such as total personal consumption.
In a more recent study, University of Michigan professor Phil Howreyexamined the predictive power of Michigan's Index of Consumer Sentiment. He showed that the ICS could improve the accuracy of predictions of recession and recovery. A sharply declining ICS does not guarantee economic contraction, but it is a pretty good warning signal.
As for predicting personal consumption, Howrey concluded that while it is "statistically significant and economically meaningful … the relationship between personal consumption expenditure and the ICS is very noisy." Which is to say that that confidence and personal spending seem to move in the same direction, but it's difficult to figure out why.