Consumer Spending:
A Sentimental Journey?

The terrorist attacks of Sept. 11 were followed by understandable predictions of a plunge in consumer spending. Those forecasts were based partly on various readings of consumer sentiment, which dropped sharply. Goodbye to one of the few pillars holding up a wobbly economy, observers prematurely concluded.

Truth turned out to be stranger than fiction. October 2001 will go down in economic history as the month Americans purchased the greatest number of autos -- ever. While that was largely attributed to auto makers' 0% financing offers, it still leaves unanswered why supposedly depressed Americans were plenty willing to visit -- of all places -- car dealerships. Harder to explain still is why the level of retail spending, after falling in September, has been higher in every succeeding month than it was in August. Indeed, even excluding autos, retail sales grew in every month during the past half-year compared with a year ago. All this while the most closely watched of the two main confidence gauges still hasn't topped pre-Sept. 11 heights.

Could it be that consumer sentiment isn't so closely tied to spending behavior after all? And if so, what does that mean for the U.S. economy now? Carl Steidtmann, chief economist of Deloitte Research, jointly owned by the accounting firm Deloitte & Touche and Deloitte Consulting, believes he has an answer.

Mr. Steidtmann compared changes in consumer confidence and consumer spending over the past 20 years. His finding: There is very little, if any, relationship between confidence and spending. There are a number of possible explanations for the disconnect, but the most important is that "spending and confidence are driven by a different set of factors," he says. Specifically, politics, disasters and war drive confidence, Mr. Steidtmann concludes, while cash flow drives spending. The twain may or may not meet.

After Sept. 11, consumer sentiment was buffeted by the terrorist attack, the anthrax mailings and fear of a second blow. But at the same time, cash flow was bolstered by lower interest rates, tax rebates, mortgage refinancings and lower energy prices.

Mr. Steidtmann's research is the latest entry in a debate that began at least as early as 1936 when economist John Maynard Keynes, in his groundbreaking book "The General Theory of Employment, Interest and Money," argued that a significant amount of the change in economic and business activity can be explained by "animal spirits," or inexplicable and unpredictable mood changes, rather than reasoned behavior. In other words, sentiment helps to determine economic activity.

In the late 1940s, George Katona, noted for attaining Ph.D.s in both economics and psychology, took Mr. Keynes's theory a step further by trying to quantify changes in consumer sentiment. The result was the University of Michigan's Index of Consumer Sentiment, which interviews about 500 individuals each month about their present financial conditions and current and future buying plans.

The Conference Board launched a competing index in 1967, with questions about how consumers felt about their job outlook. The job factor was lacking in the Michigan survey and was thought to be a significant factor influencing confidence, and thus spending. The Conference Board index, which Mr. Steidtmann used in his research, is based on a survey of 5,000 households.

But while economists and Wall Street closely watch the monthly releases of the two indexes, research has provided mixed support for the surveys' abilities to predict spending.

Federal Reserve governor Lawrence Meyer is among those who are unimpressed by the indexes. "The economy usually drives psychology and mostly not the other way around," he said recently.

Not that the two indexes are exactly alike. A 1998 paper by two economists at the Federal Reserve Bank of New York found that changes in the Conference Board index more closely tracked future changes in consumer spending than did the Michigan index. But the economists added, "our results do not prove that consumer attitudes cause changes in consumer spending."

Ken Goldstein, an economist at the Conference Board, concedes that his index won't tell investors whether spending will rise by 1% or 2% in the future, but when it comes to providing "an overall sense of whether the consumer market is building or losing momentum, this thing works like a charm."

If that's the case, consumer spending could surge in coming months along with the booming confidence numbers. The Conference Board's latest survey surged 15 points in March to 110.2, the highest level since the September terrorist attacks. The Michigan index grew by five points to 95.7, the highest level since December 2000.

But Mr. Steidtmann's preference for cash flow tells him the opposite could be the case. He is concerned about rising interest rates and energy prices and the absence of further stimulus in the form of tax rebates. "I think you are going to see consumer spending slowing down from here," he said.

Despite questions about the indexes, investors probably will keep looking to them for guidance. In an economy held up by consumers, measures that may be imperfect at predicting future spending remain better than none.

-- Steve Liesman

Write to Steve Liesman at steve.liesman@wsj.com¹

Updated April 8, 2002