What recession? What recovery? The arrival of the 21st century consumer: consumer ambiguity is likely to be a feature of the U.S. economy for some time to come

Business Economics, April, 2003 by Richard Curtin

Consumer confidence is a fairly accurate predictor of macroeconomic change. Recently, measures of consumer confidence have been giving ambiguous signals about consumer confidence and future economic growth. The factors behind the levels of consumer confidence as of the end of September 2002 are explored and provide insight into the determinants of consumer confidence and its importance as a predictor of the economy in the future. Also, the paper illustrates how the various aspects of consumer confidence can be used to interpret different facets of the economy, now and in the future.

**********

It is alleged that Winston Churchill once said that if you put two economists in a room, you would get two opinions, unless one of the two was Lord Keynes, in which case you would get three opinions. As Milton Friedman and Richard Nixon said (with different meanings), "We're all Keynesians now." The economic situation as of the end of 2002, and perhaps for some time to come, is perfectly suited to the economist who wants to end the forecast with "...on the other hand." Perhaps more importantly, consumers, who account for two-thirds of all spending, now frame their own forecasts in that same way. A broad range of offsetting changes in economic expectations has been recorded in the Surveys of Consumers conducted by the University of Michigan. Consumer expectations for jobs, incomes, interest rates, inflation, stock prices, home values, and taxes, to name just a few, have moved independently, and often in opposite directions during the past few years.

Given these divergent changes, it should be no surprise that the summary measure of trends in consumer confidence has hovered around its fifty-year average. While this middling position is hardly a positive development, the cross-currents that now exist among economic expectations do not make it unambiguously negative. Indeed, no headline-sized summary of the current state of consumer confidence is now possible as the more nuanced judgments made by consumers defy easy characterization. Consumers think it's a good time to buy vehicles but not PCs, a good time to buy homes but not household durables; consumers are pessimistic about wage growth but are optimistic about their future finances; they are more willing to incur debt, yet they have also increased their savings.

Perhaps the most surprising result has been stability in consumer confidence amid the steep falloff in business investment spending, plunging stock prices, the financial and accounting scandals, and the terrorist attacks. Who would have thought that businesses would be most susceptible to the irrational exuberance generated by the dotcom craze? More importantly, who would have thought that we would depend on consumers to provide stability to the macro economy? For the last seven consecutive quarters it has been business investment that has contracted and consumer spending that has continuously expanded. How long can that last? More importantly, will it be business or consumer spending that changes course?

An answer to this question requires an analysis of both recent developments in consumer expectations as well as the more fundamental criteria that consumers use to judge the performance of the economy. Both types of expectations have recently changed, and the both types of changes will have pervasive impact on developments in the economy during the years ahead.

Index of Consumer Sentiment

While the overall level of the Index of Consumer Sentiment is middling, it has not lost its power as a predictor of GDP (see Figure 1, both series are shown as year-to-year changes). Even after controlling for other variables typically used to predict GDP, the Index of Consumer Sentiment is a statistically significant predictor of future trends (Howrey, 2001). The survey indicated that the Sentiment Index posted its fourth consecutive small decline in September 2002. At 86.1, the Sentiment Index was just 4.3 Index points above the September 2001 low and 10.8 Index points below the year-to-date high. While the data continued to indicate that consumer confidence remained in the middling range, the recent decline indicated an even slower, but still positive, pace of growth in 2003.

Given the terrorist attacks of September 11, 2001 and the growing winds of war, these are hardly normal times. The September 2002 data provided no evidence that the prospective war with fraq had yet had much impact. Nonetheless, consumer confidence is vulnerable to the same factors that surrounded the Gulf war a decade ago. Indeed, the September 2002 level of the Sentiment Index was nearly identical to its level in the month prior to the invasion of Kuwait by Iraq. The sharp increase in gasoline prices in August 1990 caused an immediate plunge in consumer confidence, with the Sentiment Index falling by twenty-five percent in just three months.

To be sure, there have been significant developments since 1990 that could fundamentally alter consumers responses to the same set of factors. Consumers have experienced repeated and temporary surges in gasoline prices as well as repeated and successful military campaigns. While it is impossible at this early stage to know how consumers will respond, I do not think the reaction of consumers will be as extreme as in 1990.