NEW YORK — Is predicting consumer spending behavior more of an art or more of a science?
Author-consultant Joseph H. Ellis has taken a strong stand in the corner of science, an argument he presents in his book “Ahead of The Curve: A Commonsense Guide to Forecasting Business and Market Cycles” (Harvard Business School Press, $29.95), published in October.
“Stick with the numbers, lose the psychology,” asserted Ellis, advisory director of Goldman, Sachs & Co., where he serves as the firm’s principal liaison and consultant with retailers. “We have a zero percent savings rate,” Ellis said in an interview. “Americans spend what they earn. So, how much they earn really determines what they buy.”
In building his case, the author discounted the predictive power of many other economic variables to forecast how much people will spend, such as the record levels of debt with which the country’s consumers are currently saddled, fluctuations in consumer confidence and swings in the employment picture. After analyzing about 30 years of economic data, Ellis said, he has found the best predictor of consumer spending is real average hourly earnings.
Those incomes are a leading indicator of spending, while the job picture is a lagging indice, Ellis reasoned, because “the hourly earnings of the [roughly] 95 percent of those who are employed are more powerful than the 5 percent of those who are unemployed.”
Nonetheless, much of the country’s forward-looking analysis has been focused elsewhere, the author acknowledged. “Right now, it’s a morass,” he charged. “We’re all playing hunches based on all this anecdotal stuff,” he said of measures such as consumer confidence, which he described as “appealing because it’s easy to relate to,” but as only “coincident” to consumer spending. That’s because when plotted across the past five decades, people’s answers to forward-looking questions about their anticipated financial status and the economy’s health have mostly failed to prove a reliable indicator of their subsequent spending.
In one of the many charts throughout “Ahead of The Curve,” Ellis plots the paths of real average hourly earnings and real personal consumption expenditures, between 1965 and 2004, to illustrate his contention the wage picture portends consumer spending patterns. He observes that “real hourly earnings gave particularly notable advance warning of the 2000-2002 economic downturn.”
Related Articles
You May Also Like
In addition, Ellis points out an anomaly in this data: In the mid-1980s and in 2003, strong gains in consumer spending occurred despite flat-to-declining real wages, increases he attributes to federal tax cuts, as real hourly earnings are reported on a pretax basis.
Further, it’s important to keep year-over-year data in sight to identify the big picture trend, rather than relying heavily on quarter-to-quarter or month-to-month fluctuations — a chain of thinking yet to be broken.
How reliable are such year-to-year patterns for a business like fashion, where fortunes can shift sharply from season to season? It works best, Ellis said, for retailers that are broad-based, such as department stores, discounters and specialty retailers, as well as for fashion houses with multiple divisions, like Liz Claiborne and Gap Inc. Less so for, say, a Sean John or a Pac Sun. “The greater the number of [business] units and the diversity of customers, the more cyclically linked” Ellis expects real hourly earnings and consumer spending will be.