Yet after decades of these surveys, the link between them and actual behavior remains tenuous at best. We all suspect that such hard-to-measure things like feelings and attitudes shape our behavior, and they most certainly do. How they do is another question. People seem quite capable of being anxious about the world and spending money at the same time, and equally capable of feeling confident and saving.
The surveys were developed in the 1940s largely by one man, George Katona, a Hungarian émigré to the United States steeped in both economics and psychology. Katona wanted to rectify an imbalance then emerging in economics that emphasized states and systems and ignored people and their less-predictable emotions and behavior. Katona saw the world not as a mechanical system to be charted but as a product of shifting human attitudes. His surveys were part of a larger ambition to build a database of human desires, inclinations, fears, and hopes that dictated economic behavior.
Katona started with the assumption that what people believe about the state of the economy intimately shapes their spending decisions. “The better off people feel, the more they spend,” he said simply. “The worse off they feel, the less they spend.”
The sentiment surveys that are kept today are remarkably similar to what Katona first developed at the University of Michigan in the late 1940s. They ask five basic questions about what people felt about business conditions and whether now is a good or bad time to make major household purchases. At first, these surveys were avidly consumed by economic policymakers in the Federal Government. Soon, however, many began to question whether they were of much predictive use.
Katona argued that what economic models treated as soft – emotions, attitudes, and individual opinions – were crucial variable that the economic profession arrogantly ignored. Not surprisingly then, he was almost immediately question and challenged by the establishment that he was questioning and challenging.
In 1955, the Federal Reserve commission a panel chaired by future Nobel laureate James Tobin to investigate whether the survey predicted consumer behavior. The committee concluded no. Katona pushed back, saying that you couldn’t verify the surveys as predictive by going back and re-interviewing people to see if they behaved as they had said they would. You had to treat the surveys as indicators of broad trends in spending patterns.
Yet even on that score, future study failed to show a clear and reliable connection between sentiment survey and future spending or savings. Why then have they become so embedded in our culture under the mistaken assumption that they do?
Even though no one has been able to prove that the indices predict anything, few are willing to dismiss them altogether. As Katona argued, separate from what they predict, the do shine a light into the murky psyche of individuals and say something about how people are emotionally experiencing this thing called “the economy.”
In addition, there is some indication that the surveys do offer probabilities of whether people will buy big-ticket items like a car. The reason is unclear, but it may be that such items require some planning. After all few of us buy cars on impulse, with a “Hi there, I was just passing by and decided to pick up that Tesla.”
Sentiment surveys have also taken hold as one of the few consistent measures of what most of our other economic indicators lack: real people and real human emotions. In the past few years, behavioral economics has seen a surge of interest, and Katona was in many ways of that tribe, though he lacked the scientific and experimental rigor of many later behavioral economists. The enduring interest of consumer sentiment surveys speaks to the glaring gap in traditional economics and macro-economic measures: how flesh-and-blood people act and feel.
That has remained peripheral to how economics as a profession has evolved. Seeing economic systems ever more in terms of science, emotions and irrationality were left out the equations, and the result was that many assumptions turned out wrong. People are often not rational actors, and models break down.
Sentiment survey may promise far more than they deliver in terms of what they predict. Income and access to credit are much better predictors of future spending than individual intentions and feelings. But by emphasizing the human element as an ever-present and crucial aspect of how economies behave, consumer sentiment surveys filled a vital role, and continue to remind us that people are not mindless cogs in an economic machine. They are unpredictable all-too-human actors in a complicated, shifting set of systems we call “the economy.” Integrating that into our sense of what is likely to unfold in the future may not give us the crystal ball we desire, but it will at least keep us humble in the face of false certainty that we can predict how unpredictable humans will act.
Part of TPM Cafe Book Club for THE LEADING INDICATORS: A Short History of the Numbers That Rule Our World. Zachary Karabell is an author, money manager, commentator, and head of global strategy at Envestnet. He is also the president of River Twice Research, where he analyzes economic and political trends. Educated at Columbia, Oxford, and Harvard, Karabell has written eleven previous books. He is a regular commentator on CNBC, MSNBC, and CNN. He writes the weekly “Edgy Optimist” column for Reuters and The Atlantic, and is a contributor to The Daily Beast, TIME, The Wall Street Journal, The New Republic, The New York Times, and Foreign Affairs.