Why You Probably Shouldn't Pay Too Much Attention to Consumer Confidence Indicators

Consumer confidence. The phrase sounds so definitive. Like those that use it just know what consumers in all their variety are thinking. And perhaps of even greater concern, that all those faceless consumers are voting on economic conditions and the policies of those attempting to influence them.

While there has always been some skepticism about how accurate such comprehensive metrics may be, recent events have cast further doubt on the efficacy of these measures. It seems that during the US government in shutdown this past October, an event freighted with apocalyptic portents, the latest data now show that consumers pretty much went about their business rather than sitting at home cleaning their weapons, canning vegetables and praying for salvation. Unless, of course, they were government employees or contractors whose incomes had actually be cut.

Before we cast consumer confidence in the dustbin of history, however, it is worth remembering that in an age of intangibles, emotions and passions have become important to marketers. Confidence, depression, glee and worry all matter, but in different ways, at different times and to different groups.So getting at the impact of these feelings on actual spending behavior is a useful exercise.

The issue may be not that consumer confidence is a false or useless metric, but that the way it is defined, collected and interpreted may need to be reexamined if it is to continue to have meaning in an increasingly metri-centric society with access to lots of data, lots of alternative explanations or opinions about what it means - and lots of ways to figure out what's in it for them. JL

Neil Irwin comments in the Washington Post:

In terms of immediate economic impact, as long as markets don't panic, neither do consumers.
Hey, remember that time the federal government shut down, there was a debt-ceiling standoff, and consumer confidence got hammered?
It was a pretty nasty affair. Took up the whole first half of October. The Conference Board's consumer confidence index plummeted from 71.2 from 80.2, battering the fragile psyche of American households.
Now we have some more solid data on what exactly consumers did in October. And it turns out that, well, the effect was something approaching zero. Retail sales rose 0.4 percent, the Commerce Department said. Excluding volatile auto sales, the number was up 0.2 percent. Both numbers beat analysts' expectations.
On one level, this is a reminder that consumer confidence surveys have a quite poor track record of predicting actual economic activity. Here's the chart comparing percent change in the University of Michigan's consumer sentiment index with percent change in personal consumption expenditures.

Source: University of Michigan and Bureau of Economic Analysis, via St. Louis Fed's FRED database

The two have only a 4.5 percent correlation from 1978 through the present, meaning that knowing what happened to one tells you pretty much nothing about what happened to the other.
It is true that the last time there was a debt-ceiling standoff, there was a measurable decline in consumer spending and other real economic measures. But that also occurred at a time that the euro-zone crisis was at full boil, and the combination of the two created massive stock-market volatility. This time around, there were some rumbles in the financial markets, but nothing dramatic. Measures of market volatility got nowhere near their July 2011 levels.
In other words, it seems to be the case that these debt standoffs affect actual consumer behavior -- and thus economic growth -- only through financial-market channels, rather than directly. No one would argue that such episodes are good for the economy, exactly. Over time they can have a damaging impact on psychology that doesn't show up in any one month's numbers.