Opinion

Yes, We’re Confident, but Who Knows Why

Robert J. Shiller — March 2013

As housing, unemployment, the stock market, and the overall economy show signs of recovery, Professor Robert Shiller writes in the New York Times that we understand little about how people’s confidence affect these major turning points. "…[P]ublic thinking is inscrutable. We can keep trying to understand it, but we’ll be puzzled again the next time the markets or the economy make major moves."

Published in the New York Times on March 10, 2013.

The recovery in housing, the stock market and the overall economy has finally gained sustainable momentumor so it is said.

That opinion seems to be based on several salient facts. Unemployment has been declining, from 10.0 percent in October 2009 to 7.7 percent last month. More spectacularly, the stock market has more than doubled since 2009 and has been especially strong for the last six months, with the Dow Jones industrial average reaching record closing highs last week and the S & P 500 flirting with superlatives, too.

And the housing market, seasonally adjusted, has been rising. The S & P/Case-Shiller 20-city home price index gained 7 percent in 2012.

These vital signs make many people believe that we've turned the corner on the economy, that we've started a healing process. And their discussions often note one particular sign of systemic recovery: confidence. There is considerable hope that the markets are heralding a major development: that Americans have lost the fears and foreboding that have made the financial crisis of 2008 so enduring in its effects.
 
Hope is a wonderful thing. But we also need to remember that changes in the stock market, the housing market, and the overall economy have relatively little to do with one another over years or decades. (We economists would say that they are only slightly correlated.) Furthermore, all three are subject to sharp turns. The economy is a complicated system, with many moving parts.

So, amid all those complications, there are other possibilities: Could we be approaching another major stock market peak? Will the housing market's takeoff be short-lived? And could we dip into another recession?

There are certainly risks. Congress is mired in struggles over the budget crisis and the national debt. The government is questioning the risk to taxpayers in its huge support of housing through Fannie Mae, Freddie Mac, the Federal Housing Administration, and the Federal Reserve. Problems in Europe, Asia, and the Middle East could easily shift people's confidence. There have been abrupt and significant changes in confidence in European markets since 2009. Is there any reason to think that the United States is immune to similar swings?

For years, I've been troubled by the problem of understanding the social psychology and economic impact of confidence. There hasn't been much research into the emotional factors and the shifts in worldview that drive major turning points. The much-quoted consumer sentiment and confidence indexes don't yet seem able to offer insight into what's behind the changes they quantify. It also isn't clear which factors of confidence drive the separate parts of the economy.

Along with colleagues, I have been conducting surveys about aspects of stock market confidence. For example, since 1989, with the help of some colleagues at Yale, I have been collecting data on the opinions and ideas of institutional investors and private individuals. These data, and indexes constructed from them, can be found on the web site of the Yale School of Management.

I have called one of these indexes "valuation confidence." It is the percentage of respondents who think that the stock market is not overvalued. Using the six-month moving average ended in February, it was running at 72 percent for institutional investors and 62 percent for individuals. That may sound like a ton of confidence, but it isn't as high as the roughly 80 percent recorded in both categories just before the market peak of 2007.

How do these figures relate to other stock market measures? I rely on the measure of stock market valuation that Prof. John Campbell of Harvard and I developed more than 20 years ago. Called the cyclically adjusted price-earnings ratio, or CAPE, this measure is the real, or inflation-adjusted, Standard & Poor's 500 index divided by a 10-year average of real S & P earnings. The CAPE has been high of late: it stands at 23, compared with a historical average of around 15. This suggests that the market is somewhat overpriced and might show below-average returns in the future. (The use of the 10-year average reduces the impact of short-run, or cyclical, components of earnings.)

For perspective, compare today's valuation, confidence and CAPE figures to those of other important recent periods in the stock market. In the spring of 2000, a sharp market peak, only 33 percent of institutional investors and 28 percent of individual investors thought that the market was not overvalued. The CAPE reached 46, a record high based on data going back to 1871. (For the period before 1926, we rely on data from Alfred Cowles 3rd & Associates.) Yet most respondents in 2000 thought that the market would go up in the next year, so they hung in for the time being. That suggests that the 1990s boom was indeed a bubble, with investors suspecting that they might have to beat a hasty exit. They ended up trying to do just that, and brought the market down.

But then consider the valuation confidence in October 2007, another major peak, after which the stock market fell by more than 50 percent in real terms. At that peak, the CAPE was at 27—a little higher than it is now, though not extraordinarily lofty. In 2007, valuation confidence was 82 percent for institutional investors and 74 percent for individual investors, or not far from today's levels. Investors at the time didn't think that they were floating on a bubble, and they saw the probability of a stock market crash as unusually low. Yet a plunge soon occurred. The cause appears not to have been so much the bursting of an overextended bubble but the subprime mortgage crisis and a string of financial failures that most investors couldn't have known about.

Clearly, confidence can change awfully fast, and people can suddenly start worrying about a stock market crash, just as they did after 2007.

Today, the Dodd-Frank Act and other regulatory changes may help prevent another crisis. Even so, regulators can't do much about some of the questionable thinking that seems to drive changes in confidence.

For example, why is a record high in the United States stock market a reason for optimism? Nothing is remarkable about reaching a market record: the S & P Composite Index has done it 1,007 times, based on daily closes, since the beginning of 1928. That's about once every 23 trading days, on average, though the new records tend to come in bunches.

The important fact is that we haven't set a nominal stock-market record in six years. And we haven't set one in 13 years if we use the inflation-corrected S & P Composite total-return index. That this index may be about to set a record means only that we haven't made any real money in the stock market in 13 years, which hardly seems a reason for confidence.

But public thinking is inscrutable. We can keep trying to understand it, but we'll be puzzled again the next time the markets or the economy make major moves.