Sentiment Data Need Attitude Adjustment

Consumer confidence doesn't always translate into spending, and it's a poor indicator of market direction.
Author:
Publish date:


On balance, it is probably best to like what you see when you look in the mirror, a nugget of folk wisdom perhaps understood better by viewers of Oprah and Dr. Phil than by viewers of CNBC. Wall Street periodically lapses into paroxysms of self-loathing -- as was the case last Friday upon the release of the monthly University of Michigan consumer sentiment survey data -- whenever a mysterious group called "the consumer" neither acts nor thinks as desired.

Consumers -- and perhaps you know one or two -- are notoriously fickle both in expressing their attitudes and in the conversion of these attitudes into actual behavior as reflected in their purchasing decisions. Think about the percentage of expenditures in your own life best classified as nondiscretionary: items such as food, housing and utilities. You can increase or decrease some of these, depending upon either your mood or your immediate feelings of wealth, but you are unlikely to turn the thermostat up to celebrate a good day in the market.

And how many of life's little misfortunes, such as the car needing repair or your appliances staging their own version of mass civil disobedience, tend to come at the most inopportune times? Yes, you had to go plunk down some cash and thus contributed to the retail sales component of GDP, but were these expenditures motivated in any way by feelings of confidence or optimism? In addition, your reactions to your own sentiment are likely to be asymmetric; durable discretionary purchases can be postponed more easily than accelerated, especially for big-ticket items such as automobiles or major appliances.

A Misguided Relationship

Even with all of the caveats about actual retail sales and what they represent about underlying consumer attitudes in mind, the cash register is still the most sophisticated device ever developed for measuring behavior. The Commerce Department revised its classification scheme in mid-2001; a consistent history is available only going back to January 1992. Over the life of this data series, the Michigan consumer sentiment data not only have no demonstrable leading relationship to retail sales, but the opposite relationship appears to exist. The year-over-year percentage changes in total retail sales have led changes in consumer sentiment by a variable interval ranging from two to six months.

Does Consumer Sentiment Lead Retail Sales, or Vice Versa?

Source: Bloomberg

Comparing Expectations

If we switch the time frame of the comparisons from current to expected and replace the retail sales data with the

S&P 500

, we can address whether consumer expectations are or should be relevant for investors. The S&P 500 monthly data are smoothed by comparing a rolling three-month period to a rolling 12-month period in the chart below.

Consumer Expectations and Stocks

Source: Bloomberg

Here the question of which series leads and which lags is quite apparent, both on the upside and on the downside. The stock market, which tends to lead business conditions by six to nine months, tends to lead changes in consumer expectations by a similar span. Restated, the consumer expectation number is known well in advance by anyone with access to a newspaper or quote screen.

Consumer expectations, unsurprisingly, are a function of personal income as well. Consumer expectations appear to be a function of the year-over-year changes in the Commerce Department's personal income data deflated by the consumer price index. The psychological tendency to extrapolate recent trends is powerful, and in this instance it appears to have a slight optimistic bent as the relationship accelerates with stronger real income growth.

When the Present Is Good, the Future Looks Better

Source: Bloomberg

Does One Market Ultimately Lead?

If retail sales lead current consumer sentiment and stock prices lead future consumer expectations, do either retail sales or stock prices lead each other? In the limited data sample available since the Commerce Department switched its calculation basis, the relationship appears indeterminate.

Do Retail Sales Lead Stocks?

Source: Bloomberg

Once again, this conclusion is unsurprising. The so-called wealth effect, or percentage of stock markets gains likely to be converted into spending, is not linear. The monthly relationship since 1959, which would be linear with a slope of 4% if that mythic number bandied about during the bubble actually was demonstrable, shows a distinct diminishing return on stock gains.

Moreover, the bursting of the bubble, so apparent in the chart, was accompanied by an increase, not a decrease, in personal outlays.

A Diminishing Wealth Effect

Source: Bloomberg

The Sad Conclusion...

... Is we have to stop looking at individual economic data points with no demonstrable forecast content for trading direction. If consumer sentiment does not translate into retail sales, and if consumer expectations do not translate into higher stock prices, and if higher stock prices do not have a predictable wealth effect, then why should anyone pay attention to the sentiment data for trading purposes?

If this is reflective of a bad attitude, so be it.

Howard L. Simons is a special academic adviser at NQLX, a trading consultant and the author of

The Dynamic Option Selection System. Under no circumstances does the information in this column represent a recommendation to buy or sell securities. The views expressed in this article are those of Howard Simons and not necessarily those of NQLX. As a matter of policy, NQLX disclaims the private publication of materials by its employees. While Simons cannot provide investment advice or recommendations, he invites you to send your feedback to

howard.simons@thestreet.com.

TheStreet.com has a revenue-sharing relationship with Amazon.com under which it receives a portion of the revenue from Amazon purchases by customers directed there from TheStreet.com.