July 22, 2010

The Mismeasure of Mispricing: The Case of Customer Satisfaction

New research from Natalie Mizik dismantles the notion that the stock market undervalues firms that earn high marks from consumers.

The customer may always be right, but satisfied customers might matter less for some industries than others. The high cost of entry into the utility industry, for example, helps ensure little competition and a high rate of customer captivity for existing utility companies, which need not compete on customer satisfaction to maintain market share and sustain demand for products and services.

Customer satisfaction is tricky to price because, like branding strategy or research and development, it falls into the category of intangible assets, which are difficult to measure and value. For the last decade, a number of marketing experts advanced the notion that the stock market undervalues firms with high customer satisfaction scores, reporting that stock prices continue to rise in response to high scores on customer satisfaction surveys for months, rather than days, after such surveys are published. A large body of research has grown around this idea, and a number of investment advisers promoted trading strategies designed to exploit it.

But Professor Natalie Mizik was unconvinced: if the market doesn’t immediately recognize customer satisfaction, why would the stock prices of firms in industries less sensitive to customer satisfaction behave like those from industries where customer satisfaction is paramount to business success? She hypothesized that the aggregate data could be masking sector-specific phenomena.

Mizik was also curious about more than the relationship between stock price and customer satisfaction. Other researchers have pointed to the volatility and bubble-like patterns in the market following big technical revolutions: new types of firms behave differently than established ones, and the market fluctuates more than usual as it calibrates to the new industry.

Mizik’s data sets matched those used by the researchers who first identified customer satisfaction mispricing, covering the late 1990s and early 2000s: the Internet boom and bust. The emergence of online retailing clearly fell into the category of a revolution — one that struck Mizik as dramatic enough to skew results.

Working with Robert Jacobson of Diogenes Consulting in Seattle, Mizik took another look at the association between high customer satisfaction and stock prices. The researchers matched the daily and monthly stock returns of more than 100 prominent U.S. firms covered by the American Customer Satisfaction Index (ACSI) survey over the ten-year period from November 1996 to August 2006. (ACSI publishes a survey-based measure of consumer satisfaction for a broad spectrum of industries in the United States at regular intervals.)

The researchers capitalized on a new methodology in financial research — one that Professor Andrew Ang helped pioneer — using time-varying risk factor models to allow for risk premium changes over time. “Firms change; portfolio compositions change,” Mizik says, “and that must be accounted for.”

When the researchers aggregated data for all industries, their results closely replicated those of the past research. But then they segregated out utilities and Internet firms from the rest of the data sample and examined each of those three sectors — utilities, Internet firms, and all other firms — separately.

What did they find? “It is all, pretty much, about the Internet bubble,” Mizik says. “If you isolate the Internet firms — and there are only seven of them in the total sample — there is no association between ACSI and future stock returns.” The previous findings that suggest widespread mispricing of ACSI across industries appear to be driven by this small set of outliers in the Internet sector.

“Because Internet firms exhibited this dynamic, their characteristics and pattern of returns were very different from other firms. Valuation models used for them at the time by the market were probably different,” Mizik says. “These differences need to be recognized and accounted for in the model. Once you do that, all of the mispricing results for customer satisfaction disappear.”

The results suggest that financial market mispricing of customer satisfaction, if it exists, is not widespread but is instead limited to firms in the computer and Internet sector.

“Customer satisfaction has different implications in different industries, and different value implications are reflected in different valuations,” Mizik says. “It’s important to look at the same types of firms, or firms operating in a similar business environment.”

Natalie Mizik is the Gantcher Associate Professor of Business in the Marketing Division at Columbia Business School.

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>