Search

We appreciate the opportunity to respond to the commentaries and additional analyses by Fornell et al. [Fornell, C., S. Mithas, F. V. Morgeson III. 2009a. The economic and statistical significance of stock returns on customer satisfaction. Marketing Sci. 28(5) 820–825] and Ittner et al. [Ittner, C., D. Larcker, D. Taylor. 2009. The stock market's pricing of customer satisfaction. Marketing Sci. 25(5) 826–835]. Both studies have multiple theoretical and econometric limitations that challenge the validity of their arguments and findings (e.g., neither study allows for time-varying risk factor loadings in their assessments of mispricing although the composition of firms in their analyzed portfolios changes over time, Fornell et al. mischaracterize the efficient markets hypothesis, and Ittner et al. do not use standard panel data econometric methods and models). Generalizations about customer satisfaction, like any other construct, should be assessed by appropriate econometric methods and should withstand rigorous scrutiny. We believe an open, frank dialogue can help clear up misconceptions, air central issues, and advance better understanding of methods and analyses for assessing the financial market implications of marketing metrics such as customer satisfaction.

We investigate the association between information contained in the American Customer Satisfaction Index (ACSI) metric and future stock market performance. Some past research has provided results suggesting that the financial markets misprice customer satisfaction; i.e., firms advantaged in customer satisfaction are posited to earn positive future-period abnormal stock returns. We reexamine this relationship and find that statistically significant evidence of financial market mispricing of customer satisfaction is limited to firms in the computer and Internet sector. The results suggest that the mispricing anomaly reported in past research appears not to stem from a systemic failure of the financial markets to impound the financial implications of customer satisfaction into current stock price, but rather from abnormal returns achieved by a small group of satisfaction leaders in the computer and Internet sector over the period of study. Analyses based on unconditional risk covariates and analyses using conditional risk covariates estimated from short-window, high-frequency data support this finding.

The authors assess which brand asset metrics provide incremental information content to accounting performance measures in explaining stock return. The analysis focuses on the five “pillars” (i.e., central brand attributes) that form the basis for the newly updated Young & Rubicam Brand Asset Valuator model: differentiation, relevance, esteem, knowledge, and energy. Analysis shows that perceived brand relevance and energy provide incremental information to accounting measures in explaining stock returns. However, esteem and knowledge do not; that is, their effects are reflected in current-term accounting measures and in brand relevance and energy. The financial markets do not view brand differentiation as having incremental information content, but they should. Changes in differentiation are indicative of future-term accounting performance, which in turn affects stock return. These conclusions are invariant to the use of alternative accounting performance measures, risk adjustments, and the inclu...

Managers often have incentives to artificially inflate current-term earnings by cutting marketing expenditures, even if it comes at the expense of long-term profits. Because investors rely on current-term accounting measures to form expectations of future-term profits, inflating current-term results can lead to enhanced current-term stock price. We present evidence that some firms engage in this type of “myopic marketing management” at the time of a seasoned equity offering (SEO). In particular, a greater proportion of firms than is typical report earnings higher than normal and marketing expenditures lower than normal at the time of their SEO. Although they realize that firms might be undertaking strategies to artificially inflate current-term earnings, the financial markets are not adequately identifying and properly valuing the firms doing so. Our results indicate that myopic firms are able to temporarily inflate their stock market valuation, but in the long run, as the consequences of cutting marketing spending become manifest, they have inferior stock market performance. We propose some actions that might reduce the incentives for myopic behavior.

Making use of the efficient markets hypothesis as a starting point for analysis, we outline an approach for assessing financial value-relevance of marketing metrics. We illustrate the approach by investigating the association between information contained in the American Customer Satisfaction Index (ACSI) customer satisfaction metric and stock market performance. Our findings differ from previous research in that the methodological approach we use, while nesting other models, incorporates a number of efficient market considerations that previous research has not taken into account, e.g., issues ranging from unmodeled autocorrelation, to risk adjustments, to using the unanticipated component of a series in financial market analysis. We find evidence supportive of satisfaction having value-relevance. However, the ACSI metric provides statistically significant incremental value-relevance to accounting measures only for firms in the computer/Internet sector. For other sectors, i.e., the majority of firms in the sample, unanticipated changes in satisfaction do not provide statistically significant incremental information to accounting measures in explaining stock returns.

Some past research has provided results suggesting that the financial markets mis-price customer satisfaction, i.e., firms advantaged in customer satisfaction are posited to earn positive future-period abnormal stock returns. By allowing for differences across sectors in the economy, we provide evidence that any mis-pricing, if it exists, is not widespread but rather is limited to firms in the computer and internet sector. Analyses based on unconditional risk covariates and analyses using conditional risk covariates estimated from short-window, high frequency data support this finding.

We assess which brand asset metrics provide incremental information content to accounting performance measures in explaining stock return. Our analysis focuses on the five "pillars," i.e., central brand attributes, that form the basis for the newly updated Young and Rubicam Brand Asset Valuator (Y&R BAV) model. Analysis shows that perceived brand Relevance and Energy provide incremental information to accounting measures in explaining stock return. Esteem and Knowledge do not, i.e., their effects are reflected in current-term accounting measures and in brand Relevance and Energy. We find that the financial markets do not view Differentiation as having incremental information content, when, in fact, they should. Changes in Differentiation are indicative of future-term accounting performance, which, in turn, affects stock return. These conclusions are invariant to the use of alternative accounting performance measures, risk adjustments, and to the inclusion of additional brand attributes into the analysis.