An empirical analysis of the value of corporate reputation in explaining variations in price earnings ratios.

Little, Philip L. ; Jones, Elizabeth H. ; Jones, Gary H. 等

INTRODUCTION

The price earnings ratio has long been used as an investment analyst's tool to assess earnings growth potential and investment risks. Studies by Beaver and Morse (1978) and Zarowin (1990) confirm the long-held belief that variations in the price earnings ratio values are primarily influenced by expected earnings growth potential and perceived investment risk. However, both studies conclude that these factors do not explain all of the variation and suggest that further research be conducted to identify other important factors.

One such factor hypothesized by the aforementioned research is the effect of accounting method choice on variations in price earnings ratios. If two companies with equal economic circumstances choose different accounting methods to account for similar transactions, the price earnings ratios for the two companies will vary simply because their earnings per share differ. Given equal economic circumstances, the stock prices should be nearly the same. A study by Little (1998) concluded that a "quality of earnings" variable used as a surrogate measure for the effects of accounting method choices was significant in explaining some of the additional variance in price earnings ratios. Along with projected EPS growth and investment risk. However, the stud suggested that significant unexplained variance remains and that other factors should be explored.

In this study, we hypothesize that another important factor may be a corporation's reputation. Reputation can be viewed as capturing a combination of a firm's social and economic contributions. Previous research suggests that positive reputations allow firms to mitigate drops in stock prices in bear markets, charge premium prices, attract better applicants for its work force, attract investors, lower costs of capital, and enhance competitive status. (See, for example Beatty and Ritter, 1986; Caminiti, 1992; Milgram and Roberts, 1986; Fombrun and Stanley, 1990; Vergin and Qoronpleh, 1998; and Little, Jones, and Jones, 1999). If these assertions are true, then a measure of corporate reputation should contribute to explanation of variance in price earnings ratios beyond traditional measures of risk, growth potential, and quality of earnings. This research tests the significance of Fortune's reputation ratings in explaining variance in price earnings ratios in addition to the variables capturing expected earnings growth, investment risk, and quality of earnings.

BACKGROUND STUDIES

Background studies by Beaver and Morse (1978) and Zarowin (1990) conclude that indeed there are variables other than earnings growth potential and investment risk that explain differences in price earnings ratios. Beaver and Morse (1978), for example, found that earnings growth potential and risks on average explain approximately 50 percent of the variation in price earning ratios. Accordingly, they concluded that other factors important in explaining price earnings variation, such as the effects of accounting method choices.

Little (1998) examined the accounting method choice hypothesis for companies in the oil and gas industry. A variable named "quality of earnings" (cash flow operations divided by net income) was used to capture the accounting method choice phenomenon. The study concluded that "quality of earnings" did significantly explain an incremental portion of the unexplained variance in price earnings ratios after considering the effects of the traditional variables, protected EPS growth, and investment risk. However, a significant portion of variance in price earnings ratio remained unexplained, suggesting that other variables might be explored.

A number of studies have suggested that a company's reputation has a significant impact on stock prices. For example, Little, Jones, and Jones (1999) examined the relationship of a company's fortune magazine reputation rating to the size of stock price drops in stock market crashes of varying security. They found that reputation can mitigate the downward movement of stock prices especially when program trading is limited.

In another story, Vergin and Qoronpleh (1998) concluded that future stock market performance is directly related to reputation. They compared the 13-year stock price performance of firms in the upper tier of the reputation ratings with firms in the lower tier and the S&P 500 index. The upper tier firms showed an average increase of 20.1 percent in stock price over that period as compared to 1.9 percent for lower tier firms and 13.1 percent for the S&P index.

MEASURES

Financial variables

The values for the financial variables for 1992 were obtained from Value Line. The dependent variable, price earnings ratio, is represented by the value. The independent variables are five-year projects earnings per share growth, investment risk (beta) and quality of earnings (cash flow from operations2net income). These data were available from value line for 141 of the firms dated in the 1992 Fortune Magazine Annual Reputation Rating.

Corporate Reputation

The most widely-recognized measure of corporate reputation is the annual survey conducted by Fortune Magazine. For this measure, over 800 senior executives, directors, and financial analysts are asked to rate firms in their industry on a scale of zero (poor0 to ten (excellent) on each of eight attributes:

 * quality of management,

 * quality of good or services,

 * innovativeness

 * long-term investment value,

 * financial soundness,

 * ability to attract, develop, and keep talented people,

 * wise use of corporate assets, and

 * community and environmental responsibility.

This measure has been widely used. The concept is that reputation is multi-faceted construct, reflecting several stockholder groups, not just financial. While this measure purports to capture reputation as multidimensional, there is evidence that the mean value across the eight dimensions represents a unidimensional construct. Fombrun and Stanley (1990) demonstrate that not only are the scores on the dimensions highly intercorrelated with each other, but also that the scores on dimensions load on one factor. Accordingly, we use the overall reputation rating for our measure of corporate reputation.

RESEARCH METHOD

The 141 firms included in the 1992 Fortune reputation index for which, value line data were available, formed the basis of the sample. The regression model used to test our hypothesis is that corporate reputation is a significant variable in explaining variation in price earning ratios is expressed, as follows:

PE = Yo + Y1EEG + Y2B + Y3QE + Y4REP + e

Where:

PE = price earnings ratio--value line

EEG = projected five-year EPS growth--value line

B = beta per value line

QE = quality of earnings--value line:
Cash flow from operations / net income

REP = overall fortune 500 reputation rating

RESULTS

Table One displays the descriptive statistics for the variables used in the regression model for the 141 sample firms. As can be seen, the sample firms had price earnings ratios with a mean of about 18 and a standard deviation of about 8 and corporate reputation ratings with a mean of 6.61 with a standard deviation of .70. The corporate reputation mean is slightly higher than an "average" rating of 5.0 out of a possible 10.0.

Table Two displays the results of the regression model. As expected, the projected earnings per share growth variable is highly significant at the .001 level. This result is consistent with past research as well as conventional financial wisdom. Future earnings per share growth has always been considered the key factor in price earnings variation. Also, as expected, the beta (financial risk) variable is highly significant at the .0027 level. This result is also consistent with past research and conventional financial wisdom. The quality of earnings variable is significant at the .01 level confirming the finding of Little, 1998, that accounting method choices do explain some of the variation in price earnings ratios. Finally, as hypothesized in this research, the corporate reputation rating is highly significant at the .0036 level. This result indicates that a positive corporate reputation rating can lead to higher price earnings ratios even after controlling for future earnings per share growth, Beta, and accounting method choices. Thus, in addition to other benefits, a positive corporate reputation seems to aid in the goal of maximizing shareholder wealth.

Collinearity tests using the Belsley, Kuh, and Welch (1980) technique reveal that the regression model is well-conditioned. Additionally, regression coefficient signs are as anticipated. All the variables, except Beta, were expected to have a positive relationship to price-carrying ratios. For example, a higher corporate reputation rating leads to higher price earnings ratios with the other factors held constant.

CONCLUSIONS

The finding of this research indicates that positive corporate reputation ratings will lead to higher price earning ratios. This finding suggest that companies that work to improve their reputation will reap the benefits of a "reputation stock price premium" that will, in turn, help to achieve the goal of maximizing shareholder wealth.

Future research is important to determine if there are individual components of a corporation's reputation rating that is particularly important in explaining variation in price earning ratios. If so, companies could refine the process of improving their reputation by focusing on the individual components as opposed to the subjective whole.

REFERENCES

Beatty, R. P. & Ritter, J.R. (1986). Investment Banking, Reputation, and the Underpricing of Initial Public Offerings. Journal of Financial Economics, 15, 213-232.

Beaver, W. & Morse, D. (1978). What Determines Price-earnings Ratios? Financial Analysts Journal, July/August, 65-76.

Belsley, D., Kuh, E. & Welsch, R. (1980). Regression Diagnostics, New York: John Wiley and Sons.

Caminiti, S. (1992). The Payoff from a Good Reputation. Fortune, Feb. 10, 74-77.

Fombrun, C. & Stanley, S. (1990). What's in a Name? Reputation Building and Corporate Strategy. Academy of Management Journal, 33 (2), 233-258.

Little, P. (1998). Determinants of Price-earnings Ratios in the Oil and Gas Industry. Academy of Accounting and Financial Studies Journal, 2 (1), 101-105.

Jones, G., Jones, E. & Little, P. (1998). The Benefit of a Good Reputation: An Empirical Analysis. Academy of Managerial Communications Journal, 2 (1), 26-42.

Milgram, P.& Roberts, J. (1986) Relying on the Information of Interested Parties. Rand Journal of Economics, 17, 18-32.

Value/Screen II (1990). Value Line Publishing, Inc., New York, NY.

Vergin, R. & Qoronfleh, M. (1998). Corporate reputation and the Stock Market. Business Horizons, 41 (1), 19-30.

Zarowin, P (Summer 1990). What Determines Price-earnings Rations; Revisited. Journal of Accounting, Auditing, and Finance, 439-457.

Philip L. Little, Western Carolina University

Elizabeth H. Jones, Western Carolina University

Gary H. Jones, Truman State University

TABLE ONE: DESCRIPTIVE STATISTICS

Variable N Mean Standard Deviation

PE 141 18.44 7.93
EPSGR 141 2.47 1.11
BETA 141 1.10 0.18
QE 141 2.41 1.55
REP 141 6.61 0.70

Where:

PE = Price Earnings Ratios

EPSGR = Projected five-year EPS growth rate

BETA = Financing risk

QE = Cash flow from operations 2 net income

REP = Overall Fortune 500 corporate reputation rating

TABLE TWO: REGRESSION MODEL RESULTS

 Model
Variable Predicted Sign Coefficient (Prob > F)

Intercept ? -0.0436 (0.993)
EPSGR + 4.989 (0.0001)
BETA - -7.308 (0.0027)
QE + 0.988 (0.0132)
REP + 1.791 (0.0036)

Where:

EPSGR = Projected five-year EPS growth rate

BETA = Financing risk

QE = Cash flow from operations 2 net income

REP = Overall Fortune 500 corporate reputation rating