Is board quality an indicator of a firm's future performance?

Hayes, Rick Stephan ; Lee, Dong-Woo

INTRODUCTION

The world stock markets have shown extreme volatility in the first nine months of 1998. On August 31 the Dow Jones Industrial Average (DJIA) dropped 512.61 points (6.73%), the largest one-day point drop since October 1997. This came after a week which was the DJIA worst one-week loss in the history of the index. That day the Nasdaq Composite Index dropped to 1499.15, more than 25% off its all time high. The very next day, September 1, the DJIA was up 288.36 points, followed in one week (September 8, 1998) by another 380.53 increase. Stock Exchanges outside the US are also volatile. The stock exchanges in Russia (where the Russian Trading System index plunged 17% in a single day), and Eastern Asia have caused worldwide concern, increasing fears of assets being liquidated in Brazil and Argentina. The European stock markets were down by 17% from their 1998 peaks. In the same week in August when the DJIA had its worst drop, Japan's Nikkei 225 index dropped 9% to below 13915.63, a 12-year low.

What type of companies can maintain their stock prices and increase cumulative stockholder returns in times like these? In this paper, the results show that it is those companies with the highest quality board of directors that perform best.

High quality boards of directors follow the interest of shareholders by monitoring the excesses of management. Fama and Jensen (1983) describe the board of directors as the highest level control system of corporations. Fama (1980) views the role of board of directors as an internal monitoring mechanism. If board of directors effectively monitors the management decisions, the performance of such companies will exceed those of other companies with less effective boards of directors. Shivdasani (1983) argues that the ineffectiveness of the board is a cause of hostile takeovers and other agency problems.

This paper examines the future performance and other characteristics of companies with boards of directors considered to be the best and worst in the U.S. The 25 best and 25 worst corporate boards that were identified by a survey of professional stock portfolio and pension managers and corporate governance experts performed by the Business Week organization and published in December of 1997. The companies were evaluated on the criteria of board independence, stockholder accountability, board quality and corporate performance. Companies were evaluated numerically on these criteria from two points of view: professional judgment and governance standards. Professional judgment was used by the professional portfolio managers. Governance standards were based mostly on the standards of the National Association of Corporate Directors with professional judgment playing some part.

We examined if the firms rated highly or poorly performed better or worse than the market in a future period. We compared cumulative stock return, stock return subtracting the effects of the industry and market return on equity for these corporations over the period January 1, 1998 to September 8, 1998. The results show that for the best ten and worst ten and the best twenty and worse twenty, excess stock return over the market is significantly higher for companies with high quality boards of directors than for companies with low quality boards of directors. Cumulative stock return and excess stock return over the market for the 10 best boards were significantly better than the ten worst (at a p-value two-tailed of 0.009). The results indicate that independence, accountability and overall quality are important indicators of the future performance of a firm.

The rest of this paper is organized as follows: Section 2 reviews previous research on the role of corporate boards of directors. Hypothesis is developed in section 3. Section 4 explains the method Business Week used to identify the best and worst boards in America. Section 5 discusses empirical test design, and test results are presented in section 6. Section 7 provides conclusions.

THE ROLE OF CORPORATE BOARDS OF DIRECTORS

Positive theory attributes increases in shareholder wealth to effective monitoring of companies by stakeholders independent of management. If the monitoring stakeholders are also experienced in the company's core business and in service to other boards, and spend enough time in the monitoring job, actions at the company that reduce shareholder wealth would be less likely to occur. The board of directors has been viewed as an important internal corporate governance mechanism (Fama 1980). Boards supervise managerial actions, determine the level and structure of top management compensation, oversee the corporation's internal controls and accounting system, and determine the corporations major strategies and policies.

To establish the relationship between the quality of boards and corporate performance, good measures of board quality are essential. However, as indicated by Weisbach (1993), it is difficult to construct such measures. Previous studies used the proportion of outside directors, e.g, Weisbach (1988) among others, or number of shares owned by board members, e.g., Shivdasani (1993), as indirect measures of the effectiveness of boards in monitoring the activities of management.

Independence is characterized by the number of outside directors on a board, their exclusive presence on the audit, nominating and compensation committees; lack of income from the company; relationship to the company (relatives, former employees, former consultants); and "grey directors"

Weisbach (1988) found that a board dominated by outside directors increases the likelihood of a change in top management teams of poorly performing firms and attributes this to successful monitoring by outside directors.

Shivdasani (1993) used the fraction of outside directors on the board as a measure of the effectiveness of the board of directors. Insider-dominated boards imply problematic self-monitoring and particularly weak monitoring of the CEO, since the CEO is likely to be in a position to influence an inside director's career advancement within the firm.

Accountability relates to equity ownership and non-receipt of pensions from the company. Shleifer and Vishny (1986) demonstrate that equity ownership by large minority shareholders helps solve the free-rider problem in takeover bids. Shivdasani (1993) used the equity ownership by outside directors as a second measure of the effectiveness of the boards of directors in his study. Subrahmanyam, Rangan, and Rosenstein (1997) show that the ownership of outside directors is positively related with the abnormal stock return in banking industry.

The number of additional outside directorships held by each director may serve as a measure of his or her reputation as a monitor. The number of additional outside directorships held by outside directors was used by Shivdasani (1993) as his third measure of the effectiveness of the boards of directors. The positive accounting theory of monitoring is further discussed in Jensen and Meckling (1976), Morck, Shleifer, and Vishny (1988), Warner, Watts, and Wrunk (1988), and Beatty and Zajac (1994).

Previous studies use indirect measures of board quality due to the difficulties of empirically measuring the quality of boards. However, the board quality measure developed by Business Week provides an excellent opportunity to directly link the board quality and corporate performance.

DEVELOPMENT OF HYPOTHESIS

When the board is effective in performing its functions, agency costs decrease, increasing stockholder wealth (Fama 1980). Conversely, an ineffective board increases the likelihood of larger agency costs that lead to decreases in shareholder wealth. One indicator of this change in shareholder wealth may be an increased or decreased level of stock performance in relationship to the overall market. The effect of the board quality will be long-term in nature and will persist in the future periods. Therefore, the companies with good boards of directors will continue to perform better than those with poor boards of directors in the future. This leads to the hypothesis, in an alternative form, as follows:

Ha: Cumulative stock return, after adjusting for market return, will be greater for companies with best boards of directors than for companies with worst boards of directors during the period following th measurement of board qualities.

One important aspect of this hypothesis is that it is trying to predict the difference in stock performance for the period after the announcement of board quality ranking. This is important to avoid any concerns about the possibility that the historical performance of companies may affect the opinions of experts in determining the ranking of board quality. By using the cumulative stock return for the period from January 1, 1998 to September 8, 1998, the tests performed in this paper eliminate such possibilities, because the Business Week ranking was published in the December 8, 1998 issue which available to the public over one month before the beginning of the test period.

BEST AND WORST CORPORATE BOARDS OF DIRECTORS

This section explains the method used by Business Week to determine the best and worst corporate boards of directors. In scoring the best and worst boards, independence, accountability and quality of the corporations were considered from both the professional judgment view ("board performance poll") and from the governance standards view ("governance guideline analysis"). Louis Harris & Associates mailed a questionnaire to 371 of the US's largest pension funds and money managers and 50 of the nation's leading corporate governance experts, including academics, attorneys and activists (421 total). Of those queried, 103 replied, a response rate of 24.5%. The professionals surveyed were asked to identify public corporations with the most and least effect boards and grade them on a scale of 0 (poor) to 10 (excellent) on four criteria: independence, accountability to stockholders, quality of the directors, and corporate performance, which comprised the judgment view.

For the "governance guideline analysis", the 224 companies singled out as having either the most or least effective boards go another round of scrutiny. Their boards of directors were measured by Business Week by examining proxy statements in terms of a set of guidelines, or "best practices", commonly articulated by corporate governance experts. These "governance guideline analysis" scores had three components "shareholder accountability", "board quality" and "board independence"

The "board independence" component of governance standards view in the Business Week survey, a board scored points if it has no more than two inside directors; no insiders on the board's audit, nominating, and compensation committees; no outside directors who directly or indirectly draw consulting, legal or other fees from the company, , and no interlocking directorships. If outside directors meet regularly without the CEO present, extra points were given to the board.

To assess the "shareholder accountability" component of the "governance guideline analysis" score was accumulated (1) if all of the company's directors owned a minimum of $100,000 in stock, (2) no pensions were awarded to directors, and (3) if all board members stand for election each year.

To measure the "board quality" component of the "governance guideline analysis" a score was determined by awarding a company points (1) if its directors sit on no more than three boards and its retired directors sit on no more than six; (2) the board had at least one outside director with experience in the company's core business, i.e., a manager of a similar company or someone experienced in that industry; (3) at least one board member was a CEO of a company of similar size and stature; (4) if all the directors attended 75% or more of their meetings; and (5) if a board has no more than 15 directors.

The results of the board performance poll are summarized as "Survey Score" and the results of judgments based on governance guidelines are summarized as "Governance Guideline Analysis Score". To produce an overall ranking, the raw scores from the poll (professional view) and the board analysis (governance guideline analysis) were combined. A maximum of 100 points could be scored, half based on the pool and half on the analysis of proxy data.

EMPIRICAL TEST DESIGN

The sample used in this study includes the 25 best board companies and 25 worst board companies identified by Business Week's December 8, 1997 issue. Stock return data and other financial variables were obtained from the Briefing Books section of the Wall Street Journal Interactive Edition (http://www.wsj.com/). Historical stock price and Dow Jones Industrial Average data were obtained from the Web pages of Dreyfus Brokerage Services, Inc. (http://www.tradepbs.com/) and Dow Jones (http://www.dowjones.com/).

Stock performance of companies is measured by cumulative stock return (CR) for the period from January 1, 1998 to September 8, 1998. Market return for the corresponding period was subtracted from the individual company's cumulative stock return (CR) to calculate excess stock return over the market return (XMR). Excess stock return measures the relative performance of a company's stock to other companies in the market. Market return is the cumulative return of Dow Jones Equity Market Index over the corresponding period. Industry stock return is the cumulative return of the companies that belong to the same Dow Jones Industry Group. Excess stock return over the industry return (XIR) was calculated by subtracting industry return from the cumulative stock return of the company.

Return on equity represents the sum of net income for the most recent four quarters divided by the latest common equity. Debt-to-equity ratio is the ratio of long-term debt to common equity at the end of the most recent quarter in 1998. Total assets are reported as of the end of the most recent quarter in 1998.

TEST RESULTS

Table 1 reports the financial characteristics of best and worst board companies. Best board companies are larger in size than worst board companies. Median total assets for top 10 best board companies was $25,083 million, while top 10 worst board companies reported $6,930 million as the median total assets. Top 20 best companies were also larger than top 20 worst companies in size measured by total assets.

For return on equity and debt-to-equity ratio, worst board companies reported significantly wider ranges of ratios than those of best board companies. Return on equity for top 10 best board companies varied from -23.50% to 71.70% which was contrasted to top 10 best board companies whose return on equity varied from -155.60% to 422.50%. Wider variability was also noted for debt-to-equity ratio. The highest debt-to-equity ratio for top 10 worst companies was 169.66, which is substantially greater than the highest debt-to-equity ratio of 6.31 for top 10 best board companies.

Due to high volatility of ratios, mean and median values do not report consistent relationships. For return on equity, top 10 best board companies report higher median than top 10 worst board companies, however, mean was higher for top 10 worst board companies, due to an extreme value of 422.5%. The same situation was noted for debt-to-equity ratio. Mean debt-to-equity ratio for top 10 worst board companies was 17.8 which was more than 10 times higher than 1.2 for top 10 best board companies. However, median value was higher for top 10 best board companies (0.86) than for top 10 worst board companies (0.58).

The results reported in table 1 show that companies with higher quality boards of directors are larger in size and maintain more stable profitability and financial leverage than those with lower quality boards of directors.

Table 2 reports that best board companies report significantly higher stock return than worst board companies. Mean cumulative stock return was 18.58% for top 10 best board companies and -7.00 % for top 10 worst board companies. The same relationship was noted when top 20 best board companies ware compared with top 20 worst board companies. Mean cumulative stock return for top 20 best companies, 12.07%, was significantly higher than the mean for top 20 worst board companies, -1.27%.

Panel A of table 2 shows that top 10 best board companies earned 13.72% higher cumulative stock return over the market return for the period from January 1, 1998 to September 8, 1998. During the same period, cumulative stock return for top 10 worst board companies was 11.86% lower than the market return. The difference in cumulative stock return between top 10 best and worst board companies amounted to 25.59% for this period. This difference was statistically significant with a p-value of 0.009 as reported in table 3.

One interesting finding noted from the results reported in tables 2 and 3 was that there is an indication of industry clustering for the quality of board of directors. While this paper does not attempt to identify which industries tend to have higher quality boards of directors, the difference in cumulative stock return after adjusting for industry return was much smaller than the difference in stock return, which suggests that industry returns were also lower for worst board companies. Mean excess stock return over industry return was 2.38% for top 10 best board companies and -5.53% for top 10 worst board companies. However, the difference, 7.92%, was not statistically significant as reported in table 3 (p-value=0.383).

The difference in stock return between best and worst board companies became smaller as the sample include more companies from the top. The difference in stock return was 25.59% between top 10 best and worst board companies, which decreased to 13.33% when top 20 best and worst companies were compared. When top 25 companies were compared, mean difference was only 1.89% which is not significantly different from zero as reported in panel C of table 3.

The results show that companies ranked with best board quality earned significantly higher stock return in the future period, from January 1, 1998 to September 8, 1998, for those ranked with worst board quality. This supports the hypothesis that companies with good quality boards of directors maintain high management quality and continue to perform better than the companies with poor quality boards of directors.

Table 4 reports stock performance by sub-period. Because the period tested in this paper, January 1, 1998 to September 8, 1998, was a rather volatile period for stock market, additional analysis was performed to examine whether the results are due to unusual volatility of the market. For this sensitivity analysis of the results, stock performance was examined for three different sub-periods: (i) January 1 to August 28, 1998, (ii) August 31, 1998, and (iii) September 1, 1998. August 31, 1998 was the day with the second-largest point drop (512.61, 6.37%) for Dow Jones Industrial Average (DJIA). On the following day, September 1, 1998, DJIA rose 228.36 points. These two days were unusually volatile periods for stock market. August 28 was the last trading day before August 31, 1998.

Table 4 reports the percentage change in stock prices for these periods. Percentage change was calculated by subtracting the end of period price from the beginning of period price and by dividing that price difference by the beginning price. Unlike stock return reported in table 2 and table 3, cash distributions were not considered.

Panel A of table 4 shows that DJIA increased 1.81% for eight month period from January 1, 1998 to August 28, 1998. For the same period, top 10 best board companies reported 16.85% increase in stock price, which is contrasted with the 7.43% decrease for top 10 worst board companies. The mean difference between top 10 best and worst board companies was statistically significant with a p-value of 0.012. A comparison of top 20 best and worst board companies shows similar results.

For August 31 and September 1, 1998, best and worst board companies tend to move with a similar pattern. On August 31, stock prices of top 10 best board companies decreased by 7.40% which is slightly more than 5.22% decrease for top 10 worst board companies. Increases on September 1 were 5.41% and 4.14% for top 10 best and worst board companies, respectively.

The analysis reported in table 4 shows that the results of previous tables are not due to unusually volatile stock market. The difference in stock return was primarily due to changes in stock price for the period from January 1 to August 28, 1998, not the changes on August 31 or September 1, 1998.

The results support the hypothesis that companies with effective boards of directors outperform in stock return those companies with ineffective boards of directors. This is consistent with the argument that good boards of directors monitor the performance of the management more effectively than other companies. This provides evidence about the direct link between the monitoring role of board and stock performance of companies.

As an additional analysis, it is interesting to find that four (16%) of the 25 companies that were on the 1997 Worst Board list were acquired within seven months after they were ranked. Those companies were Digital Equipment Corporation (21st on the worst list), ITT Corporation (14th worst), Waste Management (16th worst) and Westinghouse (19th worst). Compaq Computer (ranked 3rd on the best board list) acquired Digital and digital stock stopped trading on June 11, 1998. ITT Corp. was bought by much smaller Starwood Hotels & Resorts, a real estate investment trust, completed in February 1998, following a takeover battle instigated by Hilton Hotels' unsolicited offer for ITT. Waste Management was taken over by USA Waste Services that completed the purchase in July 1998. Westinghouse was acquired in June 1998 by a joint venture of Morrison Knudsen and BNFL

The mergers and acquisitions of these companies from the worst board list may indicate that firms with very low quality boards cannot maintain convincing stock performance and may only be able to continue if they are acquired and reorganized.

SUMMARY AND CONCLUSIONS

This paper investigates the effect of board quality on the future performance of companies. The results show that companies with the best boards report higher cumulative stock return than for worst board companies in the period following the announcement of the board quality ranking. This finding is consistent with the argument that a good board of directors monitors the performance of management more effectively than a bad board of directors. Therefore, companies with better boards continue to outperform the companies with worse boards as reflected in stock return.

REFERENCES

Beatty, R. P. & E. J. Zajac (1994). Managerial incentives, monitoring, and risk bearing: a study of executive compensation, ownership, and board structure in initial public offerings, Administrative Science Quarterly, 39, 313-335.

Byrne, J. A. (1997). The Best and Worst Boards, Business Week, December 8, 90-98.

Fama, E. F. & M. C. Jensen (1983). Separation of ownership and control, Journal of Law and Economics, 301-325.

Fama, E. F. (1980). Agency problems and the theory of the firm, Journal of Political Economy, 88, 288-307.

Ip, G. & P. McGeehan (1998). Technology-blood letting raises specter of market revaluation. Wall Street Journal Interactive, August 31.

Jensen, M. C. & W. H. Meckling (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure, Journal of Financial Economics, 3, 305-360.

Morck, R., A. Shleifer, & R. W. Vishny (1988). Management ownership and market valuation: An empirical analysis, Journal of Financial Economics, 20, 293-315.

Shivdasani, A. (1993). Board composition, ownership structure, and hostile takeovers, Journal of Accounting and Economics, 16, 167-198.

Shleifer, A. & R. W. Vishny (1986). Large shareholders and corporate control, Journal of Political Economy, 94, 461-488.

Subrahmanyam, V., N. Rangan, & S. Rosenstein (1997). The role of outside directors in bank acquisitions, Financial Management, 26(3), 23-36.

Warner, J. B., R. L. Watts, & K. H. Wrunk (1988). Stock prices and top management changes, Journal of Financial Economics, 20, 461-492.

Webb, S. (1998). Global markets may face further storms ahead, Wall Street Journal Interactive, August 31.

Weisbach, M. S. (1988). Outside directors and CEO turnover, Journal of Financial Economics, 20, 431-460.

Weisbach, M. S. (1993). Corporate governance and hostile takeovers, Journal of Accounting and Economics, 16, 199-208.

Rick Stephan Hayes, California State University, Los Angeles

Dong-Woo Lee, California State University, Los Angeles Table 1 Financial Characteristics of Best and Worst Board Companies Panel A: Top 10 Best Board Companies Variables Mean Median Standard Deviation Total Assets (million $) 78,805 25,083 108,103 Return on Equity 28.9600 28.5500 23.2343 Debt to Equity Ratio 1.2260 0.8600 1.8754 Panel A: Top 10 Best Board Companies Variables Minimum Maximum Total Assets (million $) 6,051 318.882 Return on Equity -23.50 71.70 Debt to Equity Ratio 0.00 6.31 Panel B: Top 10 Worst Board Companies Variables Mean Median Standard Deviation Total Assets (million $) 15,042 6,930 18,862 Return on Equity 35.0500 9.9000 146.3489 Debt to Equity Ratio 17.8250 0.5750 53.3593 Panel B: Top 10 Worst Board Companies Variables Minimum Maximum Total Assets (million $) 397 57,447 Return on Equity -155.60 422.50 Debt to Equity Ratio 0.06 169.66 Panel C: Top 20 Best Board Companies Variables Mean Median Standard Deviation Total Assets (million $) 80,656 22,028 122,788 Return on Equity 37.4100 26.4000 46.7682 Debt to Equity Ratio 1.2295 0.4400 2.2299 Panel C: Top 20 Best Board Companies Variables Minimum Maximum Total Assets (million $) 3,861 420,076 Return on Equity -23.50 221.80 Debt to Equity Ratio 0.00 8.62 Panel D: Top 20 Worst Board Companies (*) Variables Mean Median Standard Deviation Total Assets (million $) 14,959 8,023 16,164 Return on Equity 20.8000 6.0000 111.2974 Debt to Equity Ratio 10.7812 0.6000 40.9502 Panel D: Top 20 Worst Board Companies (*) Variables Minimum Maximum Total Assets (million $) 164 57,447 Return on Equity -155.60 422.50 Debt to Equity Ratio 0.06 169.66 (*) The number of companies reported in this table is 17, because three companies were acquired by other companies in 1998. Definitions of Variables: Return on equity = The sum of net income for the most recent four quarters divided by the latest common equity. Debt Equity Ratio = Ratio of long-term debt to common equity at the end of the most recent quarter in 1998 Total Assets = Total assets at the end of the most recent quarter in 1998 Table 2 Stock Performance For the Period from January 1, 1998 to September 8, 1998 Panel A: Top 10 Best Board Companies Variables Mean Median Standard Deviation CR 18.5840 15.3900 21.4010 XMR 13.7240 10.5300 21.4010 XIR 2.3840 3.3550 17.7209 Panel A: Top 10 Best Board Companies Variables Minimum Maximum CR -8.44 57.79 XMR -13.30 52.93 XIR -28.61 29.65 Panel B: Top 10 Worst Board Companies Variables Mean Median Standard Deviation CR -7.0040 -8.7500 17.6765 XMR -11.8640 -13.6100 17.6765 XIR -5.5320 -7.2750 21.6279 Panel B: Top 10 Worst Board Companies Variables Minimum Maximum CR -31.77 32.38 XMR -36.63 27.52 XIR -45.42 33.66 Panel C: Top 20 Best Board Companies Variables Mean Median Standard Deviation CR 12.0660 13.9250 21.0333 XMR 7.2060 9.0650 21.0333 XIR -2.9945 1.1800 17.3851 Panel C: Top 20 Best Board Companies Variables Minimum Maximum CR -21.20 57.79 XMR -26.06 52.93 XIR -46.17 29.65 Panel D: Top 20 Worst Board Companies (*) Variables Mean Median Standard Deviation CR -1.2665 -8.040 20.5522 XMR -6.1265 -12.900 20.5522 XIR -5.5341 -6.450 19.4705 Panel D: Top 20 Worst Board Companies (*) Variables Minimum Maximum CR -31.77 40.41 XMR -36.63 35.55 XIR -45.42 33.66 (*) The number of companies reported in this table is 17, because three companies were acquired by other companies in 1998. Stock returns are reported in percent. Definitions of Variables: CR = Cumulative stock return for the period from January 1, 1998 to September 8, 1998. Cash distributions are considered reinvested as of the Ex-dividend date. XMR = CR--Market return for the same period XIR = CR--Industry return for the same period Table 3 t-tests for Equality of Means Comparison between the Best and Worst Board Companies Panel A: Comparison of Top 10 Best and Worst Board Companies Mean for Mean for Mean Top 10 Best Top 10 Worst Difference CR 18.584 -7.0040 25.5880 XMR 13.724 -11.8640 25.5880 XIR 2.384 -5.532 7.9160 P-value t-statistic (two-tailed) CR 2.915 0.009 XMR 2.915 0.009 XIR 0.895 0.383 Panel B: Comparison of Top 20 Best and Worst Board Companies (*1) Mean for Mean for Mean Top 20 Best Top 20 Worst Difference CR 12.0660 -1.2665 13.3325 XMR 7.2060 -6.1265 13.3325 XIR -2.9945 -5.5341 2.5396 P-value t-statistic (two-tailed) CR 1.942 0.060 XMR 1.942 0.060 XIR 0.419 0.678 Panel C: Comparison of Top 25 Best and Worst Board Companies (*2) Mean for Mean for Mean Top 25 Best Top 25 Worst Difference CR 5.8048 3.9167 1.8881 XMR 0.9448 -0.9433 1.8881 XIR -8.572 -1.4943 -7.0777 P-value t-statistic (two-tailed) CR 0.166 0.875 XMR 0.166 0.875 XIR -0.706 0.505 (*1) Top 20 worst board companies include 17 companies, because three companies were acquired by other companies in 1998 (*2) Top 25 worst board companies include 21 companies, because four companies were acquired by other companies in 1998. Stock returns are reported in percent. Definitions of Variables: CR = Cumulative stock return for the period from January 1, 1998 to September 8, 1998. Cash distributions are considered reinvested as of the Ex-dividend date. XMR = CR--Market return for the same period XIR = CR--Industry return for the same period Table 4 Stock Performance in 1998 by Sub-period Panel A: Percentage Change in Stock Price in 1998 Period Dow Jones Mean for Mean for Industrial Top 10 Best Top 10 Worst Average 1/1/98-8/28/98 1.8137 16.8507 -7.4259 8/31/98 -6.3665 -7.3984 -5.2188 9/1/98 3.8249 5.4128 4.1368 Period Mean for Mean for Top 20 Best Top 20 Worst 1/1/98-8/28/98 12.1291 -1.2952 8/31/98 -7.6786 -6.2553 9/1/98 4.7272 4.3664 Panel B: Comparison of Top 10 Best and Worst Board Companies Period Mean for Mean for Mean Top 10 Best Top 10 Worst Difference 1/1/98-8/28/98 16.8057 -7.4259 24.2765 8/31/98 -7.3984 -5.2188 -2.1796 9/1/98 5.4128 4.1368 1.276 Period t-statistic P-Value (two-tailed) 1/1/98-8/28/98 2.792 0.012 8/31/98 -1.520 0.146 9/1/98 0.851 0.406 Panel C: Comparison of Top 20 Best and Worst Board Companies (*) Period Mean for Mean for Mean Top 20 Best Top 20 Worst Difference 1/1/98-8/28/98 12.1291 -1.2952 13.4243 8/31/98 -7.6786 -6.2553 -1.4234 9/1/98 4.7272 4.3664 0.3608 Period t-statistic P-Value (two-tailed) 1/1/98-8/28/98 2.023 0.051 8/31/98 -1.138 0.265 9/1/98 0.297 0.768 (*) Top 20 worst board companies include 17 companies, because three companies were acquired by other companies in 1998.