PII: S1057-0810(99)80009-2 FINANCIAL SERVICES REVIEW, 7(1): 1-10 ISSN: 1057-0810 Copyright © 1998 by JAI Press Inc. All rights of reproduction in any form reserved. Shareholder Wealth Effects of CalPERS' Activism Claire E. Crutchley, Carl D. Hudson, and Marlin R.H. Jensen In the past decade, institutional investors have become more active in monitoring man- agement and voting the shares they control. The California Public Employees' Retirement System ( CaIPERS) was a leader in this wave of activism. This study investi- gates the long-term returns an investor with public information could earn by buying a portfolio of firms targeted by CaIPERS and whether the success of CaIPERS' activism depends on the aggressiveness of the targeting. The evidence supports the idea that vis- ible and aggressive activism leads to substantial increases in shareholder wealth while a quieter activism does not. I. INTRODUCTION The past several decades have witnessed a trend towards the institutionalization of the sav- ings process; more than half of all stocks outstanding are held through institutions rather than direct holdings of securities by individuals. The increased popularity of pension funds since World War II and the more recent development and expansion of Keogh plans and IRA accounts have resulted in the delegation of a large degree of corporate control to the managers of financial institutions. Prior to the 1980s such institutional investors were rel- atively passive and tended to vote shares in accordance with management's wishes. How- ever, as the holdings of institutional investors increased, so did the pressure to produce attractive returns. In an attempt to improve performance, institutional investors have become much more active in monitoring management and voting the shares they control. The California Public Employees' Retirement System (CalPERS) under the direction of CEO Dale Hanson was an early leader of this wave of activism. Each year since 1987, CalPERS has targeted a small group of firms in its portfolio that it perceives as having management problems. Articles in the Wall Street Journal have reported that investors could earn very high long-term returns by simply buying the firms that CalPERS targeted. However, in 1994 Hanson resigned, and was replaced by James Burton who continues to target finns, but whose activism is not as vocal as Hanson's. This study investigates the long-term returns that an investor with public information could earn by buying a portfolio Claire E. Crutehley, Carl D. Hudson, and Marlin R.H. Jensen ° 303 Lowder Business Building, Auburn University, Auburn, AL 36849, Fax: 3341844-4960; e-mail: claire@business.auburn.edu. 2 FINANCIAL SERVICES REVIEW 7(1 ) 1998 of firms targeted by CalPERS and whether the success of CalPERS' activism in the Hanson era differs from the post-Hanson era. II. BACKGROUND The history of CalPERS' activism can be separated into three stages; Dale Hanson, Cal- PERS' CEO from 1987-1994, spearheaded the first two. CalPERS began its aggressive shareholder activism campaign in 1987 when many boards of directors were enacting poi- son pills and staggered boards to avoid being bought out in the very active takeover market of the 1980s. The initial activism in the 1987-1990 time period was geared towards elimi- nating those poison pills and changing corporate governance structures (Nesbitt, 1994). For example, in 1987 CalPERS introduced anti-poison pill resolutions at AMR Corpora- tion and influenced management at Aluminum Company of America to withdraw their poi- son pill (Wall Street Journal [WSJ], Kilman, 1988). In 1990, CalPERS' focus changed and it began to target firms based on poor stock market performance. It examined firms in the bottom half of the Standard & Poors (S&P) 500 Index and hired consultants to prepare detailed reports that were used to select the finns to target. CalPERS first sent a letter to the targeted firms' board of directors that asked management to meet with CalPERS' staff and discuss problems with the company. In 1993 Bill Crist, the president of CalPERS' board, said, "Our objective is not to instill fear, but to encourage good performance" (The Washington Post, Vise, 1993). However, if management was not responsive to CalPERS' concerns, CalPERS took action. CalPERS introduced shareholder proposals in many firms, voted against the management slate of directors, and also retargeted firms the following year again asking for changes. For exam- ple, in 1994 CalPERS publicly announced its support for shareholder activist Robert A.G. Monks against the wishes of Sears' board of directors (Los Angeles Times, Silverstein, 1991). CalPERS also announced its intention to vote against the board of Phillip Morris when they would not meet with CalPERS' staff (WSJ, Hwang, 1995). In the early years, the fact that CalPERS targeted a firm would only become public when a shareholder resolution was proposed. However, in 1992 CalPERS began publicly announcing the list of the target firms. Hanson said, "a number of companies won't move unless they have to deal with [the problem] because it's in the public eye" (Business Week, Dobrzynski, 1992). Presently, CalPERS continues to focus on poor stock price performance and publicly announces the target firms. However, this stage is different due to the change in manage- ment styles; "To be less visible is not to be less effective," says current CEO, James Bur- ton. However, Robert Monks, a shareholder activist through Lens Corporation says, "What gave CalPERS the power was the personality of Dale Hanson" (Rehfeld, 1997). In 1995, the first year after Hanson resigned, Richard H. Koppes, General Counsel of CalPERS, announced the targeting of nine firms and promised to file shareholder proposals for com- panies who didn't make significant changes in response to CalPERS (Business Week, Schine, 1995). In 1996, Koppes left CalPERS and he now feels CalPERS is not continuing the public activism as it should; Koppes said, "Five years of going to the press, now they don't go" (Rehfeld, 1997). The size of the firms that are targeted also changed following Hanson's Shareholder Wealth Effects 3 departure. Unlike early years when very large firms were targeted, now mainly midsize companies are targeted. CalPERS' strategy changed from targeting firms such as IBM to firms such as Edison Brothers Stores and Venture Stores Incorporated. Rehfeld (1997) documents Charming Shoppes' experience with being targeted in 1996 during the post-Hanson era. The treasurer of Charming Shoppes, Bernard Brodsky, received CalPERS' letter and agreed to meet with CalPERS' representatives. In the meeting, he answered questions and documented changes being made. However, there was no follow-up to this meeting from CalPERS. After several months Brodsky called to find out whether CalPERS had more concerns and was told Channing Shoppes was fine even though over that time period they recorded losses and their stock price fluctuated greatly. There have been several empirical studies documenting the stock return effects of shareholder activism. Nesbitt (1994) documents huge gains to shareholders following the CalPERS' letter targeting a firm for poor performance. He does not find gains for firms tar- geted for corporate governance issues. These results are for companies targeted from 1990-1992, during Hanson's tenure as CEO of CalPERS. Smith (1996) documents significant positive two-day and long-term stock returns fol- lowing the announcement that CalPERS is targeting a firm for performance in 1989-1993. Wahal (1996) studies activism by several pension funds. On average, he finds neither long-term nor short-term abnormal stock returns following pension fund activism. How- ever, he finds evidence that when CalPERS pursues the activism, there are short-term (six-day) abnormal gains in the market following the day a letter is sent to management. Neither Smith nor Wahal find any evidence of improvement in accounting measures fol- lowing the activism. Strickland, Wiles, and Zenner (1996) find significant positive two-day stock returns to the announcement that the shareholder activist group United Shareholder Association (USA) negotiated an agreement with management. However, unlike the evidence on Cal- PERS' activism by Smith and Wahal, they do not find the significant returns at the announcement of the shareholder proposal. Therefore, the USA activism itself is not good news, only the successful resolution causes increased shareholder value. Opler and Sokobin (1995) study the effects of activism by the Council of Institutional Investors, and find that in the long-term there are significant gains to the shareholders of the targeted firms. The Council of Individual Investors is a group that was formed by CalPERS, but it has a low profile targeting strategy. Akhigbe, Madura, and Tucker (1997) study both insti- tutional and individual activism. They find positive long-term stock returns from activism, and find that activism instituted by individual investors leads to higher stock returns than proposals by institutions such as CalPERS. Overall, the evidence supports the idea that outsiders pursuing active monitoring can cause increases in stockholder returns. The evidence on CalPERS' activism is the stron- gest, showing positive short and long-term returns to shareholders after CalPERS' targets firms for poor performance. These studies all examine targets prior to the resignation of Dale Hanson. This paper will extend the previous papers by adding CalPERS' targets from 1994-1997. We also calculate returns following the public announcement date of the tar- gets to test whether investors with only public information could earn the high returns doc- umented in the previous studies. 4 F INANCIAL SERVICES R EVIEW 7(1) 1998 III. DATA AND METHODOLOGY The sample in this study consists of all public announcements of CalPERS' targets from 1992-1997. We collect data on stock returns for these companies, stock returns for the Standard & Poors (S&P) 500 index, evidence of significant changes in the firms following the targeting, the ownership structure and other attributes of the firm. The announcement date is the date CalPERS' target list was reported in the Wall Street Journal (WSJ) over a six-year period (WSJ, March 20, 1992; January 22, 1993; January 19, 1994; February 3, 1995; February 6, 1996; and February 11, 1997). The first date, March 20, 1992, is the first time CalPERS publicly announced its list of target companies. The initial sample contains the 63 firms that appear in the six WSJ articles (Table 1). Several firms have been targeted more than one year; this indicates that CalPERS was not satisfied with the firm's response to their initial letter. A study by Anand (1994) indicates that of the ten original targets in 1993, CalPERS only followed through with shareholder proposals on three of the firms by 1994. This indicates that seven of the ten targeted firms addressed CalPERS' concerns in some way. However, others were retargeted, such as IBM that was targeted in 1992, 1993 and 1994. This study includes only the first year in which a firm was targeted publicly, since investors could have different expectations following a second CalPERS' target. The final sample contains 47 targeted firms. TABLE 1 Firms Targeted By CalPERS as Reported in the Wall Street Journal 3/23/92 American Express Control Data Corporation Chrysler Corporation Dial Corporation Hercules Incl ITT Corporation l IBM Polaroid Ryder System Incorporated Salomon Incorporated Time-Warner USAir Group 2/03/95 Boise Cascade 2 First Mississippil Jostens Incorporated Kmart Corporation Melville Corporation Navistar International 1 Oryx Energy Corporation U.S. Shoe Company 1 Zurn Industries 1/22/93 Advanced Micro Devices Boise Cascade Champion International Chrysler Corporation 1 IBM l MacFrugals Bargains Pennzoil Polaroid 1 Sears Sizzler International Time-Warner 1 Westinghouse Electric 2/06/96 Applied Bioscience International Bassett Furniture Industries Charming Shoppes Inc Edison Brothers Stores Inc Melville Corporation1 Oryx Energy Corporation l Rollins Environmental Services Stride Rite Corporation U.S. Surgical Corporation Venture Stores Inc 1/19/94 Boise Cascade l CPI Corporation Eastman Kodak First Mississippi IBM 2 Navistar International USX Corporation U.S. Shoe Company Westinghouse Electric I Zenith Electronics Corporation 2/11/97 Apple Computer Bassett Furniture Industries l Fleming Cos Novell Inc Reebok International Ltd Rollins Environmental Services j Sensormatic Electronics Corp Stride Rite Corporation l Summit Technology Inc Sybase Inc Notes: I. CalPERS targeted this firm in the prior year. 2. CalPERS targeted this firm in the prior two years. Shareholder Wealth Effects 5 TABLE 2 Wall Street Journal Reports of Activities Associated with Targeted Firms in the Target Year Total Sample 1992-1994 1995-1997 Number of Firms Period Period Management Activities CEO Change 7 5 2 Top Manager Change 6 3 3 Restructuring 13 7 6 Divestiture 14 8 6 Debt Issue 4 4 0 Stock Issue 7 6 ! Cost Cutting 14 I 0 4 Executive Compensation Decrease 5 4 1 Executive Compensation Increase 3 3 0 Other Debt Downgrade 14 10 4 Debt Upgrade 8 8 0 The purpose of CalPERS' activism is to cause management of the targeted firms to change policies so that shareholders' (and CalPERS') wealth can be increased. The Wall Street Journal Index was examined for reports of significant structural changes in the tar- get firms in the year following the target announcement. Many firms reported major changes in operations (Table 2). Major changes include restructuring, divestitures, general cost cutting, and top management changes. Also, several firms issued either debt or equity during the target year. Managerial incentives to make significant changes and the resulting shareholder returns may be impacted by the structure of the firm' s equity ownership. Jensen and Meck- ling (1976) assert that firms with low insider ownership need more active monitoring. Firms with high insider ownership may already have the incentives to operate at a high level of efficiency. Consistent with this argument, Smith (1996) finds that CalPERS does not target firms with high insider ownership and low institutional ownership. Data on own- ership structure was collected from the Compact Disclosure data files; the ownership vari- ables collected are the percentage of stock held by insiders, institutional owners, and large (greater than five percent) owners. The success of shareholder activism may also be a function of the size of the firm; larger firms are less likely to experience a large gain, all else equal. The size of each firm is measured by the market value of its equity the year prior to targeting, which is calculated as closing price times number of shares outstanding as given by the Standard & Poors Compustat database. The measure of shareholder's excess return is the targeted company's holding period return less the stock return of the S&P 500 index. The cumulative excess performance is calculated as: n n CERin = l-I (1 + rit ) - l"l (1 +st ) t = l t = l 6 FINANCIAL SERVICES REVIEW 7(1) 1998 TABLE 3 Descriptive statistics of CalPERS targets [The mean and standard deviation (in parentheses) are listed for each variable] Total 1992-1994 1995-1997 Targeted Sample Time Period Time Period Variable N = 47 N = 27 N = 20 Market Size of the Firms (in millions of dollars) 3666.73 5474.13 a 1226.74 a (7987.91 ) ( 10,167.45) ( 1451.54) Percentage of the Stock Held by Institutional Investors 56.68 60.35 a 51.71 a (15.34) (11.33) (18.68) Percentage of the Stock Held by Insiders 5.73 6.18 5.13 (10.08) (12.38) (5.96) Percentage of the Stock Held by 5% Owners 24.69 23.18 26.73 (18.60) (20.46) (16.02) Note: aThe firms in the 1992-1994 time period are significantly different from the firms targeted in the 1995-1997 time period at the 0.05 level. where CERin is the n-day cumulative excess return for firm i, tit is the daily return for firm i, and s t is the daily return on the S&P 500 index. The cumulative return is calculated each month from the day following the announcement through the first year thereafter. Consis- tent with prior studies, we expect to see long term gains to stockholders' wealth following the public announcement of a firm being targeted by CalPERS. Cumulative monthly excess returns are examined the year following the announce- ment for the entire sample and for two sub-periods, 1992-1994 and 1995-1997. The later period corresponds to targets announced in the post-Hanson era. Using standard t-tests, we test whether the cumulative excess returns are greater than zero for each month, indicating whether a portfolio of CalPERS' targets would earn higher returns than the S&P 500 index. We also test whether the returns in the two sub-periods are statistically different. Other factors such as ownership structure, finn size, and the type of significant mana- gerial activities that were undertaken during the year after targeting are examined for effects on stockholder returns. Both firm size and the percentage of the stock held by insti- tutional investors are significantly different for the two periods (Table 3). Finally, the excess stock returns are regressed against a dummy variable for time period and the other variables. The regression results will indicate whether the relationship between the excess returns and the other variables is conditional on the time period. IV. RESULTS The mean cumulative excess return (CER) is reported monthly for the whole sample and for the two sub-periods (Table 4, Figure 1). For the total sample, the CER is positive for each month, which indicates that an investor holding a portfolio of the firms targeted by CalPERS would earn higher returns than holding the S&P 500 index. However, the excess returns are statistically different from zero only up to eight months following the announcement. For the entire year, the cumulative excess return of the portfolio of targeted firms is not different than the return of the S&P 500 index. Thus, the act of targeting has Shareholder Wealth Effects 7 T A B L E 4 C u m u l a t i v e E x c e s s R e t u r n s ( C E R ) o f C a l P E R S T a r g e t s a f te r the Wall Street Journal A n n o u n c e m e n t D a t e a Time Pehod (monks after ~e WSJ announcement Difference between CalPERS Targets time periods 1992-1994 1995-1997 Total Sample Time Period Time Period N = 47 N = 27 N = 20 Test CER CER CER Statistic 1 2.47% 0.95% 4.54% -0.77 2 6.98%* 3.57% 11.59%* -1.57 3 8.63%* 6.03%* 12.14% -0.79 4 11.05%* 8.03%* 15.18% -0.75 5 5.47% 4.76% 6.42% -0.18 6 5.80% 7.13%* 4.00% 0.38 7 8.27%* 10.41%* 5.37% 0.61 8 8.35%* 11.30%* 4.36% 0.90 9 4.09% 11.93%* -6.50% 2.24 b 10 1.82% 12.54%* -12.60% 2.91 b 11 0.65% 13.07%* -19.18%* 3.35 b 12 0.17% 14.79%* -19.56%* 3.4@ Notes: aThe cumulative excess return is calculated as the mean difference between the return for the CalPERS targets minus the return for the Standard & Poors 500 index, bThe cumulative excess returns for the firms in the 1992-1994 targeted group are significantly different from the 1995-1997 targeted group at the 0.05 level. *Significantly different from zero at the 0.05 level. 0.2 / 0.15 [ . . . . . . . . . . . . . . . . . . . . . . . . . . ;-~: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.t ................................................. / ; ................ ~ . ~ . ~ ' ~ , ~ ' , ~ , . . . . . . . . . . . . . . . . . . . . . . . : ---- * / N~ _ 4 %% F ~ . . . . ",,, " " " ~ ~ .~ .~. 0 1 2 3 4 5 6 7 8 '. 9 10 11 12 -0.05 : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0,15 ......................................................................... :.~ ............. -0.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . "~ L:."""" - -0.25 Figure 1. CalPERS Targeting: Excess Returns 8 FINANCIAL SERVICES REVIEW 7(1) 1998 some short run impact, but there is no lasting effect. There may, however, be differing effects depending on the year of the targeting. When the whole sample is split into sub-periods, two distinct patterns appear. For the 1992-1994 period, the cumulative return for the targeted firms is positive and significant. After twelve months the average return is 14.79% greater than that of the S&P which is sta- tistically different from zero with 95% confidence. For the 1995-1997 period, the perfor- mance of the targeted portfolio is only statistically above that of the S&P 500 index through month two, and by the end of 12 months, the performance on these later CaIPERS' targets is significantly below that of the S&P 500 index by 19.56%. Investors appear to find the targeting during this period to be a positive signal only in the short run. To determine whether the difference in the returns between the two sub-periods is due to differences in the firms' responses or differences in CalPERS's strategy, we regress the excess returns on ownership structure, firm size, actions taken following CaIPERS target- ing, and a dummy variable representing the time period in which the firm was targeted (Table 5). The only variable found to be significant is the time period dummy variable; firms which were targeted in the later period had significantly lower excess returns than the firms targeted in early periods. The specific actions taken by companies in the year after the target do not appear to have any effect on returns. A possible explanation is that different actions are needed in different firms, so restructuring is not superior to cost cutting; firms act in ways to enhance value. The difference between the results in the two different time periods was not caused by size differentials in the firms targeted, as size does not have a significant effect on stock returns. Another possible explanation for the significantly lower excess returns in the later time period is the performance of the overall stock market. If target firms have high returns, but the S&P 500 index had extraordinary returns, this would be shown as a negative excess return. However this is not the case. We examine the overall raw returns for the targeted TABLE 5 Cross-Sectional Regression Estimates of the Twelve-Month Cumulative Excess Returns on Various Independent Variables a (Test-statistics are in parentheses) Intercept INSIDER INST FIVE COST RESTRUCT SIZE TIME Parameter 0.31 - 0 . 7 6 0.03 0.58 0.19 - 0 . 1 9 - 0 . 3 6 -0.38*** Estimate (0.85) ( -1 .13) (0.10) (1.52) (0.93) ( -1 .43) ( -0 .78 ) ( -3 .45) Adjusted R 2 21 .16% F Value 2.76** Notes: aTbe cumulative excess return is calculated as the mean difference between the return for the CalPERS targets minus the return for the Standard & Poors 500 index. bTh¢ variable INSIDER represents the percentage of stock held by insiders in the firm as reported by Disclosure; INST represents the percentage of stock held by institutional investors as reported by Disclosure; FIVE represents the per- centage of stock held by five percent owners as reported by Disclosure; COST is a dummy variable representing 1 if the Fu'm has announced a cost-cutting measure in the Wall Street Journal during the year following the target announcement and 0 if not; RESTRUCT is a dummy variable representing I if the firm has announced a restructuring measure in the Wall Street Journal during the year following the target announcement and 0 if not; SIZE in the natural log of the stock price of the firm multiplied by the number of shares outstanding, and TIME is a dummy variable rep- resenting 1 if the firm was targeted in the 1992-1994 time period and 0 if not. ***The coefficient is significantly different from zero at the 0.01 level. ** The coefficient is significantly different from zero at the 0.05 level. Shareholder Wealth Effects 9 0.25 E | .! z= =E o 0.2 0.15 0.1 ¸ 0.05 -O.OS Figure 2. Months after targeting r loos- l~o7 Talced Firms - - - S & P 5oo [ CalPERS Targeting 1995-1997: Raw Returns vs. S&P firms and the S&P 500 index in the later time period (Figure 2). While the S&P 500 index was doing very well, the average target earned a higher return through eight months follow- ing the target. In the last four months of the year, all gains to the target f'maas were erased. V. SUMMARY AND CONCLUSION This paper examines whether investors can earn returns higher than a market index by pur- chasing a portfolio of t-n-ms after the announcement that CalPERS has targeted these firms. The evidence from this study indicates that the returns earned depend upon the era of Cal- PERS' activism. An investor following the strategy of buying the portfolio of first time tar- gets and holding these stocks for a year would have earned high returns in the 1992-1994 time period. However, an investor continuing this strategy in 1995-1997 would have earned much less than the S&P 500 after controlling for firm size, ownership structure, and the types of managerial actions taken. This later time period coincides with the era of less visible activism. A related question is whether pension fund investors benefit when managers are active monitors of firms in their portfolio. The evidence in this paper supports the idea that very visible and aggressive activism does cause substantial increases in shareholder wealth. However, a quieter activism does not yield the same results. And while CalPERS has become a quieter activist in the last few years, other pension funds and mutual funds con- tinue to aggressively monitor the firms in which they invest. In conclusion, our results indi- cate that unless management is pressured into making substantial changes investors will not benefit from shareholder activism. 10 FINANCIAL SERVICES REVIEW 7(1) 1998 REFERENCES Akhigbe, A., Madura, J., & Tucker, A. L. (1997). Long-term valuation effects of shareholder activ- ism. Applied Financial Economics, 7, 567-573. Anand, V. (1994, January 24). CalPERS gunning for poor performers, 3 companies being targeted by fund. Pensions & Investments, 22, 4, 80. Dobrzynski, J. H. (1992, March 30). Calpers is ready to roar, but will CEOs listen. Business Week, 44--45. Hwang, S. L. (1995, April 14). Calpers to vote against board of Philip Morris. The Wall Street Jour- nal, B4. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and capital structure. Journal of Financial Economics, 3, 305-360. Kilman, S. (1988, May 13). Holder resolutions against poison pills win more support at annual meet- ings. The Wall Street Journal, 4. Nesbitt, S. (1994). Long-term rewards from shareholder activism: A study of the CALPERS effect. Journal of Applied Corporate Finance, 6, 75-80. Opler, T. C., & Sokobin, J, (1995). Does coordinated institutional activism work? An analysis of the activities of the council of institutional investors. Working Paper, Ohio State University. Rehfeld, B. (1997). Low-cal CalPERS. Institutional Investor, 31, 41-49. Schine, E. (1995, February 13). This gadfly is really buzzing CalPERS is issuing fresh reprimands to laggard boards. Business Week, 48--49. Silverstein, S. (1991, April 18). CalPERS backs activist for seat on Sears board. Los Angeles Times, D2. Smith, M. P. (1996). Shareholder activism by institutional investors: Evidence from CalPERS. Jour- nal of Finance, 51, 227-252. Strickland, D., Wiles, K. W., & Zenner, M. (1996). A requiem for the USA is small shareholder mon- itoring effective. Journal of Financial Economics, 40, 319-338. Vise, D. A. (1993, January 28). New force in boardroom. The Washington Post, A1. Wahal, S. (1996). Public pension fund activism and firm performance. Journal of Financial and Quantitative Analysis, 31, 1-23.