25

JOURNAL OF ancial ECONOMICS ELSEVIER Journal of Financial Economics 42 (1996) 365 395

Corporate governance and shareholder initiatives: Empirical evidence

Jonathan M. Karpoff a, Paul H. Malatesta *'a, Ralph A. Walkling b

"School of Business Administration, University of Washington, Seattle, WA 98195, USA b Max M. Fisher College of Business, Ohio State UniversiO', Columbus, OH 43210, USA

(Received November 1994; final version received April 1996)

Abstract

Shareholder-initiated proxy proposals on corporate governance issues became popu- lar in the late 1980s as corporate takeover activity declined. We find firms attracting governance proposals have poor prior performance, as measured by the market- to-book ratio, operating return, and sales growth. There is little evidence that operating returns improve after proposals. The proposals also have negligible effects on company share values and top management turnover. Even proposals that receive a majority of shareholder votes typically do not engender share price increases or discernible changes in firm policies.

Key words: Shareholder initiatives; Proxy; Corporate governance; Performance J E L classification: G34; D23

1. Introduction

Since the middle 1980s, hundreds of publicly traded companies have received shareholder-initiated proxy proposals on corporate governance issues. These

*Corresponding author.

We appreciate helpful comments from seminar participants at Arizona State University, Ohio State University, University of Oregon, and UCLA, and also Jeff Coles, Mark Grinblatt, Tim Opler, Mitchell Peterson, Rick Smith, J. Fred Weston, Joe Williams, an anonymous referee, and Wayne Mikkelson. We thank Kevin Harper, Charles Appeadu, Heber Farnsworth, Bong-Chan Kho, Heon Jeong, Kathy Kahle, and Ami Scott for research assistance, and the Center for the Study of Financial Management and the Dice Center for Financial Economics for financial support. A previous version of this paper was presented at the 1996 American Finance Association Meetings. Peg O'Hara and The Investor Responsibility Research Center, Inc. graciously provided us with data on corporate governance proxy proposals.

0304-405X/96/$15.00 ~i~ 1996 Elsevier Science S.A. All rights reserved PII $ 0 3 0 4 - 4 0 5 X ( 9 6 ) 0 0 8 8 3- 5

366 J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365-395

proposals call for firms to rescind poison pills, reject paying greenmail, require confidential voting, or make other changes in the rules that govern relations between shareholders, directors, and top-level managers. We present empirical evidence bearing on two questions raised by the frequent use of such proposals. First, what types of companies attract such proposals? And second, how do the proposals affect firm values and operating performance?

A central tenet of shareholder activism holds that shareholder proposals ameliorate the shareholder-manager agency conflict and pressure managers to adopt value-increasing policies. Ryan (1988), Gordon (1993), and Pozen (1994), for example, argue that shareholder proposals can encourage managers to make specific value-enhancing changes. Others, including Black (1992) and Pound (1992), argue that shareholder initiatives, presumably including corporate governance proposals, offer a way to influence corporate policy with less rancor and fewer costs than are typically associated with takeover bids and proxy contests. Supporting these views is evidence that some forms of shareholder activism are associated with firm value increases. Wahal (1996), for example, examines 43 instances in which activist institutional shareholders pressured firms to make organizational changes, without resorting to the use of proxy proposals. He finds that the earliest announcements regarding such pressure are associated with positive average abnormal returns to the targets' stock. Similarly, Strick- land, Wiles, and Zenner (1996) find a positive average stock price reaction for 53 negotiated agreements between targeted firms and United Shareholders Associ- ation representatives.

Few of the purported benefits of shareholder activism, however, are evident in our 1986-1990 sample of shareholder proposals. We find that, measured by the market-to-book ratio, operating return on sales, and sales growth rate, firms attracting proposals are poor performers. These results show that proposal sponsors have reason to seek improvements in their target firms. Overall, however, we find little evidence that proposals engender performance improvements. Moreover, news about shareholder proposals generally is associated with insignificant changes in share values, and the proposals do not appear to affect top management turnover rates. Even among proposals that receive a majority of shareholder votes, we find little evidence of value increases or policy changes attributable to the proposals.

Our results are consistent with those of other researchers. John and Klein (1995) find that the probability of receiving a shareholder proposal increases if the firm reports negative income. Wahal (1996), Gillan and Starks (1995), and Strickland, Wiles, and Zenner (1996) report negligible average abnormal stock returns associated with announcements about shareholder proposals. We conclude that poor financial performance attracts shareholder corporate governance proposals. There is no persuasive evidence, however, that the proposals have substantial impact on firm values or performance.

J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365 395 367

Our paper proceeds as follows. Section 2 describes the developments that have led to the widespread use of corporate governance shareholder initiatives. Section 3 depicts four competing views about initiatives' effects on firm value and operating performance. We describe the data used to test these views in section 4. Sections 5-8 report our empirical results: Section 5 reports on the characteristics of firms attracting proposals. Section 6 reports the proposals ' stock price effects. Section 7 examines in detail the effects of proposals that gain a majority of votes. And Section 8 reports on the proposals ' effects on operating performance. Section 9 concludes.

2. Historical developments affecting shareholder activism

Pound (1992) argues that the recent trend toward shareholder activism is a logical response to historical developments in the corporate control market. In 1982, the United States Supreme Court greatly altered the structure of U.S. takeover laws prevailing at the time. The Court 's decision in Edgar v. M I T E

Corp. (457 U.S, 624) effectively invalidated the restrictive takeover laws of 37 states. Stripped of their statutory protection against hostile outside takeovers, many managers sought new takeover defenses to secure control of their firms. These new defenses included antitakeover charter amendments, poison pills, and revised state antitakeover laws. 1 Jarrell and Poulsen (1987) report that, during the first year after M I T E , 206 firms adopted antitakeover charter amendments such as supermajority clauses, fair price clauses, and classified board schemes. Only 22 firms adopted such defenses the year before M I T E . The first poison pills and the first of the new generation of state anti takeover laws were written immediately after the M I T E decision in 1982. The adoption of poison pills and state antitakeover laws accelerated after important court decisions upheld their uses as takeover defenses. 2 Danielson, Karpoff, and Marr (1995) report that by 1988 approximately half of the S&P 500 firms had poison pills, and well over half were covered by at least one p o s t - M I T E state antitakeover law.

Coffee (1991) and Pound (1992) note that by the late 1980s, charter amendments, poison pills, and restrictive state laws together posed formidable obstacles to hostile takeover activity. Though Comment and Schwert (1995)

1See DeAngelo and Rice (1983), Linn and McConnell (1983), Jarrell and Poulsen (1987), Gilson (1986), and Weston, Chung, and Hoag (1990) for descriptions and analyses of antitakeover charter provisions. Malatesta and Walkling (1988) and Ryngaert (1988) provide a detailed description and empirical analysis of poison pill takeover defenses. See Karpoff and Malatesta (1989) for a study of post-MITE state antitakeover legislation during the mid-1980s. 2The Delaware Supreme Court upheld the use of a poison pill takeover defense in Moran v. Household International (500 A. 2d 1346, Del. 1985). The U.S. Supreme Court upheld Indiana's control share acquisition law in CTS v. Dynamics Corp. of America (107 S. Ct. 1673, 1987).

368 J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365 395

argue that poison pills and state freeze-out laws had little impact on takeover frequency, nevertheless, hostile tender offer activity virtually had ceased by 1990. 3 This implies that some existing capital market mechanisms for replacing firm managers had almost stopped functioning. Gordon and Pound (1993), Romano (1993), and Pozen (1994) argue that as a consequence, many investors who sought to affect firm policies, including policies relating to takeovers, resorted to shareholder initiatives and proxy fights with incumbent managers. The increased frequency of shareholder proposals also coincides with an increase in institutional investors' attempts to influence corporate policies and governance. Bergin (1988) shows that institutional investors were generally passive before 1987, and routinely voted their proxies with management. Re- flecting institutions' increasing activism, Biersach (1990) reports that, by 1990, 90% of public employee pension funds and 83% of private pension funds supported shareholder-initiated antigreenmail resolutions.

In several cases, dissident investors have successfully used proxy voting to change firm policy. In 1989, for example, shareholders at Honeywell, led by the California Public Employees Retirement System (CalPERS), the Pennsylvania Public School Employees' Retirement System, and North American Partners (a private investment group), voted down two antitakeover initiatives proposed by management. Galant (1990) reports that two months later, the company yielded to continued stockholder pressure and announced plans to restructure. In another case, Pozen (1994) credits shareholder initiatives with a key role in Sears, Roebuck, and Company's decision to appoint new outside directors and reverse its unprofitable diversification strategy. Bergin (1988) reports that other firms for which shareholder proposals preceded changes in company policy include Gillette, General Mills, Sara Lee, and Minnesota Mining & Manufac- turing.

In the majority of cases, however, shareholder-initiated proxy proposals are defeated. During our 1986-1990 sample period, only 22 of 866 proposals received more than 50% of the votes cast - a success rate of 2.5%. Moreover, SEC Rule 14a-8 places severe limits on the force and scope of such proposals. As a result, directors and managers usually are not compelled to implement a proposal, even if it gains a majority of votes.

3. Competing views about shareholder proposals

Evidence reported by Morck, Shleifer, and Vishny (1989) indicates that external bids for control are more likely at firms with poor performance records. Hasbrouck (1985) and Lang, Stulz, and Walkling (1989) find that Tobin's

3See Mergerstat Review (1993, p. 82).

J.M. Kurpo~' et aL/Journal of Financial Economics 42 (I 996) 365 395 369

q ratios of tender offer targets are relatively low and decline over the five years before the tender offer. In addition, Palepu (1986) interprets his evidence as consistent with the hypothesis that takeovers occur when incumbent managers have performed poorly. Mulherin and Poulsen (1994) report that firms subject to dissident proxy challenges also show poor prior performance. If shareholder- initiated corporate governance proposals are made in lieu of, or to facilitate, control contests, firms that receive corporate governance proposals should be characterized by poor prior performance, too.

Proposals might be motivated by poor performance, but there are different views about the proposals' effects. Black (1990, 1992), Nesbitt (1992, 1994), and Pozen (1994) predict that shareholder activities, such as corporate governance proposals, have positive valuation effects. The proposals can precipitate specific actions, such as rescinding antitakeover measures, that potentially increase stock values. John and Klein (1995) argue that shareholder proposals can prod managers to improve the firm's operations because they are a form of '... shareholder participation which is not dominated by management'. This view is echoed by Pozen, as quoted by Dobrzynski (1990): 'Even if these things don't win, management gets the message. Management will have a more focused view now. And that may be the right result'. Shareholder proposals also can increase values if they complement broader efforts to gain control of a corporation or to exert pressure for specific policy changes. For example, shareholder proposals have been introduced at several companies, including Avon Products, Inc., Gillette Co., Lockheed Corporation, and USX, Inc., to facilitate hostile takeover bids.

A competing view holds that shareholder proposals are unlikely to cause significant organizational or policy changes. Hence, the proposals should have negligible effects on firm performance and value. This view is reflected in Jensen's (1993) observation that internal control systems, which include shareholder-initiated proposals, of publicly held corporations often fail to make managers maximize value. This view also is suggested by evidence that some shareholder proposals seek to repeal corporate charter provisions that have negligible valuation effects in the first place. DeAngelo and Rice (1983) and Linn and McConnell (1983), for example, report that upon announcement of such charter amendments as supermajority voting rules and classified boards, the average abnormal returns are insignificantly different from zero.

Another view holds that shareholder activism tends to impair firm management, degrade performance, and decrease firm value. Wohlstetter (1993), for example, claims that most shareholder activists and pension fund managers have neither the skills nor the experience to improve on managers' decisions. Activists' attempts to influence corporate decisions therefore tend to disrupt the firm's operations. Lipton and Rosenblum (1991) argue that even well-inten- tioned initiatives can distract managers and harm their abilities to manage effectively.

370 J.M. Karpoff et aL /Journal of Financial Economics 42 (1996) 365-395

Reflecting yet another view, Romano (1993) and Saxton (1994) observe that public institutional investors ultimately are subject to political control and sometimes pursue objectives other than value maximization. It is possible that public institutions use corporate governance proposals to gain influence over the target firms' decisions, and to induce the firms to pursue politically motivated, value-decreasing investments. If this view is correct, corporate governance proposals sponsored specifically by public institutions will tend to decrease firm values and operating performance.

4. Data

We test the four competing views about shareholder proposals by examining the characteristics of firms attracting proposals and the proposals' effects on firm values and operating performance. Data were obtained from the Investor Responsibility Research Center, Inc. (IRRC) on corporate governance proposals made during the 1987-1990 proxy seasons, which cover the period from March 1986 through October 1990. During this period, 866 proposals were recorded at 317 companies. We reviewed the appropriate proxy statement for each proposal to confirm its nature and to identify the dates when the proxy materials were sent to shareholders and received by the Securities and Exchange Commission. We obtained the information for proposals from 1988 90 from the Lexis/Nexis database. For proposals from 1986-87 we obtained information from the Q-Data Corp. microfiche records of the proxy materials.

To examine the characteristics of firms that attract shareholder proposals, we use financial and ownership data from Compustat and Value Line. Sufficient data to conduct these analyses are available for 269 companies, representing 522 shareholder proposal events. We refer to this as our primary sample.

To examine the valuation effects of the proposals we use data from the Center for Research in Securities Prices (CRSP) daily stock returns files for firms traded on the New York and American Stock Exchanges and Nasdaq. Sufficient data on 290 firms, representing 583 shareholder proposal events, are available. We refer to this larger group as the event study sample.

Many proposal events represent more than a single proposal. In our primary sample, 139 of the 522 proposal events represent situations in which more than one proposal was made. In 1989, for example, USAir Group, Inc. received a proposal to redeem its poison pill and a second proposal to require confidential voting on proxy matters. Since both proposals were presented in the same proxy materials, both are treated in our empirical tests as a single proposal event.

The number of proposal events increases in each year of the sample. Only six of the 522 events in our primary sample occurred in 1986 (representing several early proposals in the 1987 proxy season). By 1990, the number had risen to 181.

J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365-395 371

Many companies appear in our sample in more than one year. Of the 269 firms in our primary sample, 118 appear just once, 67 twice, 66 three times, and 18 four times. The sample of proposal events is distributed widely across industries, as measured by two-digit CRSP SIC codes. The industries with the greatest representation are chemicals (SIC = 28); transportation equipment (SIC = 37); electric, gas, and sanitary services (SIC = 49); and holding and investment companies (SIC = 67). In no case does a single industry represent more than 10% of the proposal events.

Table 1 presents the primary sample according to the type of issue raised and the sponsor's identity. There are five categories of proposal issues:

External corporate control market issues directly affect an outside bidder's ability to acquire effective control of a firm through the acquisition of shares. The 75 proposal events in this category include resolutions to rescind a poison pill, put a poison pill to shareholder vote, ban greenmail payments, and opt out of a state antitakeover law.

In the second category are 225 proposals that affect shareholders' ability to influence corporate policies through share voting. We call these internal corporate governance proposals. The most prevalent proposals of this type call for confidential or cumulative voting, or declassifying the board.

In the third category are 42 proposal events involving compensation of senior officers and directors. The majority call for limits on executive pay or require- ments that directors own company stock.

The fourth category, with 66 events, consists of other miscellaneous corporate governance issues. The miscellaneous issues primarily include proposals to involve shareholders more directly in the monitoring or appointment of directors, and proposals for the firm to conduct specific business transactions. As extreme examples, two proposals called for managers to sell the company or negotiate an acquisition agreement.

The final category, labelled 'Mixed issues', consists of events in which a firm receives in the same proxy season multiple proposals that fall into more than one of the other four categories. There are 114 proposal events in this group.

Table 1 also groups proposals by sponsor. Seventy-six proposal events involved exclusively public institution sponsors, most often CalPERS (22 events), the California State Teachers Retirement System (18 events), and the New York City Employees' Retirement System (11 events). Private institutions, including the College Retirement and Equities Fund and several unions, sponsored 48 proposal events. Most events (341), however, were sponsored by individuals. Some of the individual sponsors, including Evelyn Davis, John J. Gilbert, and Lewis D. Gilbert, are well-known for their activism at numerous companies. In 43 events the proposals are sponsored by more than one category of sponsor, and in 14 events the sponsor is not identified by IRRC.

Public and private institutions almost exclusively sponsor resolutions that affect external control or internal governance issues, but do not sponsor any of

372 J.M. Kurpoff et al./Journal of Financial Economics 42 (1996) 365 395

Table 1 Proposal sample partitioned by issue and sponsor type

Distribution of shareholder proxy proposal events from 1986-90 by issue and sponsor type. Firms receiving proposals are identified by the Investor Responsibility Research Center, Inc. The primary sample consists of 522 proposal events at 269 companies that have financial and ownership data available from Compusta t and Value Line, respectively. Multiple proposals in the same year for a given firm are treated as a single-proposal event.

Sponsor type

Public Private Individual Total by Issue type institution ~ institution f investor Mixed g Unknown issue type

External corporate 33 22 18 2 0 75 control market issues ~

Internal corporate 34 23 163 5 0 225 governance issues b

Compensation-related 0 0 41 0 1 42 issues

Other miscellaneous 0 0 57 0 9 66 i s s u e s e

Mixed issues a 9 3 62 36 4 114

Total by sponsor type 76 48 341 43 14 522

aproposals to put to shareholder vote or repeal a poison pill, put to shareholder vote or terminate a standstill agreement, opt out of a state antitakeover law, eliminate a fair price provision or reduce a fair price supermajority level, ban greenmail, repeal or require shareholder approval for antitakeover devices, reduce the vote required for shareholder action by written consent, lower a supermajority level required for merger agreements, or require a shareholder vote on a targeted share placement.

bProposals to require confidential voting, require cumulative voting, eliminate a classified board, provide shareholders equal access to proxy materials, restore shareholders" right to call meetings or propose charter amendments , or require one share, one vote rules.

CMiscellaneous proposals include those calling for shareholder ratification of auditors, changes in shareholder meeting times or locations, establishing a shareholder advisory or monitor ing commit- tee, and limits on outside directors' terms.

dProposals from more than one of the other issue type categories.

~Includes the California Public Employees Retirement System, California State Teachers Retirement System, New York City Employees Retirement System, New York City Fire, Police, or Teachers pension funds, and others.

flncludes the College Retirement Equities Fund, IBEW pension fund, United Brotherhood of Carpenters pension fund, and others.

gProposal events that include proposals from more than one of the other sponsor type categories.

J.M. Karpoff et al./Journal ~?f Financial Economics 42 (1996) 365 395 373

the events regarding compensation or miscellaneous issues, individuals, in contrast, sponsor relatively few of the external control resolutions, but most of the proposals on internal corporate governance, compensation, and miscellaneous issues.

5. Characteristics of firms receiving shareholder proposals

This section tests whether sponsors of shareholder proposals target poorly performing firms. For each event in the sample, we use the CRSP daily master file to identify the firm's two-digit SIC and its market capitalization at the beginning of the calendar year of the proposal. For each event, we identify a control firm from the same two-digit industry. We choose a control firm with market capitalization closest to that of the sample firm at the beginning of the proposal year from among firms that receive no corporate governance shareholder proposals during the sample period. When a firm has shareholder proposals in more than one year, we identify control firms separately for each year.

We examine four different indicators of financial performance, each measured at the beginning of the proposal year: the market-to-book ratio; operating return on sales, measured as earnings before interest, taxes, depreciation, and amortization divided by sales; recent sales growth rate, measured by the com- pound annual rate of change in sales over the three years preceding the proposal year; and the prior three-year cumulative abnormal stock return measured over the 750 trading days ending on the day before the proxy mailing date. To calculate abnormal stock returns, we employ the single-factor market model, using the equally-weighted CRSP index and returns expressed as continuously compounded rates. ~ We estimate the market model parameters using 200 returns from 950 through 751 days before the proxy mailing date. All data are obtained from the Compustat and CRSP files.

The four performance measures are not highly correlated, and appear to reflect different aspects of firm performance. Most pairwise correlation coefficients between the performance variables are below 0.10 in absolute value; the highest (between the market-to-book ratio and operating return on sales) is 0.17.

In investigating differences between the proposal and control firms, we account for differences in size, leverage, and ownership structure. Size is measured as the market value of equity. Leverage is the ratio of the book value long-term debt to the sum of the book values of preferred and common equity. We use Compustat data to compute both variables. Insider ownership data for both the proposal and control firms are collected from the last Value Line issue

4See Fama (1976, Chs. 3 and 4) for a discussion of the market model.

374 J.M. Karpofl' et al./Journal o f Financial Economics 42 (1996) 365- 395

that reports an analysis of the firm before the proposal yea r ) We sum the ownership of directors, officers, and insiders to obtain our measure of insider shareholdings. Data on institutional ownership are collected from Standard & Poor 's Stock Guide for the year end before the proposal year.

5. I. Univariate comparisons of proposal and control firms

Table 2 reports summary statistics on the performance and control variables for the proposal and control firms. To avoid distortions caused by very small denominators in financial ratios, we delete outliers in which the market - to-book ratio exceeds 60 or the ratio of long-term debt to preferred and common equity exceeds 25.

The results in panel A indicate that compared to the control firms, the proposal firms have relatively low market- to-book ratios, operating returns on sales, recent sales growth, and prior cumulative abnormal stock returns. For example, the mean operating return on sales is 16.8% for the proposal firms and 21.3% for the control firms. The difference of 4.5% is statistically significant at the 1% level.

We compare the control variables in panel B. Although we selected the control firms to match the proposal firms' size, they nonetheless are smaller in 65.3% of the cases. This is because in some industries the proposal firm is either the largest firm or there is no close size match. The results indicate that proposal firms have more leverage, greater institutional ownership, and less insider ownership than their control counterparts. The ownership differences could reflect the proposal firms' larger size. Previous research reports that large firms tend to have high levels of institutional shareholdings and low levels of managerial ownership. 6

To examine whether the performance comparisons reflect size differences between the proposal and control firms, we repeated the tests reported in panel A, Table 2, after restricting the sample to cases in which the ratio of the proposal firm size to control firm size is between 0.75 and 1.25. The results are qualitatively identical to those in Table 2.

The results reported in this paper are insensitive to the specific definitions of the four performance measures. For example, using operating return on assets instead of sales yields similar results. We also performed comparison tests on

5The Value Line data are computed from information contained in proxy statements, public disclosures and SEC Forms 3 and 4 (on insider trading). There is some dispute over the extent to which the Value Line data replicate ownership data disclosed in proxy statements. Anderson and Lee (1996) report that Value Line ownership data are not highly correlated with ownership data in proxy statements, while Kole (1995) reports that the correlation is high. 6See Demsetz and Lehn (1985), Mikkelson and Partch (1989), and Song and Walkling (1993).

J.M. Karpoff el al./Journal of Financial Economics 42 (1996) 365-395 375

subgroups created by partitioning the total sample according to proposal and sponsor types. The results indicate that most of the patterns reflected in Table 2 also appear in each subsample. (We note, however, that there are some excep- tions to this statement. In comparing the firms' previous stock returns, one-year cumulative abnormal returns are not significantly lower for the proposal firms than for the control firms. Also, several deviations from the overall patterns appear when miscellaneous issue proposals are isolated.)

5.2. Multivariate logistic regressions

We estimate logistic regressions to investigate the effects of firm characteristics on the probability of a firm receiving a corporate governance shareholder proposal. The dependent variable in these regressions has a value of one for the proposal firms and zero for the control firms. The independent variables include performance measures and measures of firm size, institutional and insider ownership, and leverage. Firm size is measured as the natural log of the market value of equity. Firm size is highly collinear with the number of institutional shareholders (the correlation coefficient is 0.73), so we estimate separate regressions using each variable. There are 1044 total cases (522 proposal events and 522 matched controls), but in only 709 cases do we have complete data on all variables. The results reported in Table 3 are based on these 709 cases. The evidence is qualitatively identical when we estimate each regression by using the larger sample that is available when we require sufficient data only for the particular test.

Each of the first four regressions in Table 3 estimates the effect of a different performance measure on the probability of receiving a shareholder proposal. The market-to-book ratio, operating return on sales, and recent sales growth all are negatively and significantly related to the likelihood of receiving a proposal. The coefficient for the prior three years' cumulative abnormal stock return, however, is not statistically significant. In most cases, the evidence on the control variables is consistent with the inferences derived from the univariate comparisons. For example, the probability of receiving a proposal is positively related to firm size, the number of institutional investors, leverage, and the percentage of shares owned by institutional investors. Unlike the univariate results, however, in the logistic regressions the probability is not significantly related to the percentage of shares owned by insiders. Regressions 5 and 6 report results when all performance variables are included simultaneously. In regression 5 we include firm size as a control variable, and in regression 6 we include the number of institutions. The results are similar to those of the previous four regressions, except that the coefficients for the market-to-book ratio are smaller and less significant.

We estimated regressions using only subsets of the control variables. We also varied the performance measure, alternating our use of prior market-adjusted

376 J.M. Karpoff et al./"Journal of Financial Economics 42 (1996) 365 395

~ . ~ ~ ~ o

. .~ . ~ o "~

.~ .~ ~ .~ ~- ~ ~

~- ~ ~ ~

g N N ~ ~

~ . ~

• -~ ~_ ~ ~ = ~

~ i e ~" ~ ~

8~A~

z ~

q ~

I I I I I I F I

I

i I v

o o

,-&

I -

0 ~

J.M. Karpo~]' et a l . / Jourml l q l Financia l Economics 42 (1996) 365 395 377

E c'i c.~ ~.~

,,~ r.i ~, ~ ~ t-i E ~ - -

©

r~ c

z ~ Z

G

y.

,i E

Z

?,

~J

2

378 J.M. Karpoff et al./'Journal of Financial Economics 42 (1996) 365 395

=

= ~ ~ ' - ~ o

"~ ~ .~ ~

.~ ~ ~ . ~ .2

o ' ~ "~"~

= '-' = u ' ~ ~ .=_

= ~ ~ "£ ~

• ~ ~ ~ ~ ~ , .~

o ~.o ' , S ~ 5 . - " , - ~ '-- o

= 0 ~ ~ r " ,!:~ .

= .~ ~ , ~ ~ ~ ,_. "~ .~

'- ° ° ~ ' - ~i ~ i . Z ~-' #, ~., " = o ..=

2

~3 ,;b =

#,

,..0

,~ ' r !

I I I I I I I I I I I I

oc ~..~ ~ b.. ~ ~ ~,'-~ ¢'N ~ o,, , 430o ~1 oo o'~ o ~, ~ "~ £.4 i '~ t ' q "~ ~"~ ~ ,.-;,",4 ~ -4 ,--: ~ ,~ ,,-q - ; , - 4 o .<

o - , & ~ - ~ - ~ - - S o _ ~ i~- o ', ~ ,"h, ~o I~. t th I ~ tt.~ ,~% t-..i ' , o ~ , - 4 ~ ,""4 ~r - ,4 r , , i ~ ,'-4 .".,4 ~ q 5

I I I I I I I I I I I I

. t .c

r4 ,'4

i ,20

q~

I I I I

J

I I I I

I : I I

,.5

J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365-395 379

stock returns, operating return on assets, and return on book equity. In these sensitivity tests, the probability of receiving a shareholder proposal is negatively related to the market - to-book ratio, operating return on sales or assets, return on equity, and recent sales growth, and is not significantly related to prior stock returns. We also estimated logistic regressions for subgroups of firms partitioned by issue type and sponsor type. The results are similar to those reported in Table 3. As with the univariate comparisons, the only subgroup with notably different results from the group as a whole consists of firms receiving proposals classified as dealing with miscellaneous issues. Unlike the results for the overall sample, the probability of receiving a proposal on a miscellaneous issue is not significantly related to recent sales growth or to the percent of shares owned by institutions.

We analyze the sensitivity of the probability of receiving a shareholder proposal to changes in the market- to-book ratio, operating return on sales, and recent sales growth. In each comparison, all regressors are set at their mean values and the performance variable is perturbed by one standard deviation. Using coefficients from the first regression in Table 3, a one-standard-deviation increase in the market - to-book ratio decreases the probability of attracting a proposal from 54.2% to 48.9%. The influence of operating return on sales is similar: Using coefficients from regression 2, a one-standard-deviation increase in the operating return on sales corresponds to a decrease in proposal probability from 57.4% to 48.1%. Using regression 3 estimates, a one-standard-deviation increase in the recent sales growth rate decreases proposal probability fl-om 55.3% to 44.4%."

Our results are consistent with some of those reported by John and Klein (1995), who examine shareholder proposals made in the 199142 proxy season at S&P 500 companies. John and Klein report that in their sample, the probability of receiving a corporate governance proposal increases with firm size and the report of negative net income in the prior year. Unlike us, however, John and Klein report a negative relation between institutional ownership and the likelihood of receiving a proposal.

Pound (1992), Romano (1993), and others argue that shareholder proposals are made in lieu of external control contests. Nonetheless, firms that receive shareholder-initiated corporate governance proposals differ in some respects from firms receiving takeover bids. We find a positive relation between the probability of receiving a proposal and firm market capitalization, whereas Mikkelson and Partch (1989) and Song and Walkling (1993) find that the

7The probabilities of attracting a proposal when all regressors are evaluated at their means approximately equals the fraction of proposal firms in the Table 6 regressions. Of the 709 observations used to calculate the coefficients, 391, or 55%, represent proposal firms. Evaluated at the mean regressor values, the estimated proposal probabilities differ slightly from 55% because the logistic ['unction is nonlinear.

380 J.M. KarpolJ et al./Journal of Financial Economics 42 (1996) 365 395

likelihood of becoming a takeover target is negatively related to size. Our results indicate that the probability of receiving a shareholder proposal is negatively, but insignificantly, related to the percentage of shares held by insiders. Mikkel- son and Partch (1989) and Song and Walkling (1993) find that the probability of receiving a takeover bid is negatively related to insider ownership, although Ambrose and Megginson (1992) report that the relation is not statistically significant.

Overall, our results indicate that the probability of attracting a corporate governance proposal increases with firm size, leverage, and institutional shareholdings. The probability decreases with the market-to-book ratio, operating return on sales, and recent sales growth, indicating that corporate governance proposals are made at poorly performing firms.

6. Wealth effects of shareholder corporate governance proposals

To test hypotheses about shareholder proposals' wealth effects, we perform an event study. To calculate abnormal returns, we employ a Dimson (1979) type market model with one lagged and one leading market index return in addition to the contemporaneous market return. (Results are similar using a model with only the contemporaneous return.) Our empirical procedure treats explicitly the serial dependence in forecast errors that arises from estimation error in the market model parameters, as discussed by Mikkelson and Partch (1988) and Salinger (1992) for single-factor models, s

Our tests focus on three different event dates: the date The Wall Street Journal first published an article referring to the proposal, the date the firm mailed proxy materials to shareholders that included a discussion of the proposal, and the date of the shareholder meeting.

6.1, Abnormal returns around WSJ announcements'

We discovered articles in The Wall Street Journal for only 27 events in our sample, involving 24 companies. That so few proposals are reported suggests that The Wall Street Journal's editors do not consider proposals especially newsworthy. Consistent with this judgment, the average abnormal returns, reported in Table 4, are small. Event days - 1, 0, and 1 denote the day before, the day of, and the day after The Wall Street Journal article. The average abnormal returns on these days do not differ significantly from zero. There is

SA technical appendix containing details of our empirical procedure for multi-factor models is available upon request.

Table 4 Cumula t i ve average a b n o r m a l s tock re turns a r o u n d key dates for shareho lder p roposa l s

Cumula t i ve average a b n o r m a l dai ly s tock re turns for firms exper iencing shareho lder in i t ia ted corpora te governance p roxy p roposa l events from 1986 t h r o u g h 1990. C o l u m n 2 repor ts resul ts for 27 events at 24 firms, measu r ing t ime from the da te on which The Wall Street Journal first pub l i shed an art icle referring to the p roposa l event. C o l u m n 3 repor ts results for 583 events at 290 firms measur ing t ime from the proxy mai l ing date. C o l u m n 4 repor ts results for 258 firms exper iencing 534 events measu r ing t ime from the shareho lder mee t ing date. C o l u m n 5 repor ts results for the 16 events a t 15 firms at which at least one p roposa l received a major i ty of votes, measu r ing t ime from the shareho lder mee t ing date. N o r m a l re turns are es t imated using a D i m s o n (1979) style m a r k e t mode l with one l agged and one lead ing term in add i t ion to the c o n t e m p o r a n e o u s m a r k e t re turn term. The equal ly-weighted C R S P index is used as a proxy for the m a r k e t portfolio. The z-statist ic, which is d i s t r ibu ted s t andard no rma l under the null hypothes i s of 0 a b n o r m a l performance, is in parentheses . The percentage of a b n o r m a l re turns greater t han 0 is in square brackets .

Event da te C u m u l a t i o n Shareho lde r interval re la t ive WSJ Proxy Shareho lder mee t ing date to the event date a n n o u n c e m e n t ma i l ing da te mee t ing date (winning proposals)

[ - 20, - 23

[ - 1 0 , - 2 ]

[ - 1 ,0]

- 1

[ - 1 , 1 ]

[1, 10]

[1 ,20]

[ - 20, 203

N u m b e r of events N u m b e r of firms

0.27l 0.16)

[48.1] 1 . 8 9

(1.64) [59.2]

- 0.223 - 0.42)

[55.53

- 0.326 - o.87) [37.0]

O.lO3 (0.27)

[55.53 - 0.191 - 0.51) [44.4]

- 0 . 4 1 4

- 0.63) [37.0]

- 0.79 - 0.65) [29.6] 8

- 2.57 - 1.46) [37.0]

- 3.06 - 1.16) [37.0]

27 24

- 0.401 - 0.275 0.485 - 1.19) ( 0 . 7 4 ) (0.20) [46.3]" [44.9] 8 [43.7]

- 0.192 - 0.063 0.378 - 0.84) ( - 0.25) (0.23) [44.5] c [46.6] [43.7]

- 0.117 - 0.055 1.090 - 1.11) ( - 0.48) (1.44) [48.0] [47.0] [50.0]

- 0.053 - 0.045 1.023 - 0.72) ( 0.56) (1.92)" [47.5] [46.4] [50.0]

- 0.063 - 0.009 0.066 - 0.89) ( - 0.12) (0.12) [46.8] [47.3] [50.0]

- 0.058 - 0.077 - 0.218 - 0.79) ( - 0.96) ( - 0.41) [45.6] 8 [46.6] [37.5]

- 0.175 - 0.132 0.871 - 1.36) ( - 0.94) (0.94) [46.8] [46.4] [56.2]

0.133 - 0.179 - 0.116 - 0.55) ( 0.68) ( - 0.07) [47.5] [48.6] [43.7]

0,142 - 0.174 - 0.853 (0.41) ( 0.46) ( - 0.34)

[49.9] [48.5] [43.7]

- 0.375 - 0.505 0.721 - 0.72) ( - 0.89) (0.19) [48.8] [46.6] [50.0]

583 534 16 290 258 15

~,8~ Stat is t ical s ignif icance at the 10%, 5%, and 1% levels, respectively, for two- ta i led tests.

382 J.M. Karpoff et al./Journal of Financial Economics 42 (1996) 365-395

weak evidence of a negative stock price reaction after The Wal l Street Journal press date. Only 29.6% of the sample firms have positive cumulative abnormal returns over days + 1 through + 10 relative to the press date. This fraction differs significantly from one-half at the 5% level.

6~2. Abnormal returns around proxy mailing dates

Table 4 also reports event study results when time is measured from the proxy mailing date. These results are based on the much larger sample of 583 proposal events at a total of 290 firms. The average cumulative abnormal returns are negative over most intervals around the proxy mailing date, but are statistically insignificant. Some of the nonparametric test results, however, are insignificant. For the intervals [ - 2 0 , - 2 ] , [ - 10, - 2 ] , a n d for day +1 relative to the proxy mailing date, the fraction of firms with positive abnormal returns is significantly below one-half. Overall, there is no indication that the mean abnormal returns are positive around the proxy mailing date.

We conduct several tests to assess the relations between abnormal returns and proposal characteristics. In one type of test, we estimate cumulative average abnormal returns for various time intervals around the proxy mailing date for the different categories of proposals and sponsors distinguished in Table 1, and find that most of them differ insignificantly from zero. The exceptional estimate is for compensation-related proposals. The average abnormal return for these proposals summed over the period from one day before to one day after the proxy mailing date is - 0.885%. This estimate is statistically significant at the 5% level. However, multivariate regression results fail to confirm a significant difference between compensation-related proposals and other proposals.

We perform several cross-sectional, weighted least squares regressions of abnormal returns, summed over the proxy mailing date and the day before, on dummy variables that identify resolution and sponsor categories. 9 Conjecturing that the ex post voting outcomes serve as a proxy for the proposals' ex ante chances of success, we also include regressors that measure the fraction of votes cast in favor of the proposals. In these regressions, however, none of the estimated coefficients is statistically significant. These results indicate that abnormal returns around the proxy mailing date are not related to proposal type, sponsor identity, or the fraction of votes subsequently cast in favor of the proposal.

The stock price evidence does not support the hypothesis that public institutions' proposals influence target companies to make value-decreasing investments. The abnormal returns for proposals sponsored by public institutions are

~These regressions weight observations by the inverse of the estimated standard deviation of the two-day abnormal return. Details of the procedure are available from the authors on request.


not significantly different from zero or from the abnormal returns for other proposals. There also is little evidence that high-profile institutions' proposals have unusual effects on stock returns. None of the average two-day abnormal returns for proposals sponsored by CalPERS, the California State Teachers Retirement System, or the State of Wisconsin |nvestment Board are significantly different from zero. Similar to this result, Smith (1996) reports a zero mean stock price reaction to announcements that CalPERS had targeted a firm for pressure. Wahal (1996) also reports a zero mean stock price reaction to proposal target- ings by nine prominent institutions.

6.3. Abnormal returns around shareholder meeting dates

We also calculate abnormal stock returns around the meeting dates when shareholders voted on the proposals. Typically, voting results are available to investors within hours, or at most a day or two, of the annual meeting. Preliminary vote tallies are often announced at the meeting. 1° Forty-nine proposals in our sample were disallowed by the Securities and Exchange Commission or not voted on for other reasons, leaving shareholder vote data on 534 proposal events at 258 firms. The results are reported in Table 4. For various cumulation intervals that include the shareholder meeting date, the average abnormal returns are negative, but not significantly different from zero. The point estimates also are small. For example, the average two-day meeting day abnormal return is - 0 . 0 5 5 % . These results suggest that, on average, shareholder proposals' voting outcomes have negligible valuation effects.

The last column of Table 4 reports results for those cases in which a proposal received a majority of votes. For day - 1, the average abnormal return is 1.023%. This is significantly greater than zero at the 10% level. Though not reported in the table, cross-sectional weighted least squares regressions indicate that the two-day abnormal return at the shareholder meeting is positively and significantly related to a dummy variable that identifies proposals with at least 50% favorable votes. These results suggest that successful proposals could be associated with stock price increases, a possibility that is examined in more detail below.

7. Additional evidence on proposals' effects

7.1. A closer look at proposals that win

Sixteen proposal events, involving 22 proposals at 15 different firms, garnered a majority of shareholder votes. Using The Wall Street Journal lndex and the

I°We are grateful to John Wilcox, Board chairman of the proxy solicitation firm Georgeson & Company, for information about when shareholder voting results become known to investors.

384 J.M. Katpoff et al./Journal of Financial Economics 42 (1996) 365 395

Lexis/Nexis database, we compiled a history of each of the 15 companies over a six-year period beginning three years before each successful proposal 's proxy mailing date. Table 5 provides information on these proposals. We conjecture that proposal victories are at least partly unanticipated, and that these are the proposals that are most likely to pressure managers for change. Hence, if shareholder proposals affect share values and corporate policies, we are most likely to observe that these effects result from successful proposals.

Table 5 contains very little evidence to support the hypothesis that successful proposals typically generate wealth gains. The average abnormal two-day return on the annual meeting date is positive (1.09%), but significant only at the 15% level for a two-tailed test (z = 1.44). Moreover, this average is heavily influenced by one positive outlier. Panel A, Table 5 shows that on May 4, 1989, Avon Products, Inc. shareholders approved a proposal to rescind the firm's poison pill. The associated two-day abnormal return is 12.58%. This abnormal return reflects an important confounding event: On May 3, 1989, the press reported that Amway Corporat ion and investor Irwin Jacobs had acquired 10.3% of Avon's shares and intended to seek control of the firm. Avon Products ' abnormal return over the May 3~4 period therefore reflects the pending takeover bid as well as the poison pill recision vote. Excluding this event, the two-day abnormal return on the annual meeting date for the remaining 15 success events is 0.33%, which differs insignificantly from zero (z = 0.41). Hence, discounting the 1989 Avon Products outlier, there is no persuasive evidence that successful initiatives increase share values.

It is possible that we detect no stock price changes attributable to successful proposals because we focus on the wrong dates. Our procedure assumes that investors first learn of vote outcomes on or about the annual meeting dates. Unfortunately, for nine of the 16 successful events, we are unable to determine when voting results were first made available to investors. Press reports indicate that for at least seven of the events, vote outcomes were revealed on the day of the annual meeting. In at least one of these seven events, the vote outcome was anticipated in press reports on the meeting day. The average abnormal two-day return on the annual meeting date for this set of seven events, -0.07%, differs insignificantly from zero (z = 0.87).

7.2. Longer-term effects" oJsucces~'[id proposals

We also look for direct evidence of policy changes ensuing from the successful proposals. Of the 16 events reported in Table 5, five occurred as the target firm was attempting to defeat a hostile takeover bid, two occurred within one year after the firm defeated a hostile control bid, and nine involved firms not in a control battle at the time (although four of these nine firms had been subjects of takeover rumors). Specific policies called for in a proposal subsequently were adopted by Chock Full O'Nuts, Champion International, Kmar t Corp., and

J.M. Karp(~ff et al./Journal o f Financial Economics 42 (1996) 365 395 385

Santa Fe Southern Pacific. In addition, several firms underwent other major policy changes following the successful proposals. In most cases, however, other events probably account for the policy changes. For example, external control battles, and not the shareholder proposals, most likely motivated restructuring efforts at USX Corp., Lockheed Corp., Milton Roy Co., and National Inter- group Inc. Large outside investors or control battles may have prompted the less extensive changes undertaken by Chock Full O'Nuts, Avon Products, and Gillette.

Perhaps the most influential among the 22 successful shareholder proposals involved Santa Fe Southern Pacific Corp. On May 24, 1988, shareholders approved a proposal to rescind the firm's poison pill. At the time, Santa Fe's managers were resisting a takeover bid by the Henley Group, Inc., and a separate bid for influence by another large shareholder, Olympia and York Develop- ments, Ltd. On the day of the shareholder meeting, The Wall Street Journal reported that the proposal was likely to pass, stating that the proposal's '... ability to attract majority support could put increased pressure on the company and make it more open to an eventual takeover' (Rose, 1988). The 3.48% two-day abnormal return at the shareholder meeting suggests that investors regarded the proposal as value-increasing. In the following years the firm undertook a major restructuring effort along the lines that had been proposed by the hostile bidders. Although Santa Fe repelled the hostile bids and retained its poison pill, the shareholder proposal presumably encouraged managers to undertake the restructuring effort.

The Santa Fe Southern Pacific case notwithstanding, most successful shareholder proposals seem to have had little impact on firm operations or policies. We can find no evidence in Lexis/Nexis or The Wall Street Journal Index, for example, that successful shareholder proposals had any influence at Armco Inc., Consolidated Freightways Inc., First Bank System Inc., Kmart Corp., Outboard Marine Corp., and Ryder Systems Inc.

In two cases, proposals succeeded even as an outside control bid failed. At its 1990 meeting, Lockheed Corp. shareholders approved resolutions calling for confidential voting, redemption of the firm's poison pill, and opting out of the Delaware state takeover law. At the same time they rejected Harold Simmons' proxy bid for control of Lockheed's board. The two-day abnormal return for Lockheed is negative ( - 0.63%). Similarly, in May 1990, USX Corp. shareholders approved resolutions calling for a declassified board and redemption of the firm's poison pill. But at the same time, management defeated a proxy challenge for control by Carl Icahn. The two-day abnormal return for the shareholder meeting is - 2.41%. In both cases, the abnormal returns reflect the outcomes of both the shareholder proposals and the control bids. Even if the proposals had positive valuation effects, the effects appear to be small compared to the valuation effects associated with the failed bids for control.

386 J.M. Karpq)f et al./Journal o f Financial Economics 42 (1996) 365 395

e',

..~ =

.o

o

¢.

~b

.E

-a ©

o

~o

.o

b

Q r- "~

%

o

0

-5

2

o

,q.

<

c~

i..)

0

0

0

d r-,

*d

©

P. 0

<

o ,-4 I I I

©

~ ~ ~ == ~

; z

%

d ~ o

c5 I I

~. -5

"F-. ~. "E.

(5 Y. °

J.M. Karpoff et al./Journal of Financial Economics 42 (I996) 365 395 387

I I I

<

= ~ = = = ~'

2

~a 9

E

E

~e

Z.~ g~ ' ~ .~

r m ~

~ . ~


7.3. CEO turnover Jbllowing shareholder proposals"

Our case studies of the 16 success events do not reveal any systematic relation between the proposals and subsequent changes in firm policies. Another indication of a change in firms' policies is a change in its top manager. To draw inferences about proposals' effects from a larger pool of firms, we examine chief executive officer (CEO) turnover in the years following a proposal. We obtain information on CEOs from the Forbes 800 for firms in our primary sample. For 111 firms, the CEO data are available for both the firm and its matched control firm.

Gilson (1989) reports that CEO turnover frequently is related to financial performance. We therefore estimated logistic regressions to control for firm performance as well as the presence of a shareholder proposal. In each logistic regression, the dependent variable is one for all firms changing CEOs within one year after the proxy mailing date, and zero otherwise. The independent variables include the control and performance variables reported in Table 3, and the CEO's age. A dummy variable is set equal to one for all firms that experienced their first shareholder proposals (in our sample) during the previous year. When all four performance variables are included as regressors, the coefficient on the dummy variable representing proposal firms is 0.12 and the ratio of the coefficient to its standard error is 0.18. When we consider CEO turnover within three years following the firm's first shareholder proposal, the coefficient is 0.43 and the ratio of the coefficient to its standard error is 1.07. These results indicate that the probability of a change in CEO is insignificantly related to a previous shareholder proposal.

8. Operating performance changes following shareholder proposals

Our evidence indicates that shareholder proposals are not associated with systematic share value changes or unusual CEO turnover. Even among proposals that win, there is little evidence of policy changes that can be attributed to the proposals. These results suggest that proposals have little effect on firms' operations. In this section we test directly the hypothesis that proposals change operating performance, by examining changes in operating returns on sales and assets. All data for the performance tests are taken from Compustat.

For each firm that received at least one proposal, we calculate the change in each performance measure from one year before the firm's first proposal year through one after the firm's last proposal year in our sample period. We also calculate the change from one year before the first proposal through three years after the firm's last proposal year. We then compute an adjusted performance change, which equals the change in the proposal firm's performance minus the change in a control firm's performance. The control is the firm with the same

J.M. Karp(~ff el al./Journal of Financial Economics 42 (1996) 365 395 389

Table 6 Operating performance changes for proposal and control firms

Comparisons of performance changes for firms receiving shareholder-initiated corporate governance proposals and control firms matched pairwise by industry and performance before the proposal year. Firms receiving proposals are identified by the Investor Responsibility Research Center, Inc. When comparing changes in sales and operating return on sales, the control is the firm with the same two-digit SIC that has the closest operating return on sales to the proposal firm in the year before the firm's first proposal in our sample. When comparing changes in assets and operating return on assets, the control is the firm with the same two-digit SIC that has the closest operating return on assets to the proposal firm in the year before the firm's first proposal. Financial data are from Compustat.

Mean Mean (median) (median) t-statistic % of matches value for value for on paired Number for which proposal control difference of proposal value firms firms (Wilcoxon matched > (%) (%) z-statistic) pairs control value

Panel A: Comparisons ji'om one year before the.first proposal to one year ajter the last proposal

Average annual rate 5.3 8.0 - 2.33 b 205 43 of change in sales (5.2) (7.0) ( 2.36)"

Change in operating - 0.8 - 1.3 0.74 204 54 return on sales ( - 0.2) ( 1.0) (1.05)

Average annual rate 7.4 7.8 - 0.26 207 51 of change in assets (6.3) (5.7) (0.18)

Change in operating - 1.3 2,0 1.31 202 52 return on assets ( 0.6) ( - 1.1) (1.64)

Panel B: Comparisons ji'om one year before the first proposal to three years after the last proposal

Average annual rate 6.3 10.7 3.31 ~ 187 44 of change in sales (6.1) (8.1) ( - 2.59) ~

Change in operating 0.2 - 0.7 1.07 187 51 return on sales ( - 0.2) ( - 0.9) (0.72)

Average annual rate 9.5 10.7 - 0.59 189 55 of change in assets (7.7) (8.3) ( - 0.31)

Change in operating 1.4 - 2.2 1.21 188 57 return on assets ( - 1.0) ( 1.3) (1.86) a

~,b.~ Statistical significance at the 10%, 5%, and 1% levels, respectively, for two-tailed tests.

t w o - d i g i t C o m p u s t a t S I C a n d p e r f o r m a n c e m e a s u r e t h a t m o s t c l o s e l y m a t c h e s

t h a t o f t h e p r o p o s a l f i r m in t h e y e a r b e f o r e i ts f i r s t p r o p o s a l . B a r b e r a n d L y o n

(1995) r e p o r t t h a t a d j u s t e d p e r f o r m a n c e c h a n g e s t h a t m a t c h p r o p o s a l a n d

c o n t r o l f i r m s b y p r i o r p e r f o r m a n c e y i e l d t e s t s w i t h g r e a t e r p o w e r a n d r e l i a b i l i t y

t h a n m a n y o t h e r m e a s u r e s .


Table 6 gives summary statistics on the proposal and control firms' changes in sales, assets, and returns on sales and assets. The number of matched firm pairs varies from 187 to 207, depending on data availability. The average annual rate of change in sales is significantly higher for the control firms: From one year before a firm's first proposal through one year after its last proposal, for example, the rate of change is 5.3% for proposal firms and 8% for the controls. Both proposal and control firms experience average decreases in operating return on sales. Using the one-year-before to one-year-after changes reported in panel A, the average change in operating return on sales is - 0 . 8 % for the proposal firms and - 1.3% for the controls. Neither the proposal firms' nor the control firms' changes in operating return on sales are statistically significant, nor are the changes significantly different from each other.

Although proposal firms' sales grow relatively slowly after receiving proposals, their asset growth rates are closer to those of the control firms. There is slight evidence that proposal firms' change in operating return on assets differs from that of the controls. In panel B of Table 6, the proposal firms' average change in return on assets ( - 1.4%), measured from one year before the first proposals through three years after last proposals, is greater than that for the controls ( - 2.2%). The Wilcoxon z-test indicates that this difference is statistically significant at the 10% level. The t-statistic for the paired difference, however, is insignificant. Moreover, in panel A, in which performance changes are measured from one year before first proposals through one year after last proposals, the difference in return on assets changes also is insignificant.

We also investigated the relation between the changes in operating performance and the persistence, intensity, and influence of shareholder proposals' sponsors targeting the firm. We associate persistence with the number of years that a firm receives at least one proposal of any type, and intensity with the total number of proposals of all types received over the sample period. To measure influence, we create dummy variables based on the number of proposals sponsored by public institutions and by private institutions. We treat one prominent public institution, CalPERS, separately. The view that shareholder proposals prompt value-increasing operating changes implies that performance changes are positively related to proposal persistence and intensity. If public or private institutions are more influential than individuals, performance changes should be positively related to the number of proposals sponsored by such institutions.

We estimated several regressions to test these conjectures. The dependent variable alternately is the change in operating return on sales or in operating return on assets. Because we measure performance changes over different numbers of years for different firms, the regression error is likely to be hetero- skedastic. Accordingly, we use weighted least squares. The weight for each observation is the inverse of the square root of the number of years over which the performance change is measured. In addition to dummy variables reflecting


proposal persistence, intensity, and sponsor identity, we include dummies that reflect how frequently firms received proposals related to external control, internal governance, and compensation issues.

By design, each dummy variable used in the regressions partitions the observations at the median value of some underlying measure. For example, the median number of years in which firms received proposals over the sample period was one. Hence, firms receiving proposals in two or more years were subject to relatively persistent proposal activity. For these firms, the variable 'Proposals in more than one year' equals one. It equals zero for all other firms. We construct similarly dummy variables associated with proposal intensity, sponsor influence, and proposal types.

In one regression, the dependent variable is the change in operating return on sales, measured from one year before the firm's first proposal through one year after the firm's last proposal. In another regression, the dependent variable is this same change minus the contemporaneous change in operating return on sales for the control firm. In a third regression we measure the adjusted operating return on sales change through three years after the firm's last proposal in our sample. In none of these three regressions is the F-statistic significant. These results indicate that neither unadjusted nor adjusted operating returns on sales are significantly related to the persistence, intensity, or other characteristics of proposal activity.

We also perform regressions in which the dependent variable is the operating return on assets (ROA). All operating ROA measures are significantly lower for firms receiving a large number of internal governance proposals. These results suggest that the marginal relation between internal governance proposals and changes in operating ROA is negative. The two adjusted operating ROA measures also are positively related to the number of years a firm receives proposals. This result suggests that persistent proposal pressure over multiple years is associated with improvements in operating return on assets. The F-statistic, however, is significant in only one of the regressions. Furthermore, in sensitivity tests the persistence dummy coeffÉcient is not consistently significant. Hence, the result is not robust.

In our regression analysis, the coefficient on a dummy variable for compensation proposals often is negative, although in most cases it is not statistically significant. Negative coefficients are consistent with our event study finding that mailing proxy materials containing information on compensation proposals is associated with negative abnormal stock returns. The results suggest that compensation-related proposals hamper operations and destroy value. How- ever, the evidence is weak.

We also investigated operating performance changes for various subsets of the sample. These subsets include firms at which a proposal won a majority vote, and firms at which CalPERS initiated proxy proposals. In none of these cases can we reject the null hypothesis that all coefficients in the regressions equal zero.


9. Conclusion

We find that the probability of receiving a shareholder corporate governance proposal is negatively related to a firm's market-to-book ratio, operating return on sales, and recent sales growth. This indicates that proposals are targeted at poorly performing firms. Unlike some other forms of shareholder activism, however, proposals have little effect. Wahal (1996) reports that activist institutions' efforts to promote organizational changes through negotiations with top managers are associated with gains in the target firms' stock prices. Strickland, Wiles, and Zenner (1996) report that firms that were targeted for pressure by the United Shareholders Association, and which negotiated settlements with the group, also experienced positive average abnormal stock returns. In contrast, the average effect of shareholder corporate governance proposals on stock values is close to, and not significantly different from, zero. This holds whether we focus on initial press announcements of the proposals, proxy mailings that publicize the proposals, or meeting dates at which shareholders vote on the proposals. Furthermore, average abnormal returns are not systematically related to proposal type or sponsor. Proposals sponsored by institutions have insignificant stock price effects.

We find little evidence that shareholder proposals engender firm policy changes. The probability that a firm changes its CEO is not significantly related to the prior receipt of a shareholder proposal. Even proposals that gain 50% support do not appear to alter their target firms' policies or stock values. Some of the firms that attract these proposals restructured their operations, but these changes were most often accompanied by external control threats.

An exception to the general pattern of our findings is that firms that receive shareholder proposals experience smaller subsequent declines in operating ROA than do control firms matched by industry and prior performance. This difference is statistically significant at the 10% level in one of our tests. The change in operating ROA is also positively related to the number of years in which a firm receives at least one proposal. This result is consistent with the view that persistent shareholder pressure, in the form of shareholder proposals over more than one year, prompts beneficial operating changes.

Other than this exception, most evidence indicates that proposals have little effect on operating performance. Sales growth declines for firms that receive proposals in relation to sales growth for control firms. Changes in operating return on sales are not significantly larger for proposal firms than their controls, and are not significantly related to the persistence or intensity of proposal pressure, or to the sponsors' identity. Changes in operating ROA are not related to the pressure's intensity or sponsors' identity. Proposals sponsored by CalPERS, for example, are not associated with unusual performance improvements. There also is evidence that the receipt of internal governance proposals is negatively related to the change in operating ROA.

j.3,1. Karpoff et al./Journal of Financial Economics 42 (1996) 365 395 393

J u d g i n g f r o m p r i o r s tud ies , m a n y r e s e a r c h e r s a n d p r a c t i t i o n e r s t h i n k s h a r e -

h o l d e r a c t i v i s m p r o m o t e s v a l u e - i n c r e a s i n g pol ic ies . H o w e v e r , o u r f i n d i n g s o n

s h a r e h o l d e r - i n i t i a t e d c o r p o r a t e g o v e r n a n c e p r o p o s a l s d o n o t s u p p o r t th i s view.

T h e r e is n o p e r s u a s i v e e v i d e n c e t h a t t h e s e p r o p o s a l s i n c r e a s e f i rm va lues ,

i m p r o v e o p e r a t i n g p e r f o r m a n c e , o r i n f l u e n c e f i rm pol ic ies .

References

Ambrose, Brent W. and William L. Megginson, 1992, The role of asset structure, ownership structure, and takeover defenses in determining acquisition likelihood, Journal of Financial and Quantitative Analysis 27, 575 589.

Anderson, Ronald C. and D. Scott Lee, 1996, Ownership databases: Which brand is right for you?, Working paper (Texas A&M University, College Station, TX).

Barber, Brad M. and John D. Lyon, 1995, Detecting abnormal operating performance: The empirical power and specification of test-statistics, Journal of Financial Economics 41,359-399.

Bergin, Paul R., 1988, Voting by institutional investors on corporate governance issues in the 1988 proxy season (Investor Responsibility Research Center, Washington, DC).

Biersach, Jeffrey W., 1990, Voting by institutional investors on corporate governance issues in the 1990 proxy season (Investor Responsibility Research Center, Washington, DC).

Black, Bernard S., 1990, Shareholder passivity reexamined, Michigan Law Review 89, 575 591. Black, Bernard S., 1992, Institutional investors and corporate governance: The case for institutional

voice, Journal of Applied Corporate Finance, Fall, 19 32. Coffee, John C., Jr., 1991, Liquidity versus control: The institutional investor voice, Columbia Law

Review 91, 1277 -1338. Comment, Robert and G. William Schwert, 1995, Poison or placebo? Evidence on the deterrent and

wealth effects of modern antitakeover measures, Journal of Financial Economics 39, 3 43. Danielson, Morris G., Jonathan M. Karpoff, and M.Wayne Marr Jr., 1995, Corporate governance

and firm performance: Empirical evidence, Working paper (University of Washington, Seattle, WA).

DeAngelo, Harry and Edward M. Rice, 1983, Antitakeover charter amendments and stockholder wealth, Journal of Financial Economics 1 l, 329-360.

Demsetz, Harold and Kenneth Lehn, 1985, The structure of corporate ownership: Cause and consequences, Journal of Political Economy 93, 1155 1177.

Dimson, Elroy, 1979, Risk measurement when shares are subject to infrequent trading, Journal of Financial Economics 7, 197 226.

Dobrzynski, Judith H., 1990, Shareholders unfurl their banner: 'Don't tread on us', Business Week, June 11, 66.

Famm Eugene F., 1976, Foundations of finance (Basic Books, New York, NY). Galant, Debbie, 1990, Putting new muscle into proxy voting, institutional Investor 24, Feb., 161. Gillan, Stuart L. and Laura T. Starks, 1995, Relationship investing and shareholder activism by

institutional investors, Working paper (University of Texas at Austin, Austin, TX). Gilson, Ronald J., 1986, The law and finance of corporate acquisitions (Foundation Press, Mineola,

NY). Gilson, Stuart C., 1989, Management turnover and financial distress, Journal of Financial Eco-

nomics 25, 241 262. Gordon, Lilli, 1993, New deal for shareholders, The Wall Street Journal, Feb. 4. Gordon, Lilli A. and John Pound, 1993, Information, ownership structure, and shareholder voting:

Evidence from shareholder sponsored corporate governance proposals, Journal of Finance 48, 697 718.


Hasbrouck, Joel, 1985, The characteristics of takeover targets, Journal of Banking and Finance 9, 351-362.

Jarrell, Gregg A. and Annette B. Poulsen, 1987, Shark repellents and stock prices: The effects of antitakeover amendments since 1980, Journal of Financial Economics 19, 127-168.

Jensen, Michael C., 1993, The modern industrial revolution, exit, and the failure of internal control systems, Journal of Finance 68, 831-880.

John, Kose and April Klein, 1995, Shareholder proposals and corporate governance, Working paper (New York University, New York, NY).

Karpoff, Jonathan M. and Paul H. Malatesta, 1989, The wealth effects of second generation state takeover legislation, Journal of Financial Economics 25, 291-322.

Kole, Stacey R., 1995, Measuring managerial equity ownership: A comparison of sources of ownership data, Journal of Corporate Finance 1, 413~435.

Lang, Larry H.P., Ren6 M. Stulz, and Ralph A. Walkling, 1989, Managerial performance, Tobin's q, and the gains from successful tender offers, Journal of Financial Economics 24, 137 154.

Linn, Scott C. and John J. McConnell, 1983, An empirical investigation of the impact of 'antitakeover amendments' on common stock prices, Journal of Financial Economics 11,361 399.

lipton, Martin and Steven A. Rosenblum, 1991, A new system of corporate governance: The quinquennial election of directors, University of Chicago Law Review 58, 187 253.

Malatesta, Paul H. and Ralph A. Walkling, 1988, Poison pill securities: Stockholder wealth, profitability, and ownership structure, Journal of Financial Economics 20, 347--376.

Mergerstat Review, 1993 (Merrill Lynch Business Advisory Services, Schaumberg, IL). Mikkelson, Wayne H. and M. Megan Partch, 1988, Withdrawn security offerings, Journal of

Financial and Quantitative Analysis 23, 119-143. Mikkelson, Wayne H. and M. Megan Partch, 1989, Managers' voting rights and corporate control,

Journal of Financial Economics 25, 263 290. Morck, Randall, Andrei Shleifer, and Robert W. Vishny, 1989, Alternative mechanisms for corpo-

rate control, American Economic Review 79, 842-852. Mulherin, J. Harold and Annette B. Poulsen, 1994, Proxy contests, shareholder wealth and

operating performance: An analysis of the 1980's, Working paper (University of Georgia, Athens, GA).

Nesbitt, Stephen L., 1992, Rewards from corporate governance, Working paper (Wilshire Associ- ates, Los Angeles, CA).

Nesbitt, Stephen L., 1994, Long-term rewards from shareholder activism: A study of the CalPERS effect, Journal of Applied Corporate Finance 6, Winter, 75 80.

Palepu, Krishna G., 1986, Predicting takeover targets, Journal of Accounting and Economics 8, 3 35.

Pozen, Robert C., 1994, Institutional investors: The reluctant activists, Harvard Business Review, Jan. Feb., 140 149.

Pound, John, 1992, Raiders, targets, and politics: The history and future of American corporate control, Journal of Applied Corporate Finance 5, (:~18.

Romano, Roberta, 1993, Public pension fund activism in corporate governance reconsidered, Columbia Law Review 93, 795 853.

Rose, Frederick, 1988, Santa Fe Southern's annual meeting features 'landmark' poison pill vote, The Wall Street Journal, May 24, 7.

Ryan, Patrick J., 1988, Rule 14a-8, institutional shareholder proposals, and corporate democracy, Georgia Law Review 23, 97 184.

Ryngaert, Michael, 1988, The effect of poison pill securities on shareholder wealth, Journal of Financial Economics 20, 377418.

Salinger, Michael, 1992, Standard errors in event studies, Journal of Financial and Quantitative Analysis 27, 39 54.

J.M. Karp()ff' et al./Journal of Financial Economics 42 (1996) 365 395 395

Saxton, Jim, 1994, A raid on America's pension funds, The Wall Street Journal, Sept. 29, AI4. Smith, Michael P., 1996, Shareholder activism by institutional investors: Evidence from CalPERS,

Journal of Finance 51,227 252. Song, Moon H. and Ralph A. Walkling, 1993, The impact of managerial ownership on acquisition

attempts and target shareholder wealth, Journal of Financial and Quantitative Analysis 28, 439-457.

Strickland, Deon, Kenneth W. Wiles, and Marc Zenner, 1996, A requiem for the USA: Is small shareholder monitoring effective?, Journal of Financial Economics 40, 319 338.

Wahal, Sunil, 1996, Public pension fund activism and firm performance, Journal of Financial and Quantitative Analysis 31, 1 23.

Weston, J. Fred, Kwang S. Chung, and Susan Hoag, 1990, Mergers, restructuring, and corporate control (Prentice-Hall, Englewood Cliffs, N J).

Woh|stetter, Charles, 1993, Pension fund socialism: Can bureaucrats run the blue chips?, Harvard Business Review 71, Jan. Feb., 78.