Standstill agreements and market response based on ownership structure of rget firms

I approach this task by cross-sectionally analyzing the abnormal returns during a three-day window for the announcement date of the standstill agreements using several variables that pro

This manuscript has been reproduced from the microfilm master UMI

f f l m s the text directly from the original or copy submitted Thus, some thesis and dissertation copies are in typewriter face, while others may

be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrougfa, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge Contact UMI directly

to order.

A Bell & Howell Information Company

300 North Z eeb Road Ann Arbor Ml 4 8 1 0 6 -1 3 4 6 USA

313/761-4700 800/521-0600

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

OWNERSHIP STRUCTURE OF TARGET FIRMS

byFarooq I Chaudhry

A Dissertation Presented in Partial Fulfillment

of the Requirements for the Degree

UMI Microform 9539534 Copyright 1995, by UMI Company All rights reserved.

This microform edition is protected against unauthorized

copying under Title 17, United States Code.

UMI

300 North Zeeb Road Ann Arbor, MI 48103

OWNERSHIP STRUCTURE OF TARGET FIRMS

byFarooq I Chaudhry

has been approved July 1995

APPROVED:

, Chairperson

Supervisory Committee

ACCEPTED:

Dean of the College

Dean, Graduate College

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

A standstill agreement is a voluntary contract between a firm and a substantial stockholder that limits the stockholder's ownership of voting shares to a maximum percentage for a specific period Management does not need approval of the shareholders

to enter the firm into the agreement Specific characteristics of the agreement can vary by situation but they can eliminate or restrict the ability of hostile stockholders to gain control

of the firm through contractual agreement, repurchase of stock, or both

The finance literature provides mixed evidence o f the roles that institutional ownership and blockholder ownership play in monitoring firm management or how monitoring affects managerial actions Institutional and blockholder ownership may substitute for insider holdings to align the interest of managers with those of the stockholders through monitoring, especially when managerial holdings are small When insider holdings are large, the alignment benefits may be offset by the entrenchment effects However, external monitoring may still provide some discipline to managerial actions

I examine whether actions of target firms' managements in entering standstill agreements are in the best interest of their nonparticipating shareholders I approach this task by cross-sectionally analyzing the abnormal returns during a three-day window for the announcement date of the standstill agreements using several variables that proxy for the ownership structure of target firms The announcement period abnormal return for 249 observations is a statistically significant -0.5 percent However, investors do not suffer any loss when you consider abnormal returns for the whole control event A non-linear

iii

and the announcement period abnormal returns Positive abnormal returns exist for all firms in the pre-event period, except for a group of firms with over 25 percent insider ownership The average announcement period return is negative for NYSE firms, but it

is positive for non-NYSE firms All non-NYSE firms, except those having insider ownership between 5 to 25 percent, experience a negative average abnormal return of 10 percent or more after the agreement Institutional ownership has a negative correlation with the announcement effect in the under 5 percent group This result suggests that institutional monitoring in this group is ineffective The opposite effect exists for the 5 to

25 percent group, suggesting that when management enters a standstill agreement it does

so for the good of the firm When both a confidentiality and standstill agreement is present, significant abnormal returns exist and the firms have a change o f control within three years o f the agreement

iv

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

I am most grateful to Dr George Gallinger for his time and patience in reading and guiding me in the research process I am also deeply grateful to Dr Herbert Kaufman and

Dr Linda Martin for their help on my dissertation committee and their constructive comments Special thanks to Peter Luan for help with computer problems and Dr Ibrahim Helou for his continued moral support in completing this project I am also indebted to Dr Richard Smith for encouraging me to work on the topic of standstill agreements

My greatest debt is to my parents who instilled in me the desire to seek knowledge

My parents died in a car crash when I was fourteen years old and it is on occasions like this that I miss them terribly I am sure they would have been very proud of this accomplishment, and how I wish I had them around to join me in my joys and sorrows

v

List of T a b l e s vii

1 Introduction 1

2 Literature Review and Hypothesis D e v elo p m e n t 7

2.1 Convergence-of-Interest Hypothesis vs Management Entrenchment Hypothesis 10

2.2 Other H y p o th e sis 13

2.2.1 Ownership Structure and Corporate Value 13

2.2.2 Blockholders as Monitors 19

2.2.3 Institutional M o n ito rin g 20

3 Data and Methodology .24

3.1 Identification o f Sample and Measurement of Abnormal R e t u r n s 24 3.2 Modeling Abnormal Returns Occurring on ED A Y 29

3.3 Data Summary D e sc rip tio n 36

4 Empirical Results .44

4.1 Average Abnormal Return Results 44

4.2 Cross-sectional Analysis of the Announcement E f f e c t 59

5 Conclusions 82

R e fe re n c e s 89

vi

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

1 Definition of dummy variables corresponding to different time periods relative tothe announcement o f standstill agreement 27

2 Definitions and expected signs of the coefficients of the variables used in the

cross-sectional regression explaining the announcement effect of standstill agreements 30

3 Frequency distribution of standstill agreements by year, by number o f firms and

by stock listing 37

4 Summary of descriptive statistics for 249 standstill agreement events pertaining to

202 firms from 1978 to 1992 39

5 Insider stock holdings, insider block holdings, other block holdings, institutional

ownership, and percent of target stock owned by the bidder prior to the standstill event for the total sample and by NYSE and non-NYSE subgroups 42

6 Average abnormal stock returns reported for firms by exchange o f listing and

various important c a te g o rie s 51

7 Average abnormal stock returns reported in three ownership subgroups for total

sample and by stock lis tin g 57

8 Model 1: Ordinary least squares regression analysis of various ownership variables

and other factors to explain the abnormal returns generated on the announcement

o f standstill a g re e m e n ts 66

9 Model 2: Ordinary least squares regression analysis of various ownership variables

and other factors to explain the abnormal returns generated on the announcement

o f standstill agreements from 1978 to 1992 The number o f institutional stockholders is added to this model 73

10 Model 3: Ordinary least squares regression analysis of various ownership variables

and other factors to explain the abnormal returns generated on the announcement

o f standstill agreements from 1978 to 1992 t-statistics are corrected for heteroscedasticity (White (1980)) and are noted in parenthesis under the respective coefficients An exchange dummy variable is added to this model 80

V ll

A standstill agreement is a voluntary contract between a firm and a substantial stockholder that limits the stockholder's ownership of voting shares to a maximum percentage for a specific period.1 Management does not need approval of the shareholders

to enter the firm into the agreement Specific characteristics of the agreement can vary by situation but they generally include the following features The agreement can eliminate

or restrict the ability of hostile stockholders to gain control of the firm through contractual agreement, repurchase of stock, or both The agreement often allows a substantial stockholder to make either a friendly or an unfriendly cash tender offer for all shares outstanding In return, management of the target firm promises not to erect any takeover defenses for a certain period Frequently, a targeted repurchase or "greenmail" purchase for all or part of the substantial stockholder's equity in the target firm accompanies the agreement.2 Finally, the agreement usually requires potential bidders to vote as directed

by the target firm 's management

Popular perceptions of why management of a target firm enters into a standstill agreement are threefold First, Linn and McConnell (1983) argue that a standstill agreement places management in a better position to bargain with raiders on behalf of stockholders The agreement allows time for a bidding war to occur with a higher price for the stockholders as the expected outcome However, it is puzzling why target managers wait for an initial bid to do something about maximizing firm value Target management's

^ h e use of standstill agreements between debtor and creditors, when debtor is in technical default of loan agreement, is not the subject of this study

2Greenmail is an offer by management to repurchase the shares of a subset of shareholders at a premium The offer is not made to other shareholders

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

Kim (1983); that is, management is reluctant to do anything until forced by outside parties Second, the standstill agreement may provide management with cover to accept longer term projects that appear to be unprofitable in the short run Stein's (1988) managerial myopia argument points out that the capital market systematically undervalues the delayed payoffs on long-term investments Takeover pressure leads managers to focus more on short-term profits rather than on meeting long-term objectives A standstill agreement can relieve the pressure of having to attain short-term results If management is able to concentrate its effort on long-term investment projects, shareholders can benefit.3 Third, the agreement may reduce expected legal costs In most cases of managerial resistance to

a takeover attempt, bidders file lawsuits against the firm and its management Bidders claim that management's stewardship of the company leads to entrenched officers, which

is detrimental to stockholders Regardless of the lawsuit's merit, the cost of a protracted legal battle can be substantial to the target firm, thereby lowering returns to stockholders

Limited empirical research exists about standstill agreements Dann and DeAngelo (1983) studied 30 firms that entered standstill agreements and/or negotiated premium targeted repurchases They found that the agreements and repurchases resulted in statistically significant negative average returns to stockholders.4 Dann and DeAngelo

3For example, management might be able to boost the stock price by selling off productive assets that shareholders are unable to value properly If left unsold, the assets may have little effect on current earnings and may be undervalued by shareholders Consequently, this sale has an immediate impact on income and may cause an upward revaluation of the company's stock

4Many standstill agreements do not involve repurchase of stock from a bidder but

stockholder interest hypothesis Mikkelson and Ruback (1991) find contrary evidence for targeted repurchases but not for standstill agreements They conclude that large blockholders and targeted repurchases generally help, or at least do not harm, nonparticipating stockholders of the repurchasing firm They also conclude that the importance of a standstill agreement is unclear because it does not restrict investments by other potential blockholders.

Mikkelson and Ruback's (1991) logic is as follows When you view the standstill agreement as an isolated event, there is a stock price decline at the agreement date This result, they claim, leads Dann and DeAngelo (1983) to conclude that management of the target firm harmed nonparticipating shareholders However, when you view the standstill agreement as the outcome of an ongoing corporate control process, the average stock price decline, at either the announcement date or the stock repurchase date, suggests a downward revision in the market's expectations This price revision could be caused by

a realization that expected profitability of available investment choices is too high or management chose an inferior investment alternative Based upon Mikkelson and Ruback’s (1991) research, it is not appropriate to evaluate managerial actions by observing just the announcement period returns For this reason I analyze abnonnal returns to firms entering standstill agreements from the Schedule 13D filing date through to the change-of-control date

simply place a maximum limit of percentage ownership Many targeted repurchases do not involve a standstill agreement

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

aforementioned studies I examine whether actions of target firms' managements in entering standstill agreements are in the best interest of their nonparticipating shareholders.

I approach this task by cross-sectionally analyzing abnormal returns during a three-day window for the announcement date of the standstill agreements using several variables that proxy for the ownership structure of target firms I also follow Mikkelson and Ruback’s (1991) argument and analyze firms involved in standstill agreements from the filing date

of Schedule 13D through the change-of-control date The data set includes 249 standstill events for the period 1978 to 1992 This sample consists of 217 concluded agreements and

32 cases where there is mention of a possible agreement The announcement period abnormal return for the 249 observations is a statistically significant -0.5 percent The announcement period abnormal return for 217 concluded agreements is a highly significant -1 percent The announcement period abnormal return for 32 potential agreements is about

4 percent, which is also significant

There is a simple explanation for the different magnitude of abnormal returns for the two subgroups In most cases a standstill agreement is the outcome of an ongoing corporate control process It begins with the accumulation o f shares of a target firm by a blockholder (potential bidder) who wishes to play a more active role in the affairs of the company The process ends with a standstill agreement and subsequent repurchase of shares by the firm from the blockholder The increase in average stock price associated with an accumulation of shares indicates that initiating the corporate control process benefits shareholders However, market participants are unable to distinguish

that will end with a more favorable outcome for the target firm 's stockholders, such as a takeover bid.

Because of the weak support found for effectiveness of institutional monitoring, the benefits derived from such external monitoring may outweigh the potential public cost of damages if there were no such restrictions The fact that institutions provide some monitoring has a public policy implication for the government to relax some o f the regulatory restrictions imposed on certain institutions because o f the potential benefits of institutional ownership For example, the Investment Company Act of 1940 prevents a mutual fund from calling itself a diversified fund if it owns more than 10 percent of the stock of any single company Also, the Internal Revenue Service allows only diversified funds to pass income through to shareholders untaxed Insurance companies must typically limit equity investment to 20 percent of their assets and no single investment can exceed

2 percent of the assets These restrictions result from regulatory concern about the public costs of under-diversified institutions and may be costly if they unduly limit institutional monitoring [see Eakins (1990)].5

In the next section, I provide an overview of the relevant finance literature This literature provides the basis for development of hypotheses and empirical predictions Section 3 describes the sources and methods used in data collection, data description and

5 An article in The Wall Street Journal (April 27, 1992) indicated that the SEC has

taken steps to simplify the procedure to be followed by institutional stockholders or blockholders of a firm to communicate with each other Institutions are not required to file proxy documents in order to discuss issues of mutual interest outside the amiual stockholder meetings

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

relationship between corporate control variables and abnormal returns at the standstill agreement date Section 4 presents the empirical results and a discussion of how the results support the hypotheses and testable implications Section 5 states the conclusions and outlines future research avenues.

Bhagat and Jefferis (1987) argue that a conflict of interest between management and shareholders cannot explain the persistence of standstill agreements, greenmail and targeted repurchases The authors state that the failure of shareholders to deal with a standstill agreement and/or targeted repurchase through contracting suggests that these actions are not a pure transfer of wealth If they were, shareholders would contractually bond management not to engage in such practices While termination of a takeover threat through a standstill agreement and/or a targeted repurchase always benefits incumbent management through continued tenure, shareholders may also benefit If management forgoes any resistance to a takeover and ignores the use of a standstill agreement, stockholders are likely to accept a tender offer at a lower price than the true worth of the shares Shareholders apparently recognize that some standstill agreements and/or targeted repurchases are in their best interest while others are not, so they do not completely prohibit such actions.6 This decision process by shareholders is consistent with the theory developed by Leland and Pyle (1977) They show that managerial equity ownership conveys information to outside shareholders about management's private valuation of the firm By observing the proportion of insider ownership and voting control of the firm, shareholders can infer the motives and incentives of management in entering a standstill agreement.

Ruback (1988) examines stock prices of target firms involved in unsuccessful control contests and finds significant abnormal positive returns of 31 percent associated

6Management can easily prohibit greenmail or standstill agreements without legislation, announcing a policy that prohibits the board and management from taking such actions and abide by this policy

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

with the initial announcement of a control contest At the termination announcement, these firms realized statistically significant negative abnormal returns of about 10 percent No significant abnormal returns occur in the three years following termination o f the takeover offer Ruback concludes that negative termination announcement returns indicate that the failed control contest is costly to shareholders of target firms A failed contest also means that opposition to a takeover bid can be harmful because potential bidders may abandon the takeover attempt when target management resists O f course, potential bidders may instead choose to increase the offer price, thereby benefiting target shareholders Ruback argues that managerial opposition to tender offers is a gamble; the target firm may obtain another higher bid or not He concludes that losing the gamble imposes costs on shareholders Similarly, resistance to change-of-control through a standstill agreement and/or a targeted repurchase may not always be a negative Target shareholders benefit

if the firm 's value increases because of a change of control or other managerial actions after entering a standstill agreement Stockholders lose in this managerial gamble if no value-enhancing change occurs

M ikkelson and Ruback (1991) analyzed abnormal returns to stockholders of 111 observations from the initial accumulation of blocks of stock through subsequent targeted repurchases Thirty-three o f the observations were subject to a prior control event However, a tender offer for control or a merger proposal preceded only four targeted repurchases.7 A standstill agreement accompanied 39 of the repurchases They find an

7Mikkelson and Ruback (1991) include an observation as a control event when The

Wall Street Journal reports that a blockholder sought or considered seeking either board

representation or control

increase in the average stock price in response to the accumulation of shares and the announcement of the blockholding The average return is 7.4 percent from the time of the initial blockholding until the repurchase However, Mikkelson and Ruback argue that the average stock returns mask the heterogeneity among repurchases The authors find that the average stock price change for a repurchase with a standstill agreement is positive but statistically insignificant for the initial-accumulation-to-repurchase period Yet, the average abnormal return at an announcement of a targeted repurchase with a standstill agreement

is significantly negative The authors suggest that much of the cross-sectional difference between the abnormal returns is due to standstill agreements

Mikkelson and Ruback (1991) find no obvious pattern to the use o f standstill agreements There is no evidence of a higher incidence of standstill agreements associated with known corporate "raiders" as opposed to non-raiders The frequency o f change of control for the repurchasing firm in the three years following the repurchase is virtually identical for repurchases with and without a standstill agreement Thus, standstill agreements do not seem to target active investors who pose a threat to invest again, nor

do they successfully thwart changes in control Furthermore, the authors cannot explain why standstill agreements may harm shareholders of firms in which management repurchases shares

The persistent use of standstill agreements and the failure of capital markets to restrict the ability of management to enter such agreements suggests that an important function is served by these agreements The following discussion develops hypotheses and testable implications related to standstill agreement announcements

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

2.1 Convergence-of-Interest Hypothesis vs Management Entrenchment Hypothesis

Shleifer and Vishny (1986a) argue that standstill agreements and other approaches aimed at eliminating a bidder are not necessarily contrary to a shareholders' best interest hypothesis They show that management's defensive tactics can encourage a bidding contest, thereby increasing the expected takeover premium and improving shareholder welfare Management may encourage uninformed potential acquirors to bid by providing information necessary to realize a profit However, prior to revealing the private information, target managements generally require potential acquirors to sign a confidentiality and standstill agreement

An earlier study by Jensen and Ruback (1983) suggests that without the cooperation

of target management, bidders may encounter unwanted hurdles They argue that a target firm can benefit from delay in the release of private information to a potential acquiror to promote information acquisition by other possible acquirors In this way the target firm conceals a "white knight" in the hopes that another bidder will surface so that the target firm can capture a larger share of the takeover gains.8

Shleifer and Vishny (1986a) conclude that in a signaling equilibrium context share prices always fall after payment of greenmail, even when managers appear to be maximizing the long-run value of the firm There are at least two possible reasons for the decline in share prices First, the market may respond unfavorably to information that

information collection is costly and target firm management wants other potential acquirors to collect more information about its firm If the target firm releases the information that it has found a white knight, other firms may stop investing in the information collection process

market may have a negative reaction upon learning that the target firm entered into a greenmail agreement, a standstill agreement, or some other managerial action to resist a takeover.

Dann and DeAngelo (1983) note that when contracting is costless, managers and stockholders resolve their conflicts by designing contracts to ensure that management follows a stockholder wealth-maximization policy The possibility of conflict of interest remains when contracting is costly However, sufficient constraints exist to limit the extent

of management's deviation from stockholder wealth maximization Manne (1965) argues that the threat of a takeover motivates management to act in the best interest of non­controlling shareholders An implicit assumption of his argument is that managers do not take unilateral actions which blunt the takeover threat to the detriment of stockholders Williamson (1975) does not entirely agree and notes that managers act opportunistically

to reduce any takeover threat, thereby expanding their perquisite consumption possibilities

at stockholder expense Standstill agreements and privately negotiated repurchases are opportunistic actions that incumbent management takes to reduce the threat of a takeover

by a large blockholder

The convergence-of-interest hypothesis states that a standstill agreement with a new blockholder can increase firm value through more effective monitoring of management, through innovative ideas offered by a blockholder, or through takeover interest in the firm

by other investors The target management's decision to repurchase an existing blockholder's stake can help nonparticipating shareholders by facilitating a takeover by a

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

third party, by eliminating a trouble-making investor or by eliminating a managerial

impasse about corporate strategy At the other extreme is the management entrenchment

hypothesis It suggests that target management may use stock repurchase with a standstill agreement as a device to entrench itself by rejecting a higher-valued takeover bid

Stockholders do not know ex ante which firms will enter standstill agreements Nor

do they know if firms entering standstill agreements will eventually realize a value- enhancing change in control Stockholders only know that some standstill agreements are advantageous for them while others are not If stockholders believe that management is acting in their best interests by entering a standstill agreement, a positive announcement period abnormal return supports the convergence-of-interests hypothesis On the other hand if stockholders believe that management is entering a standstill agreement to entrench itself, the stock price should fall at the announcement date A negative abnormal return supports the management entrenchment hypothesis

Ownership structure of target firms is an observable variable Investors can use this information to infer the incentives motivating management in entering a standstill agreement and making other control related decisions A significant correlation between the ownership structure and the abnormal performance in the pre-announcement period, the announcement period and the post-announcement period provides support for the alternate hypotheses depending on the direction of the correlation Analysis of abnormal returns in the pre-announcement and post-announcement periods and their correlation with ownership structure and other observable variables may allow the formulation of a trading rule by an interested observer of the process

2.2 Other Hypothesis

Jensen and Meckling (1976) and Leland and Pyle (1977) develop theory to show how insider ownership can affect the value of a firm Brickley, Lease and Smith (1988), Morck, Shleifer and Vishny (1988a,b), Pound (1988), Shleifer and Vishny (1986a, b), and Stulz (1988), among others, have shown the impact of block ownership and institutional ownership on the value of a firm The following discussion reviews the relevant literature and develops hypotheses relating to the impact of ownership structure on a firm 's share price because of the announcement of a standstill agreement

2.2.1 Ownership Structure and Corporate Value

The generally accepted theoretical view is that stockholders of a firm are a widelydispersed and homogeneous group of relatively uninvolved absentee owners Berle andMeans (1932) argue that when managers own little o f the firm 's equity and shareholdersare too dispersed to enforce value maximization, management deploys corporate assets tobenefit itself rather than shareholders Thus, the firm 's performance depends on thedistribution of share ownership between managers and others Jensen and Meckling (1976)

>

provide a theoretical foundation of the relationship between corporate value and management ownership They divide shareholders into insiders (agent) and outsiders (principal) Inside shareholders manage the firm and have exclusive voting rights Outside shareholders have no decision making rights However, both classes o f shareholders receive the same dividend per share of stock held A principal-agent problem arises because the inside shareholder can augment the dividend stream by consuming additional

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

perquisites In this framework, insider owner-managers have incentive to adopt investment and financing policies that benefit them but reduce the payoff to outside stockholders According to Jensen and Meckling (1976), the costs of deviating from a value maximization strategy decline as management ownership increases As their stakes rise, managers pay a larger share of these costs and are less likely to squander corporate wealth.

According to the convergence-of-interest hypothesis, market value increases with management ownership As the proportion of shares owned by insiders increases, there

is greater alignment of insiders' interests with those of outside stockholders Insiders who own equity are less likely to enter a standstill agreement if they believe the agreement is not in their best interests Larger managerial holdings imply greater convergence of interests and a smaller likelihood of entering the agreement to entrench management In

a cross-section of firms entering standstill agreements, market response to the announcement of the agreement should be positively correlated to the level of insider stockholdings

Demsetz (1983) and Fama and Jensen (1983) point out offsetting costs of significant managerial ownership These researchers argue that when management owns

a small stake, it will act in shareholders' best interests based on signals received from the capital market [Easterbrook (1984), Rozeff (1982)], based on forces operating in the managerial labor market [Fama (1980)], and based on the threat of an outside takeover [Jensen and Ruback (1983), Manne (1965), M artin and McConnell (1991)] In contrast,

if management controls a substantial fraction o f the firm's equity, it may have enough voting power or influence to guarantee continued employment with the firm at attractive

salaries and may indulge in non-value-maximizing behavior The entrenchment hypothesis predicts that corporate assets can be less valuable when management is free of any controls

on its behavior The main motivation o f management in entering a standstill agreement may be entrenchment

Stulz (1988) offers a competing argument to the Jensen and Meckling (1976) view

He focuses on the importance of the takeover market for disciplining corporate managers Stulz states that the premium a hostile bidder must pay to gain control of a target firm increases as the fraction of equity owned by management increases However, the probability that the takeover will succeed decreases as equity owned by management increases When management owns a small fraction of the shares outstanding, it is more likely that a hostile takeover will succeed at a premium that is less than the maximum the

bidder is willing to pay, ceteris paribus As managerial equity ownership increases, the

probability of a successful hostile takeover for a given premium declines If management controls a sufficiently large fraction of votes and always opposes takeover attempts, the value of the outside shares is lowest because investors never make a tender offer Empirical evidence exists to support Stulz’s views Bradley and Kim (1985) show that tender offers for firms in which insider ownership exceeds 20 percent are rare Weston (1979) reports that no firm in which insiders own over 30 percent equity has ever been acquired in a hostile takeover Thus, Stulz's reasoning implies that there is a unique insider ownership value that maximizes the value of the firm This reasoning suggests a curvilinear relationship between the value of the firm and the fraction of shares owned by insiders

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

Morck, Shleifer, and Vishny (1988b) agree with Stulz that the firm's value changes

as a function of insider ownership However, they disagree as to the direction of change with ownership They argue that the relation between ownership and value depends on which force dominates over any particular range of managerial equity ownership Management's natural tendency is to allocate the firm's resources in its own best interests This decision may conflict with the interests of outside shareholders and may negatively impact the value of the firm However, as management's equity ownership increases, its interests are likely to coincide more closely with those of outside shareholders as pointed out by Jensen and Meckling (1976) Thus, increased insider ownership should have a positive effect on firm value

Morck etal ( 4 9 p i f -p«mtbut that the convergence-of-interests hypothesis predicts

that larger stakes should be associated v ^ h a higher market valuation o f the corporation Any prediction of the entrenchment hypothesis is less clear-cut The problem is that entrenchment is not only the consequence of voting power Some managers, by virtue of their tenure with the firm, status as a founder, or even personality, can entrench themselves with relatively small stakes Other managers in firms with large outside shareholders or an active group of outside directors may be only weakly attached to the job despite high ownership Even if higher percentage ownership allows deeper entrenchment, diminishing returns may occur before reaching the 50 percent ownership level Theoretical arguments alone cannot unambiguously predict the relationship between management ownership and market valuation of the firm's assets Thus, the relation between corporate value and ownership structure is an empirical issue

Morck et al (1988a) find a positive relationship between inside ownership and

Tobin's q in the zero to 5 percent ownership range, a negative and less pronounced relationship in the 5 to 25 percent range and a further positive relationship beyond 25 percent ownership They conclude that the conditions necessary for entrenchment increase with management ownership beyond 5 percent These conditions are not much different for firms with greater than 25 percent board ownership than they are for ownership in the

20 to 25 percent range A study by Hermalin and Weisbach (1991) finds that there is a non-monotonic relation between Tobin’s q and the fraction of stock owned by all present and former chief executive officers still on the board of directors However, their results

are different from those of Morck et al Hermalin and Weisbach find that the relation

between Tobin’s q and CEO stock ownership is positive between 0 percent and 1 percent, negative between 1 percent and 5 percent, positive between 5 percent and 20 percent, and negative thereafter

Demsetz and Lehn (1985) estimated a simple linear relationship between profitability and ownership by large shareholders (as opposed to ownership by management) and found no correlation Morck, Shleifer and Vishny (1988b) estimated their non-linear specification using the profit rate in addition to Tobin's q and found that

it is inappropriate to impose a simple linear structure on the data They failed to find support for Demsetz and Lehn's argument that ownership structure does not matter

McConnell and Servaes (1990) find a strong curvilinear relation between Tobin's

q and the fraction of shares owned by insiders The relationship is positive for low levels

of insider ownership and negative for high levels of insider ownership The curve is

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

maximized in the 38 to 49 percent range.9 This result broadly supports Stulz's (1988) theoretical arguments and is consistent with Jensen and Meckling (1976) and Morck, Shleifer and Vishny's (1988b) theories at low levels of insider ownership.

Jensen and Meckling's (1976) convergence-of-interest argument suggests that a positive correlation should exist between the price change resulting from the announcement effect of standstill agreements and the proportion o f insider ownership The likelihood of management entering a standstill agreement should increase with the level of insider shareholdings Stulz (1988) posits a theory which differs from Jensen and Meckling's (1976) theory at the lower end of management ownership The reason is that increased shareholder welfare from higher managerial ownership results from a more effective opposition to takeovers and not from better alignment of management and shareholder interests as suggested by Jensen and Meckling (1976) Thus, the coefficient for insider ownership regressed on cumulative abnormal returns at the announcement date of standstill agreements should be positive to support a convergence-of-interest hypothesis A negative coefficient supports the management-entrenchment hypothesis, as embodied in Stulz's theory His theory relates to the entrenchment hypothesis at the high end of insider ownership because high management ownership effectively precludes a takeover

M cConnell and Servaes (1990) also find a strong positive relation between q and shares held by institutional investors, but the relationship with blockholders is not significant

2.2.2 Blockholders as Monitors

Another aspect of the literature focuses on the pressure that equity blockholders and institutional investors can exert on management to force the firm toward value maximizing behavior Shleifer and Vishny (1986b) argue that takeovers can be successful only when the bidder previously acquires a large minority ownership in the firm The potential takeover threat that large blockholders can exert works as an effective device for monitoring management However, a shareholder owning a small proportion of the outstanding shares has little incentive to invest heavily in voting or to take actions to discipline management The shareholder is unlikely to affect the outcome In contrast, a shareholder owning a large block of stock has a strong incentive because more of the benefits from participating in the voting process and monitoring management are

internalized Shleifer and Vishny predict that, ceteris paribus, the presence of large-block

equityholders will cause managers to be less entrenched, which should result in a positive effect on the market value of the firm Monitoring by blockholders should be more effective as the percentage of equity they control increases Thus, it is less likely that management will enter a standstill agreement simply to entrench itself In this situation there should be a positive relationship between cumulative abnormal returns at the standstill agreement announcement and the degree of blockholder monitoring, as measured

by the block size

In McConnell and Servaes’ (1990) nonlinear regression of Tobin's q against ownership, they find that their results are inconsistent with the hypothesis that blockholders have an independent effect on corporate value They argue that their results

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

are consistent with the evidence of Holdemess and Sheehan (1988), Mikkelson and Ruback (1985) and Barclay and Holdemess (1991) When McConnell and Servaes combine block ownership with ownership by officers and directors, the coefficients in the regressions are significant in explaining the change in corporate value They argue that the effects of block ownership and inside equity ownership may interact in an undetermined way McConnell and Servaes further argue that their results are consistent with Shleifer and Vishny's (1986b) theoretical argument that blockholders are effective external monitors.

I extend the above arguments to the announcement effect o f standstill agreements

If blockholder monitoring is effective, management will enter an agreement only if management believes the agreement is in the best interest of the stockholders Hence the announcement effect of a standstill agreement should be positively correlated with the size

of the block ownership

2.2.3 Institutional Monitoring

Growth in institutional stock ownership raises interest about how institutional investors affect corporate voting Roe (1990) expresses some concern that institutions may

side with management by following the Wall Street Rule: Vote with management or when

sufficiently disgmntled, merely sell the stock He argues that legal and political constraints may prevent or discourage institutions from actively monitoring managers Such monitoring could restrict the activities or tax status of these institutions.10 These types of

10Other restrictions include proxy solicitation rules, disgorgement of short-term trading profits, double/triple taxation of profits if classified as a undiversified fund/institution by IRS, and so on See Roe (1990) for details

Vishny (1986a) suggest that cumulating shares into blocks creates stronger incentives for principals to monitor the activities of their agents and voting is one of the few observable shareholder monitoring activities Theoretical arguments suggest that institutions have greater incentives to invest in the voting process than a shareholder owning a small fraction of the firm 's shares because institutions control a large proportion of equity.

Brickley, Lease and Smith (1988) examine the relationship between the ownership structure of a firm 's common stock and votes on management initiated antitakeover amendments They present evidence indicating that institutional investors vote more actively on antitakeover amendments than do other shareholders and institutions oppose those proposals that appear to be harmful to shareholders Further, they document a strong positive correlation between the level of institutional ownership and the percentage of no­votes cast This result is inconsistent with the argument that the typical institutional owner

is a "rubber stamp" for management or is less critical of management than noninstitutional shareholders The evidence offered by Brickley, Lease and Smith suggests that certain institutional investors, such as mutual funds, endowments, foundations and public pension funds, are more likely to oppose management than are banks, insurance companies and trusts These latter institutions frequently have a conflict of interest; they derive benefits from lines of business under management control

Demsetz and Lehn (1985) find that institutional investors can benefit from their expertise in identifying undervalued assets These investors actively seek to increase their ownership of undervalued assets to take advantage of expected appreciation of the assets

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

and shares held by institutional investors They conclude that institutional ownership reinforces the positive effect of insider ownership on corporate value Pound (1988) investigates proxy contests He finds that large outside shareholders, particularly fiduciaries, are more likely to support management than to support dissidents in proxy contests and the probability that management prevails increases with the fraction of shares owned by institutional investors Pound concludes that the current proxy solicitation rules make institutional monitoring costly, more difficult, and less effective for dissident groups

as compared to target management The current legal environment favors incumbent management

According to Pound's (1988) efficient-monitoring hypothesis, institutional investors have greater expertise and can monitor management at a lower cost than can small atomistic shareholders When a target firm's management knows that institutional investors and blockholders are effective monitors, it is less likely to enter a standstill agreement to entrench itself In these circumstances, management is more likely to enter a standstill agreement to benefit the stockholders Hence, the announcement effect of a standstill agreement should be correlated positively to the level of institutional ownership, which is

a proxy for external monitoring Institutional investors and blockholders, as potential monitoring agents, may not be able to prevent managers from unilaterally entering a

standstill agreement These agents can punish management, ex post, if they perceive

managers to have entered standstill agreements to entrench themselves Punishment can take the form of a proxy contest, a takeover or removal of managers at a later date Klein

and Rosenfeld (1988) find evidence of such action Their study reports above-average management turnover within one year of a greenmail payment.

The conflict-of-interest hypothesis suggests that institutional investors are coerced into voting their shares with management because of other profitable business relationships with the firm For example, an insurance company may hold a significant portion of a corporation's stock and concurrently act as the corporation’s primary insurer Voting against management may significantly affect the firm's business relationship with incumbent management and others as well, whereas voting with management results in no obvious penalty Pound (1988) argues that this behavior may result from the current regulatory structure, which places no burden on fiduciary managers to vote or to disclose voting behavior to beneficial owners He finds support for the hypothesis that there is an inverse relationship between institutional ownership in a proxy contest target firm and the probability o f dissidents winning the proxy initiative Extending the conflict-of-interest argument of institutions to standstill agreement announcements, I argue that the level of institutional ownership should be correlated negatively to the announcement effect of standstill agreements

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

This section outlines the data and methodology Section 3.1 describes the sources

of data and approach used to identify firms entering standstill agreements There is also

a discussion of the details of the agreements, dates and periods for measurement of cumulative abnormal returns Section 3.2 describes the modeling of abnormal returns on the announcement day and the set of corporate control variables used to explain these returns Section 3.3 summarizes statistics of the data set

3.1 Identification of Sample and Measurement of Abnormal Returns

A text search of the Dow Jones News Retrieval Service (DJNR) for the period January 1979 to April 1990 identified all articles appearing in The Wall Street Journal (WSJ) and Dow Jones News Wire Service (DJNR) containing the words "standstill

agreement." The initial sample consisted of 295 domestic target firms potentially entering

or already having entered standstill agreements Whenever a WSJ or DJNR reference mentions a particular firm having entered a standstill agreement, I made every effort to identify the first public announcement in the financial press about the agreement If there was no public announcement about the existence of a standstill agreement at the time of the agreement or if the exact date of the announcement could not be determined, then I deleted the observation from the sample.11 This strategy resulted in the exclusion of 31

n For example, B F Goodrich made no public announcement of a standstill agreement when it repurchased the 4.9 percent stake owned by Carl Icahn on November

9, 1984, at a premium above the market price Carl Icahn promised not to buy any shares

in B.F Goodrich for five years and not to influence management during this period The firm did not make any public announcement about the targeted repurchase or standstill agreement The press release by B.F.Goodrich only indicated that the firm had purchased

a block of shares for general corporate purposes The first indication about Goodrich

observations I excluded an additional 25 observations because returns data were unavailable from CRSP/NASDAQ tapes The final sample consists o f 249 observations

o f standstill agreement events (involving 202 firms) The sample includes 32 cases of standstill agreements being discussed or negotiated but not finalized The news retrieval articles indicate that either the target firm had requested a standstill agreement with the potential bidder or the bidder had offered to enter a standstill agreement in return for confidential information

The first publication of news in The Wall Street Journal about the firm having entered into a standstill agreement is "day 0" or ED AY If only the Dow Jones Wire

Service reports the announcement of the standstill agreement, then this date is "day -1" or

EDAY-1 The event day for firms that mentioned the possibility of, but did not conclude,

a standstill agreement is the first disclosure of it in the financial press Calculation of the total announcement effect includes all standstill agreement announcements or possible agreements 90 days before and 90 days after EDAY I use a three-day window around each announcement to calculate announcement period abnormal returns This three-day period includes the announcement day and days immediately before and after the announcement Once having identified EDAY for each firm, I draw 1001 trading day stock returns from CRSP and NASDAQ returns files for the period EDAY-500 to EDAY+500

I partition the returns into four non-overlapping periods which require identification of two more important dates: the Schedule 13D filing date and the change-of-control date

having entered a standstill agreement came about a year later when the Securities and Exchange Commission censured the management for not making a proper disclosure about the standstill agreement and the greenmail payment to Carl Icahn

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

The filing of Schedule 13D by a bidder is an important event date, signifying that

a blockholder has reached the 5 percent or more ownership level of a firm ’s outstanding shares.12 The capital markets respond to the filing of Schedule 13D (DATE13D) with a positive stock price revaluation because of an increase in the probability of a value- enhancing change in control or because of increased monitoring Mikkelson and Ruback (1991) and Ruback (1988), among others, also find an increase in the stock price after the filing o f Schedule 13D For my sample, the average length of time between filing of the schedule and the subsequent EDAY is 134 days The definition of the change-of-control day (CCDATE) is the first public announcement of a successful agreement of change in control between the bidder and the target firm The average length of time between EDAY and a subsequent change of control of the firm is 350 trading days If dates for either a Schedule 13D filing or change of control are not available, I arbitrarily chose EDAY-134 and ED A Y +350 as the days, respectively I use the Lease, Masulis and Page (1989) model to calculate abnormal returns for the portfolio of firms entering standstill agreements Their approach, shown as equation 1, allows for a shift in beta to simultaneously estimate abnormal returns over different time periods, which I define using dummy variables

it i m t i l t l i2 t2 i3 t3 it \ *

12The Williams Act requires stockholders to report a 5 percent level of ownership

in a Schedule 13D filing within 10 days of its attainment

r it is the continuous compound return to security i on day t and r mt is the continuous compound return to the CRSP equally-weighted market index over day t The dummy variables dtl, dt2, and d^ equal one if t is in the nth event period (as shown in Table 1), and they equal zero otherwise The error term eit is i.i.d ~ N (0 , a) and E(eit,eit.)= 0 for t=£t' 0U, 0,2 and 0 i3 isolate the components of tire security's daily abnormal return caused

by the different event in periods

Table 1Definition of dummy variables corresponding to different time periods relative to the announcement of a standstill agreement _

Standstill agreement announcement period

EDAY-500 to DATE13D-2 or EDAY-136 to EDAY+ 500

d4 = l

Dummy variable dj equals one for the three-day window around EDAY and fordays that the financial press discusses the standstill agreement (announcement period).13Dummy variable d2 equals one for all days in the pre-announcement period (between

13The dummy variable d, equals 1 on three-day windows around all announcements and mention of a standstill agreement in addition to the EDAY This allows me to capture the total effect of the announcement of the agreement These results are compared with only the EDAY and ignore the other announcements Most of the announcement effect is due to the EDAY

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

DATE13D and the EDAY), whereas dummy variable d3 equals one for all days in the post-announcement period (between EDAY and CCD ATE) Dummy variable d4 equals one

in the period before the Schedule 13D filing date and after the change-of-control date The dummy variables dl5 d2, d3 and d4 are zero on all other dates The intercept <Xj in equation

1 captures the abnormal return component corresponding to d4

This model allows me to follow the suggestion by Mikkelson and Ruback (1991) that it is inappropriate to look at abnormal returns at the announcement of the standstill agreement to decide if target management acted in the best interest of all stockholders My period for analysis extends from the Schedule 13D filing day to the change-of-control day Calculating abnormal returns for this total period allows me to evaluate returns to an investor by subperiod from a corporate control event that begins with accumulation of a block of stock through a change in control

The total abnormal return to firm i over event period n is AR in = con0n, where con

is the number of days in the event period A R in measures the continuously compounded holding period abnormal return from investing in stock i over the nth-event period In a sample of K firms, the equally-weighted cross-sectional average abnormal return is

Equation 3 tests the statistical significance of AAR under the null hypothesis of zero abnormal return Assuming that the N non-contemporaneous events are independent, the test-statistic is

1 N 0

where oein is the standard error of 0 in Z is approximately standard normal for large samples when ordinary least squared estimates replace the true values of 0 in and a ein

3.2 Modeling Abnormal Returns Occurring on EDAY

The next step in the methodology is to analyze factors contributing to any abnormal

returns occurring during the three-day standstill agreement window I researched The Wall

Street Journal Index (WSJI) for a period of three years before and three years after the

standstill agreement (or until December 1994) for each company This search used the DJNR industry category code of “mergers, acquisitions, takeovers (tnm in.)” for all

articles published in The Wall Street Journal and Barron’s These articles provided

information about the standstill agreements and control activities and allowed identification

of several variables Table 2 summarizes the variables and indicates the expected sign for each variable’s coefficient based upon the discussion in Section 2 14

14Dummy variables representing existence of poison-pills, resolution o f lawsuits with standstill agreements and continuous variables representing ownership structure after the agreement were also regressed against the announcement effect There was no significant correlation observed

Reproduced with permission of the copyright owner Further reproduction prohibited without permission.

BODSEATS is a dummy variable that equals 1 if management promises +

board representation as part of the standstill agreement;

otherwise, it equals zero Management's willingness to allow board representation by the blockholder is akin to subjecting itself to monitoring by the new board member BODSEATS should have a positive announcement effect because of the additional monitoring

CNSSA is a dummy variable that equals 1 if the target firm enters a +

confidentiality and standstill agreement; otherwise, CNSSA equals 0 Target firm management provides private

information to the bidder to justify a higher asking price for the firm CNSSA should have a positive stock price effect on the event day

EXCHANG is a dummy variable that equals 1 if the target firm 's stock +

trades on the American Stock Exchange or over the counter; or otherwise, it equals zero There is no obvious reason why the market’s response to a standstill agreement announcement should differ for stocks trading on different exchanges