Who opts out of state antitakeover protection?: the case of Pennsylvania's SB 1310.

Despite evidence that shareholders of takeover targets reap considerable abnormal returns (see, for example, Jarrell, Brickley, and Netter, 1988, and Jensen and Ruback, 1983), numerous states have enacted legislation intended to impede the takeover market. Between 1982 and 1990, more than 70 major

laws designed to restrict takeover activity were adopted by 35 states.(1) Perhaps the most prominent example of this type of legislation is Pennsylvania's Senate Bill 1310 (SB 1310), containing five provisions that make hostile takeover attempts prohibitively expensive. These provisions are (i) the stakeholder provision, (ii) the control share provision, (iii) the disgorgement provision, (iv) the severance provision, and (v) the labor contracts provision.(2) SB 1310 allowed Pennsylvania firms to opt out of (i) the disgorgement provision alone; (ii) the control share and severance pay provisions together; (iii) the disgorgement, control share, and severance pay provisions together; or (iv) all five provisions of SB 1310. To opt out, the board of directors was required to adopt a bylaw to the firm's articles of incorporation within 90 days of the statute's effective date, i.e., April 30, 1990. After this date, a shareholder vote was required to opt out.

The relatively complete obstruction of the market for corporate control and the board of directors' ability to opt out of the law's provisions provide a unique environment within which to examine three potential determinants of the opt-out decision. First, the presence of firm-level takeover defenses may affect the opt-out decision. If, for example, the board believes that takeover protection is important (as evidenced by the adoption of firm-level takeover defenses or the control of voting rights by insiders), it may prefer the additional protection of state antitakeover laws. Alternatively, if a board values an active market for corporate control, it may prefer to opt out of state antitakeover protection.

Second, several theories suggest that information asymmetry between managers and shareholders may affect the opt-out decision. Stein (1988) argues that this information asymmetry (which is likely to be higher for R&D-intensive firms) can cause managers to act suboptimally if they are forced to concentrate on current stock prices. It may, therefore, be optimal to protect managers of these firms from the market for corporate control. Hirshleifer and Thakor (1992) indicate that, in the presence of information asymmetry between managers and shareholders, managers are more likely to seek takeover protection to safeguard their human capital. Bizjak, Brickley, and Coles (1993) develop a model in which firms with high information asymmetry favor compensation contracts that focus on long-run stock returns to prevent managers from overinvesting or underinvesting. Consistent with their model, Gaver and Gaver (1995) report that managers of high-growth firms (i.e., high information asymmetry firms) derive a larger portion of their compensation from long-term incentive compensation. Overall, these theories suggest that firms subject to a greater degree of information asymmetry may be less likely to opt out of the law.

Third, institutional shareholder pressure may affect the opt-out decision. Szewczyk and Tsetsekos (1992) document that Pennsylvania firms lost approximately $4 billion during the law's adoption process. There was, however, a positive stock price reaction for firms announcing that they opted out of the law. Recognizing this potential wealth loss, the California Public Employees Retirement System (Calpers) sent letters to selected Pennsylvania firms in which it held an ownership position to "strongly encourage [them] to opt out" of the law. It is an empirical question, however, whether Calpers was successful in pressuring firms to opt out and whether it reduced its ownership in firms that did not opt out.

The results documented in this paper indicate that firms that opted out of SB 1310 were less likely to have a poison pill in place prior to the law's enactment and had lower insider control of voting rights. These findings are robust to differences in firm size and the monitoring activities of blockholders and outside directors. This result suggests that some boards value takeover defenses (whether at the firm or state level) to insulate themselves from the takeover market, while others firms choose to remain subject to an active market for corporate control.

Second, we find that firms that opted out spent less on R&D than firms that did not opt out, thus providing support for Hirshleifer and Thakor (1992) and Stein (1988). This result is also robust to differences in firm size, the level of insider voting rights control, and the existence of poison pills. We do not, however, find any significant differences in changes in R&D spending around the law's adoption.

Finally, we find that Calpers was not able to significantly influence the opt-out decision. Furthermore, Calpers did not significantly reduce its ownership in firms that it unsuccessfully pressured to opt out of the law. The insignificant change in ownership may not be surprising, however, given that approximately 80% of Calpers' investment portfolio is indexed and that sales decisions, on balance, are made by outside money managers.

This paper is organized as follows. We provide an overview of the Pennsylvania antitakeover law in Section I, describe the sample construction in Section II, present the empirical results in Section III, and offer concluding remarks in Section IV.

I. The Pennsylvania Antitakeover Law

The development of SB 1310 began in July 1989 when the Belzberg family of Canada announced its intention to acquire Armstrong World Industries of Lancaster, Pennsylvania. Almost immediately thereafter, the Pennsylvania legislature, at the behest of Armstrong's management, began to compose SB 1310. Within a year, on April 27, 1990, Pennsylvania Governor Robert P. Casey signed the bill into law. Its five provisions rendered it one of the most restrictive antitakeover laws in the country. The law did, however, provide a 90-day period after enactment during which a firm's board of directors could opt out of some or all of the law's provisions and, thus, relinquish some degree of protection provided by the state against hostile takeover attempts.

SB 1310's five provisions, which are designed to help firms resist hostile takeover attempts, significantly alter the relation between managers and shareholders of firms incorporated in Pennsylvania. The first provision authorizes directors to consider the short- and long-term impact of any proposed corporate change on all stakeholders affected by such actions. Stakeholders include, but are not limited to, shareholders, employees, customers, suppliers, creditors, and members of the local community. Directors, furthermore, do not have to weigh the impact of a change on one stakeholder group, including shareholders, more heavily than any other. This provision also reaffirms the applicability of the business judgment rule. Specifically, any decision regarding a change in control to which a majority of outside directors has assented is considered to be in the firm's best interest. Moreover, there is no greater burden of proof or justification required for change-in-control decisions than is necessary for any other board act. The final section of the first provision provides that only the corporation, or shareholders acting on behalf of the corporation, may bring suit against directors for breach of fiduciary duty.

The second provision, known as the control share provision, removes the voting rights of shareholders who control 20% or more of the firm's voting stock until both a majority of all shareholders and all disinterested shareholders vote to restore those rights. Disinterested shareholders are those who are neither the bidder nor the officers of the target. This provision, furthermore, specifies a minimum period of time that disinterested shareholders must own their shares before voting rights are awarded. The purpose of this section of the law is to preclude short-term arbitrageurs from participating in decisions that would ostensibly harm the firm's long-term shareholders. This section, moreover, provides the board with broad latitude in calling for a meeting to restore voting privileges.

The third provision, known as the disgorgement provision, allows the firm, or any shareholder acting derivatively for the firm, to sue a controlling person or group to seek the remission of all profits from sales of the firm's equity or equity derivative securities within 18 months after the acquisition of the controlling interest. A controlling person or group is a person or group that has "acquired, offered to acquire, or, directly or indirectly, publicly disclosed or caused to be disclosed the intention of acquiring voting power" of at least 20% of a firm's voting shares or "that announces that it may seek to acquire control of a corporation through any means." This provision does not distinguish between friendly and hostile tender offers nor does it distinguish between ultimate success or failure in acquiring a controlling interest.

The fourth provision provides for a lump-sum severance payment to employees released subsequent to a change in control. The maximum amount of the payment is 26 times an employee's average weekly compensation during the three months preceding termination. Finally, the fifth provision prevents firms from terminating or impairing most labor contracts subsequent to a change-in-control transaction.

Firms were permitted to opt out of all provisions of SB 1310, the control share provision alone, the disgorgement provision alone, or both. Since the fourth and fifth provisions are triggered by control share approval, firms opting out of the control share amendment are also exempt from those provisions. To opt out, the board of directors of existing corporations were required to adopt a bylaw to their articles of incorporation within 90 days of the effective date of the statute. After the 90-day period, a majority of shareholders had to vote to adopt an amendment to their firm's corporate bylaws to opt out.

II. Sample Construction

We identify a comprehensive sample of 142 NYSE-, ASE-, and NASDAQ-listed firms incorporated in Pennsylvania from Corporate Text/PC Plus COMPUSTAT. Nine utilities are eliminated, since utilities are heavily regulated and are not commonly subject to hostile takeover attempts. Two firms that were acquired before the enactment of SB 1310 are also eliminated from the final sample.(3)

We gather data on the percentage of voting rights controlled by officers, directors, and blockholders from proxy statements and on aggregate institutional ownership from the Standard and Poor's Stock Price Guides. In order to calculate the ownership of the top five institutions, as well as to obtain aggregate institutional ownership figures for a sample of 2,000 firms, we also obtain institutional ownership data from Compact Disclosure. Information on Calpers' equity portfolio is obtained from Calpers' annual reports. We classify blockholders either as non-insider blockholders (i.e., all blockholders except for executives, directors, and ESOPs) or as non-institutional blockholders (all non-insider blockholders minus institutional investors and trusts).

Board of directors data are also gathered from proxy statements. Board composition is measured using the procedure outlined by Gilson (1990), who classifies directors as insiders, independent-outsiders, or affiliated-outsiders. Insiders are directors who are members of the firm's current management team. Independent-outsiders have no personal or professional relationship with the firm other than in their capacity as directors. Affiliated-outsiders have a personal or professional relationship with the firm but are currently not managers.

We use proxy statements, The Wall Street Journal Index, and data provided by the Pennsylvania Chamber of Business and Industry to determine whether the firms in our sample opted out of some or all of SB 1310's provisions. We determine whether our sample companies had firm-level takeover defenses in place prior to the enactment of SB 1310 by searching The Wall Street Journal Index, proxy statements, the appendix to Jarrell and Poulsen (1987), and Corporate Control Alert. Accounting data are collected from COMPUSTAT.

In the following tests, we use all available data. Data are not available for 21 firms, which reduces the maximum sample size to 110. Moreover, sample sizes vary across tests since we lose observations as a result of the limited availability of proxy statements, institutional ownership information, and R&D data.

III. Empirical Results

This section describes the various empirical tests we performed and presents their results.

A. Univariate Analysis

We initially characterize our sample firms by examining selected firm attributes, focusing primarily on variables that influence takeover activity. The sample is divided into firms that opted out of some or all of the provisions of SB 1310 (opt-out firms) and those that did not opt out of any of the provisions (non-opt-out firms). Panel A of Table 1 shows that the 48 opt-out firms have median asset and sales values of $202 million and $234 million and the 62 non-opt-out firms have values of $53 million and $68 million respectively. Firms opting out of the law are, in general, larger and more likely to be listed on the NYSE or ASE than on the NASDAQ. The size difference is important because it suggests that tests of differences between opt-out and non-opt-out firms must control for firm size.

In Panel B of Table 1, we examine various measures that can influence the probability and success of a takeover. These measures suggest that there are no significant differences in the use of firm-level defenses (poison pills, supermajority provisions, and other antitakeover amendments) between opt-out and non-op-out firms. In addition to explicit firm-level takeover defenses, however, insider control of voting rights also affects the probability of receiving a bid, the premium offered, and the likelihood of success.(4) Stulz (1988), for example, posits that, as managerial control of voting rights increases, the probability of receiving a tender offer decreases, whereas the expected premium of any offer received increases. In fact, once insiders control 50% of the voting rights, a hostile acquisition cannot be completed successfully. In practice, however, insiders can successfully thwart a hostile takeover attempt even if they control less than 50% of the voting rights. Panel B shows that the voting rights of opt-out firms are less likely to be controlled by insiders, where insider control is a situation in which managers and directors control more than 25% of the firm's voting rights.(5,6) We also define a combined measure that is equal to one if the firm is controlled by insiders, has a poison pill, a supermajority provision, or some other type of antitakeover charter amendment and zero otherwise. This combined measure is also lower for firms that opted out. Overall, our univariate results suggest that firms without takeover defenses in place preferred to opt out of the law.

Panel C provides comparisons of other takeover-related firm attributes. The results indicate that there is not a significant difference in the frequency of golden parachutes between opt-out and non-opt-out firms. In addition, there is no evidence that firms subject to hostile takeover attempts prior to the law's adoption are more or less likely to opt out. A possible explanation for this result is that prior hostile takeover attempts are not a good indicator of the likelihood of future hostile takeover attempts. Firms may have adopted antitakeover amendments since the first unsuccessful attempt, thus making new attempts less likely.

Because of the importance of voting rights control, we analyze the percentages of voting rights controlled by insiders, non-institutional blockholders, and non-insider blockholders over the three years surrounding the adoption of SB 1310 in Table 2.(7) Insider control is important since it can significantly affect takeover probability and success. Voting rights control by blockholders is important because large shareholders have greater incentives to monitor [TABULAR DATA FOR TABLE 1 OMITTED] managerial actions, particularly in the event of a takeover (Shleifer and Vishny, 1986).(8)

We find that over all three years, insiders control a larger percentage of voting rights in non-opt-out firms. To determine if the difference in the level of voting rights control is driven by differences in firm size, we also tabulate the results for firms smaller or larger than the [TABULAR DATA FOR TABLE 2 OMITTED] median. The results for the subsets suggest that firms opting out have lower insider control of voting rights in both size categories, although the differences are no longer significant. The lack of significance within the size categories is most likely due to the small sample sizes. The mean and median percentages of voting rights controlled by the both non-institutional and non-insider blockholders (Panels B and C, Table 2) do not differ across opt-out and non-opt-out firms. This result also does not appear to be driven by differences in firm size. Thus, it appears that blockholders did not influence the opt-out decision.

Agrawal and Mandelker (1990), Brickley, Lease, and Smith (1988), and Jarrell and Poulsen (1987) report a negative relation between the adoption of antitakeover measures that are harmful to shareholders and the level of institutional ownership. Given the negative stock price impact of SB 1310, institutions may have encouraged firms to opt out. In Table 3, we present the level of and changes in institutional ownership from two years before to two years after the adoption of SB 1310. We find that there is no significant difference in the levels of institutional ownership between opt-out and non-opt-out firms; institutions do not appear to have influenced the opt-out decision. Given this lack of influence, we explore whether institutions divested their shares after the law's adoption. Mean institutional ownership in opt-out firms increases from 37.7% to 46.6%, and in non-opt-out firms from 34.4% to 37.1% during the five-year period around the law's adoption. These changes do not differ significantly between opt-out and non-opt-out firms and also do not differ significantly for firms that are smaller or larger than the median firm in the sample.

We perform two robustness checks in analyzing the level and changes in institutional holdings. First, we calculate the aggregate institutional ownership in a sample of 2,000 non-Pennsylvania firms over the five-year period surrounding the adoption of the law. The tests in Table 3 are then replicated by examining the level and changes in institutional holdings of the Pennsylvania sample and subsamples relative to the non-Pennsylvania sample. The results (not reported in the interest of brevity) indicate that the holdings, and changes in holdings, of institutions in Pennsylvania firms did not differ from those in non-Pennsylvania firms. Second, we examine the changes in ownership of the top five institutions over the period. The examination is motivated by the finding in Ambrose and Megginson (1992) that institutional holdings are concentrated in the top five institutions. The results (not reported) show that the ownership positions of the top five institutions do not change significantly over the period surrounding the adoption of the law. These robustness checks reinforce the conclusion that, in general, institutional investors did not reduce their holdings in Pennsylvania firms that did not opt out. One institutional investor that was openly critical of SB 1310 and did react to its adoption is Calpers. We focus on Calpers' reaction to the law in Subsection III, D.

[TABULAR DATA FOR TABLE 3 OMITTED]

There is some evidence that independent outside directors monitor managers and help align the interests of shareholders and managers (e.g., Hermalin and Weisbach, 1991; Lee, Rosenstein, Rangan, and Davidson, 1992; Shivdasani, 1993; and Weisbach, 1988). Since the board of directors controls the decision to opt out of the law, it is important to examine the relation between board composition and the decision to opt out. We present the results of such an analysis in Table 4. In Panel A, we document that firms opting out of SB 1310 have larger boards than those that do not opt out. When this sample is categorized into firms smaller or larger than the median, however, the differences are no longer statistically significant.

Although there is no significant difference in the percentage of independent-outside directors between [TABULAR DATA FOR TABLE 4 OMITTED] opt-out and non-opt-out firms (see Panel B), Panel C indicates that firms opting out of the law have a higher percentage of affiliated-outside and independent-outside directors. In the year prior to the law's enactment, for example, the percentage of affiliated-outside and independent-outside directors is 68% for firms that opt out and 62% for firms that do not opt out.

Independent-outside directors with significant reputational capital may be more likely to protect shareholder value. To investigate the effect of reputation, we classify independent-outside directors with three or more outside directorships as "professional" directors. The results in Panel D imply that opt-out firms have a higher percentage of professional directors than non-opt out firms. The tests based on the two size categories suggest, however, that this effect may be driven by size. The non-opt-out sample is dominated by smaller firms that typically have fewer professional directors.

Firm size is also correlated with several other variables of interest (e.g., ownership structure).(9) Thus, while the univariate statistics reveal important differences in firm-level takeover defenses and insider control of voting rights, it is important to analyze the interactions between the variables. For example, the board of a potential takeover target may not adopt an explicit takeover defense because insiders already control a significant proportion of its voting rights. Accordingly, we analyze the decision to opt out in a multivariate setting.

B. Multivariate Analysis

We estimate the following multivariate logistic regression model for the year of the law's enactment:

P(PA firm opts out) = f[control variables (i.e., firm size, control of voting rights by blockholders, outside directors on the board), presence of a poison pill, control of voting rights by insiders] (1)

We include control variables for blockholder voting rights control and the percentage of outside directors on the board, because as mentioned earlier, blockholders and outside directors can influence the takeover process. We include firm size in the regression model because prior research shows that larger firms are less likely to be takeover targets (Shivdasani, 1993), more likely to resist a tender offer, and less likely to be successfully acquired (Cotter and Zenner, 1994, and Raad and Ryan, 1995) and because there are systematic size differences between opt-out and non-opt-out firms.

Two independent variables are of particular interest. The first one is a dummy variable equal to one if the firm had a poison pill in place prior to the law's adoption and equal to zero otherwise. If the coefficient for the poison pill variable is negative, then our results indicate that firms with poison pills were less likely to opt out of SB 1310. The second variable captures the incremental effect of insider control of voting rights on the opt-out decision. We use two different measures of insider control. In model (i), we measure insider control of voting rights as the proportion of voting rights controlled by officers and directors, and in model (ii), we construct an indicator variable that equals one if insiders control more than 25% of the firm's voting rights, and zero otherwise. Negative coefficients for these variables imply that insider control of voting fights is associated with preference for state antitakeover protection.

We present the regression results in Section 1 of Table 5. Across all model specifications, our results indicate that larger firms are more likely to opt out. This result is consistent with the relation between firm size and the takeover process documented in prior research as well as the univariate results in Table 1. The proportion of voting rights controlled by non-institutional blockholders and the proportion of affiliated-outside and independent-outside directors fail to enter the regressions significantly. This lack of significance is also consistent with the univariate results in Tables 3 and 4.

The negative coefficients of the poison pill dummy variable indicate that firms with a poison pill in place were less likely to opt out of SB 1310. In addition, both variables used as proxies for insider control of voting rights are negative, even in the presence of the poison pill dummy variable. Thus, firms with a higher proportion of insider voting rights control were less likely to opt out. This result appears to be at odds with Pound (1992), who finds that insider ownership is higher for firms that opt out than for firms that do not opt out. Three factors can explain these differences. First, Pound uses insider ownership and not control of voting rights. Second, our sample does not include utilities, whereas he does not drop any particular industry. Finally, his comparisons are univariate, whereas we also perform a multivariate analysis of the decision to opt out.

C. SB 1310 and R&D

Pennsylvania lawmakers and boards of firms not opting out of SB 1310 argue that the removal of the threat of hostile takeovers allows managers to focus on the long-term interests of the corporation.(10) This view also has some academic support. Stein (1988), for example, argues that information asymmetry between managers and shareholders [TABULAR DATA FOR TABLE 5 OMITTED] [TABULAR DATA FOR TABLE 6 OMITTED] can cause managers to act suboptimally (myopically) if they are forced to concentrate on current stock prices. Since information asymmetry is high for R&D-intensive firms, it may therefore be optimal to protect managers of these firms from takeovers. Stein's (1988) model and his discussion (pp. 76-78) indicate that the absolute level of R&D spending is important in the managerial myopia scenario. Hirshleifer and Thakor (1992) also suggest that information asymmetry between shareholders and managers is important. In their model, however, managers are more likely to seek protection from the takeover market to safeguard their human capital.

The opt-out provision in SB 1310 allows us to provide direct evidence on the R&D intensity of firms that prefer takeover protection. We compare the mean and median values of R&D, standardized by total assets, for opt-out and non-opt-out firms in the five years around the law's enactment. We present the results in Table 6. These results are based on smaller sample sizes because R&D data are available for only 64 firms for the [-1,+1] window and for only 42 firms for the [-2,+2] window. Our comparison indicates that opt-out films spend approximately half as much on R&D as firms that did not opt out. Moreover, the average increase in R&D expenditures over the [-1,+1] window is marginally lower for opt-out firms than for non-opt-out firms. Over the [-2,+2] window, however, the changes in R&D expenditures are not significantly different between the two groups. These findings are in contrast to the finding of Meulbroek, Mitchell, Mulherin, Netter, and Poulsen (1990), who report that firms reduce R&D spending after adopting antitakeover charter amendments.

We also examine industry-adjusted R&D for the opt-out and non-opt-out samples in Panel B. Sample sizes are reduced for industry-adjusted R&D expenditures since COMPUSTAT does not provide industry R&D expenditures for 22 firms for which we have firm-specific R&D data.(11) The results show that there are no significant differences in either level or changes in industry-adjusted R&D expenditure between the opt-out and non-opt-out firms. The lack of significance for the industry-adjusted ratios is not, however, inconsistent with Stein (1988) and Hirshleifer and Thakor (1992); their models focus on firm-level information asymmetry, not relative to an industry benchmark.

We also examine the effect of R&D intensity on the opt-out decision in a multivariate setting in order to control for firm size. More specifically, we augment models (i) and (ii) in Section 1 of Table 5 with both firm and industry-adjusted R&D expenditures.(12) As shown in models (i) and (ii) of Section 2 of Table 5, the results of Section 1 are robust to the addition of standardized R&D expenditures. Moreover, the coefficient for R&D is negative and confirms that firms with high R&D intensity were less likely to opt out of the law, even when other firm attributes are held constant.

D. Calpers and SB 1310

Calpers is widely recognized as one of the most active and visible institutional shareholders. On May 11, 1990, Calpers sent letters to selected Pennsylvania firms to "strongly encourage [them] to opt out" of the antitakeover law, since the law would "be harmful to both Pennsylvania companies and their shareholders." Table 7 presents a list of the 30 sample firms in which Calpers held an equity position, provides the percentage of its equity ownership in these firms, indicates whether the firms received the Calpers letter, and shows whether the firms opted out.(13) The data suggest that Calpers did not send letters to some firms in which it held small ownership positions.

We analyze whether Calpers was able to affect the opt-out decision by estimating a multivariate logistic model similar to Equation (1). More specifically, we examine whether receipt of a letter from Calpers and Calpers' percentage ownership in the firm relate to the opt-out decision. Neither of these Calpers variables appears to explain the opt-out decision. Hence, we do not present the results in the paper for the sake of brevity. This result suggests, however, that Calpers did not affect the opt-out decision.(14)

We also examine changes in Calpers' investment in Pennsylvania firms around SB 1310's adoption. In Table 8, we present results for differences in the percentage of equity ownership by Calpers in our sample firms relative to Calpers' aggregate equity portfolio. The data in panel B suggest that over the short [-1,+1] window, Calpers reduced its ownership in firms that opted out of the law as compared to firms that did not opt out. There is not a significant difference between the two sets of firms, however, over the [-2,+2] window. Panel C shows that, as expected, Calpers had greater percentage ownership in firms to which it sent a letter than in those to whom it did not. Moreover, Calpers reduced its ownership in firms to which it sent letters significantly more than to firms to which it did not send letters. Finally, the results in Panel D indicate that there is a larger decline in the percentage of Calpers' ownership in firms to which it sent a letter and did not opt out. This difference is not significant, however, using t-tests as well as nonparametric tests. We also compare Calpers ownership in firms that received the letter and opted out with the firms that received the letter and did not opt out. We do not find a significant difference in the level or the changes in Calpers ownership between these two sets of firms over the [-2,+2] window.

Table 7. Calpers Ownership Around the Adoption of SB 1310

This table presents the list of sample firms in which the
California
Public Employees Retirement System (Calpers) owned equity in 1989.
Percentage of equity ownership in each firm for 1989 (year -1), the
year prior to the adoption of SB 1310, and 1991 (year +1), the year
after the adoption, are provided. A "Y" under the column labeled
"Letter" indicates that Calpers sent a letter to the firm
requesting
that it opt out of the law, and an "N" indicates that Calpers did
not send a letter. In addition, a "Y" in the last column indicates
that the firm opted out of the law, and an "N" indicates that the
firm did not opt out.

                      Percentage of ownership in the years
                       relative to the adoption of SB 1310

                              Year -1     Year +1   Letter   Opt
Out

 1 ALCOA                       1.2          1.2       Y         N
 2 AMP Corp.                   1.1          0.7       Y         N
 3 Armstrong World Ind.        1.0          0.5       Y         N
 4 Betz Labs                   2.0          2.4       N         N
 5 CDI                         1.6          1.6       N         Y
 6 Charming Shoppes            1.1          0.6       Y         N
 7 Check Pointe Systems        2.2          2.2       N         N
 8 Comcast                     0.1          0.6       Y         N
 9 Conrail                     1.2          0.6       Y         Y
10 Crown Cork & Seal Co.       0.0          0.0       N         Y
11 Deb Shoppes                 0.6          0.6       N         Y
12 Dravo Corp.                 0.2          0.0       N         Y
13 H.J. Heinz                  1.0          0.7       Y         Y
14 Equitable Resources         2.5          1.2       N         Y
15 Mellon Bank Corp.           0.5          0.7       Y         Y
16 Mine Safety Appliances      0.8          0.0       N         Y
17 Mylan Labs                  1.3          1.3       N         Y
18 Pep Boys                    2.9          0.9       N         Y
19 P.H. Glatfelter             0.3          0.4       Y         Y
20 PNC Fin. Corp.              0.7          0.6       Y         Y
21 PPG Industries              1.1          0.8       Y         N
22 Scott Paper                 0.7          0.7       Y         N
23 Sun Company                 0.5          0.6       Y         Y
24 Tasty Baking                1.5          1.5       N         Y
25 U.S. Healthcare             0.2          0.4       N         N
26 VF Corp.                    1.0          0.9       Y         Y
27 Weis Markets Inc.           1.2          1.2       N         Y
28 West Co.                    1.9          2.0       N         Y
29 Westinghouse Elec. Corp.    1.1          0.7       Y         Y
30 Zurn Ind.                   1.1          0.9       Y         N

Mean                           1.09         0.88

Median                         1.10         0.70

[TABULAR DATA FOR TABLE 8 OMITTED]

The insignificant differences may result from the small size of the sample of firms that received letters and did not opt out. Another explanation for this result is that a large fraction of Calpers' equity portfolio is indexed, making divestiture of specific stocks difficult. Indeed, the Calpers' annual report for 1992 suggests that 80% of Calpers' equity portfolio is indexed. Moreover, the report suggests that the remaining 20% is allocated to external money managers, and the composition of this portion of the portfolio is also not under Calpers' direct control.

IV. Conclusions

The ability of Pennsylvania firms to opt out of SB 1310, Pennsylvania's restrictive antitakeover law, allows us to document important determinants of this opt-out decision. Our results indicate that firms that opted out were larger, had lower insider control of voting rights, and were less likely to have had a poison pill in place. These results imply that firms with managers who are insulated from the external market for corporate control are more likely to retain the additional protection provided by SB 1310. In contrast, boards that value an active takeover market are willing to make the decision to opt out of the law. In addition, we document that monitoring by blockholders and outside directors did not seem to affect the opt-out decision. We also find that firms that opted out of the law spent less on R&D than firms that did not opt out. This result provides support for the proposition that managers of R&D-intensive firms seek antitakeover protection because of information asymmetries between themselves and the shareholders.

Finally, we also document that Calpers sent letters to most Pennsylvania firms in which it owned shares requesting that they opt out of SB 1310. The opt-out decision of Pennsylvania firms did not appear, however, to be influenced by Calpers' ownership or by the fact that the firm received a letter from Calpers. We also find that Calpers did not significantly reduce its ownership in firms that it unsuccessfully pressured to opt out. One possible explanation for this result is that indexing reduces Calpers' ability to selectively divest its holdings.

Most of the work on this paper was done while Sunil Wahal was a doctoral candidate and Ken Wiles was an Assistant Professor of Finance, both at the University of North Carolina. We thank Anup Agrawal, David Blackwell, Oyvind Bohren, James Cotter, David Denis, the Editors, Craig Holden, Jonathan Karpoff, Claudio Loderer, Gordon Phillips, Henri Servaes, Dennis Sheehan, Anil Shivdasani, Scott Smart, Karl Snow, Jeremy Stein, George Tsetsekos, Jim Wahlen, two anonymous referees, and seminar participants at the University of North Carolina, Indiana University, and the Financial Management Association annual meetings for helpful comments and suggestions. We also thank Carl Ackermann, Hilda Eubanks, Chet Friedman, and Shihchien Tang for providing valuable research assistance, and Michele Doherty from Corporate Control Alert for providing data on managerial entrenchment devices.

1 Indiana, North Carolina, Ohio, and Washington, for example, enacted antitakeover laws to protect Cummins Engine, Burlington Industries, Goodyear Tire and Rubber Co., and Boeing, respectively (see Karpoff and Malatesta, 1990).

2 We provide an overview of these provisions in Section I.

3 Compared with other papers that study SB 1310, our sample is the most comprehensive sample of firms affected by the law.

4 Cotter and Zenner (1994), Mikkelson and Partch (1989), Shivdasani (1993), Song and Walkling (1994), Walkling (1985), and Walkling and Long (1984), among others, provide evidence on the relation between managerial ownership and the likelihood of managerial resistance to takeover bids and takeover likelihood, premiums, and success.

5 Six firms in our sample have multiple classes of securities with different voting rights. Managers and directors, in general, own proportionally more of the securities with superior voting rights. We, therefore, measure voting rights instead of beneficial share ownership to determine the degree of insider control. Our subsequent empirical results are qualitatively unchanged if the percentage of common stock ownership by insiders is used instead of the percentage of voting rights controlled by insiders.

6 There is anecdotal evidence that insiders controlling Pennsylvania firms have responded differently to SB 1310's enactment. Salem Corp., for example, with controlling shareholder Victor Posner, did not opt out of SB 1310. This result is interesting since Holderness and Sheehan (1985) report 7% average gains to shareholders for 31 firms that were targets of block acquisitions by Victor Posner between 1977 and 1982. Conversely, Fischer & Porter, with controlling shareholder Jay H. Tolson, reincorporated in Pennsylvania in 1990, adopted an antitakeover charter amendment with provisions similar to the provisions of SB 1310, and eventually opted out of SB 1310. This anecdotal evidence suggests that individual firm characteristics may influence the response of controlling shareholders to the law. As a consequence, we explore these issues in greater detail in our multivariate analysis in Table 5.

7 For the tests in Tables 2 and 4, we require that three of the firm's proxy statements be available from Corporate Text/PC Plus COMPUSTAT or the SEC File: one dated after July 26, 1990 (event time = +1), the last date that Pennsylvania firms could opt out of the provisions of SB 1310; one dated at least one year before July 26, 1990 (event time = -1); and one dated between July 26, 1989 and July 26, 1990 (event time = 0). This requirement reduces the sample to 91 firms.

8 We also analyze compensation data for the top officers identified in the proxy statement and determine the present values of the compensation components according to Murphy (1985). We examine changes in the structure and level of compensation around the opt-out decision where the structure of remuneration contracts is estimated based on the ratios of cash compensation, option compensation, and any other compensation identified in the proxy statements to the sum of all compensation identified in the proxy statements. Our results (not reported in the paper) do not reveal differences in the level or structure of compensation contracts between opt-out and non-opt-out firms.

9 We also examine the relation between the ownership of independent-outside directors and the opt-out decision. These results (not reported in the paper) suggest that director ownership does not relate to the opt-out decision.

10 PPG's March 8, 1991, proxy statement illustrates this viewpoint. It contains the following response by the board of directors to a shareholder proposal to reincorporate the firm outside of Pennsylvania. The board of directors "determined that it was in the best interests of the Company to remain subject to all the provisions of the Act." The proxy also implies that the supporting statement of the shareholder sponsoring the amendment does not rely on studies focusing on the long-term effects of the Act but that "... the attention focused on such studies underscores one of the major problems the Act seems to alleviate: a misguided preoccupation with short-term price movements." In addition, William George, President of the Pennsylvania AFL-CIO, writes in the Pennsylvania AFL-CIO newsletter of June-July 1990, "The law discourages Wall Street raiders who have no intention of operating or building a business, but who seek to enrich themselves by manipulating stocks or breaking-up healthy companies to cash in on assets, jobs and employee pensions."

11 Industry average R&D expenditures provided by COMPUSTAT (based on four-digit SIC codes) are used to calculate industry-adjusted R&D ratios for the opt-out and non-opt-out samples. Specifically, the industry average R&D expenditure (divided by total assets) is subtracted from the firm's R&D expenditure. COMPUSTAT provides industry R&D data when the following conditions are satisfied: (i) 60% of the companies in the four-digit SIC code have R&D data and (ii) the "dominant" firm in the industry has R&D data (a dominant firm is defined as one for which the ratio of the firm's assets to the aggregate assets of the industry is greater than one divided by the number of firms in the industry). Firms with annual sales less than $50 million are excluded from the industry average.

12 The sample for these tests consists of 104 firms for which accounting, ownership, and board composition data are available. R&D figures 2 for year 0 are available for 71 of our sample firms. Following Holthausen and Larcker (1991), we set R&D expenditures equal to zero for the firms for which we do not have these data. We also estimate the models in Section 2 of Table 5 with these 71 observations only. These regressions yield qualitatively similar coefficients and significance for all variables, except for firm size, which loses significance. Our results are also qualitatively unchanged if we normalize our data with sales instead of total assets.

13 During the same period, Calpers owned equity in CoreStates Financial Corp., H.H. Robertson, Philadelphia Electric, Rorer Group, and SmithKline Beecham, which are also incorporated in Pennsylvania but not included in our sample. Each of these firms, except H.H. Robertson, received letters from Calpers requesting that the firms opt out of the provisions of SB 1310.

14 We also compute, but do not report, the abnormal returns for Pennsylvania firms around the date the letter was sent and find that the average abnormal returns do not differ significantly from zero. We do not find a significant difference between firms that did and did not receive the letter. Two potentially conflicting effects may influence our results. The market may perceive the letter as positive news that a large influential shareholder is monitoring management's opt-out decision, which may increase the probability that the firm will opt out. Conversely, the market may perceive the letter as a signal that Calpers was unable to induce management to opt out of the law in less formal negotiations and that the probability that Calpers will convince management to opt out is small.

References

Agrawal, A. and G. Mandelker, 1990, "Large Shareholders and the Monitoring of Managers: The Case of Antitakeover Charter Amendments," Journal of Financial and Quantitative Analysis (June), 143-161.

Ambrose, B.W. and W.L. Megginson, 1992, "The Role of Asset Structure, Ownership Structure, and Takeover Defenses in Determining Acquisition Likelihood," Journal of Financial and Quantitative Analysis (December), 575-589.

Bizjak, J.M., J.A. Brickley, and J.L. Coles, 1993, "Stock-based Incentive Compensation and Investment Behavior," Journal of Accounting and Economics (January/April/July), 349-372.

Brickley, J.A., R.C. Lease, and C.W. Smith, 1988, "Ownership Structure and Voting on Antitakeover Amendments," Journal of Financial Economics (January/March), 267-291.

Cotter, J.F. and M. Zenner, 1994, "How Managerial Wealth Affects the Tender Offer Decision," Journal of Financial Economics (February), 63-97.

Gaver, J.J. and K.M. Gaver, 1995, "Compensation Policy and the Investment Opportunity Set," Financial Management (Spring), 19-32.

Gilson, S., 1990, "Bankruptcy, Boards, Banks and blockholders: Evidence on Changes in Corporate Ownership and Control when Firms Default," Journal of Financial Economics (October), 355-388.

Hermalin, B.E. and M.S. Welsbach, 1991, "The Effect of Board Composition and Direct Incentives on Firm Performance," Financial Management (Winter), 101-112.

Hirshleifer, D. and A. Thakor, 1992, "Managerial Conservatism, Project Choice, and Debt," Review of Financial Studies (Vol. 5, No. 3), 437-470.

Holderness, C.G. and D.P. Sheehan, 1985, "Raiders or Saviors? The Evidence on Six Controversial Investors," Journal of Financial Economics (December), 555-579.

Holthausen, R.W. and D.F. Larcker, 1991, "Financial Performance and Organizational Structure," University of Pennsylvania Working Paper.

Jarrell, G.A., J.A. Brickley, and J.M. Netter, 1988, "The Market for Corporate Control: The Empirical Evidence Since 1980," Journal of Economic Perspectives (Winter), 49-68.

Jarrell, G.A. and A.B. Poulsen, 1987, "Shark Repellents: The Effects of Anti-takeover Amendments Since 1980," Journal of Financial Economics (September), 127-168.

Jensen, M. and R.S. Ruback, 1983, "The Market for Corporate Control: The Scientific Evidence," Journal of Financial Economics (April), 5-50.

Karpoff, J.M. and P.H. Malatesta, 1990," PA Law: State Antitakeover Laws and Stock Prices," Financial Analysts Journal (July-August), 8-10.

Lee, C.I., S. Rosenstein, N. Rangan, and W.N. Davidson III, 1992, "Board Composition and Shareholder Wealth: The Case of Management Buyouts," Financial Management (Spring), 58-72.

Meulbroek, L.K., M.L. Mitchell, J.H. Mulherin, J.M. Netter, and A.B. Poulsen, 1990, "Shark Repellants and Managerial Myopia: An Empirical Test," Journal of Political Economy (October), 1108-1117.

Mikkelson, W.H. and M.M. Partch, 1989, "Managers' Voting Rights and Corporate Control," Journal of Financial Economics (December), 263-290.

Murphy, K.J., 1985, "Corporate Performance and Managerial Remuneration: An Empirical Analysis," Journal of Accounting and Economics (April), 11-42.

Pound, J., 1992, "On the Motives for Choosing a Corporate Governance Structure: A Study of Corporate Reaction to the Pennsylvania Takeover Law," Journal of Law, Economics and Organizations (October), 656-672.

Raad, E. and R. Ryan, 1995, "Capital Structure and Ownership Distribution of Tender Offer Targets: An Empirical Study," Financial Management (Spring), 46-56.

Shivdasani, A., 1993, "Board Composition, Ownership Structure and Hostile Takeovers," Journal of Accounting and Economics (January/April/July), 167-198.

Shleifer, A. and R.W. Vishny, 1986, "Large Shareholders and Corporate Control," Journal of Political Economy (June), 461-488.

Song, M.H. and R.A. Walkling, 1994, "The Impact of Managerial Ownership on Acquisition Attempts and Target Shareholder Wealth," Journal of Financial and Quantitative Analysis (December), 439-457.

Stein, J., 1988, "Takeover Threats and Managerial Myopia," Journal of Political Economy (February), 61-80.

Stulz, R.M., 1988, "Managerial Control of Voting Rights: Financing Policies and the Market for Corporate Control," Journal of Financial Economics (January/March), 25-54.

Szewczyk, S.H. and G.P. Tsetsekos, 1992, "State Intervention in the Market for Corporate Control: The Case of Pennsylvania Senate Bill 1310," Journal of Financial Economics (February), 3-23.

Walkling, R.A., 1985, "Predicting Tender Offer Success: A Logistic Analysis," Journal of Financial and Quantitative Analysis (December), 461-478.

Walkling, R.A., and M. Long, 1984, "Agency Theory, Managerial Welfare and Takeover Bid Resistance," Rand Journal of Economics (Spring), 54-68.

Weisbach, M.S., 1988, "Outside Directors and CEO Turnover," Journal of Financial Economics (January/March), 431-460.

Sunil Wahal is Assistant Professor of Finance at the Krannert Graduate School of Management, Purdue University, West Lafayette, IN. Kenneth W. Wiles is with The Development Group Inc., Salisbury, NC. Marc Zenner is Assistant Professor of Finance at The Kenan-Flagler Business School, The University of North Carolina at Chapel Hill, Chapel Hill, NC.