How corporate governance is made: the case of the golden leash.

• Author: Cain, Matthew D.
• Position: III. Empirical Analysis through Conclusion, with footnotes and appendices, p. 678-702

3. EMPIRICAL ANALYSIS

   The saga of the golden leash, the Wachtell Bylaw, and the ISS recommendation presents a natural experiment to test how investors value corporate governance. We begin, in Sections A and B, by analyzing the thirty-two companies that had adopted the Wachtell Bylaw to determine if there is any clear pattern to explain their actions. In Section C, we then conduct time series analysis of the share prices of the thirty-two companies around principal dates surrounding the adoption and repeal of the Wachtell Bylaw. Our goal is to discern if investors value these governance changes, and if so, whether they assign them negative or positive values. Finally, in Section D, we consider the effects of the Wachtell Bylaw on a wider set of firms subject to activist attack.

   1. Data Description

      We obtained from ISS a list of companies that adopted a Wachtell Bylaw on or after May 9, 2013, the date Wachtell first proposed a model director compensation bylaw. To confirm that no other corporations belong in this list, we searched 10-K Wizard for bylaw amendments addressing director compensation; we then cross-checked the companies against a similar list that Innisfree, a proxy solicitation service, had compiled and provided to us. We confined our search to the period from May 8, 2013, the date of the Wachtell Memorandum, until November 27, 2013, the date when Provident published the voting results for its annual meeting. We excluded companies that adopted the bylaw in conjunction with an initial public offering or as a prelude to being spun off from a parent company. The majority of companies adopting a bylaw during this time period used the Wachtell language. We arrived at a list of thirty-two companies.

      We also obtained from ISS a list of all companies that subsequently repealed their director compensation bylaw. We confirmed that there were no other repeals by searching the 10-K Wizard database, confining our search to any repeals implemented between November 26, 2013 and June 30, 2014. Our search returned a list of twenty-seven companies.

      We then searched the SEC EDGAR database to confirm the date that each company publicly filed its disclosure of an adoption or repeal. Incorporation data for each of these companies is obtained from the EDGAR database. Institutional Ownership information is obtained from Thomson Reuters while information on staggered boards and shareholder activism is obtained from FactSet SharkRepellent. (143)

      We measured activism by searching the FactSet SharkRepellent database for any proposal by an institutional shareholder for substantive changes to corporate organization or operations from May 8, 2012 (one year prior to the Wachtell Memorandum) through November 25, 2014 (the one year anniversary from the day before the Provident annual meeting). Voting results and majority voting policies are also obtained from the FactSet database as well as from filings in the SEC EDGAR database. Accounting information is obtained from Compustat. Stock price information is obtained from the Center for Research in Security Prices database and Yahoo! Finance. Compensation data is obtained from Execucomp. Institutional Ownership information is obtained from Thomson Reuters. Finally, information on staggered boards and shareholder activism is obtained from Factset SharkRepellent.

   2. Analysis--Adopters and Repealers

      In this subsection, we examine the characteristics of the thirty-two companies that adopted the Wachtell Bylaw. Table II, Panel A describes certain accounting and compensation data for these companies.

      There is no pattern in terms of size, income, or revenue linking the companies which adopt these bylaws. The median market capitalization of these companies as of year-end 2013 is $3.961 billion but the mean market capitalization is $7.162 billion, with a standard deviation of $9.939 billion, which indicates a significant right-skewed range of size for the companies in our sample. Similarly, the assets and revenue figures for the sample have a wide range which is right-skewed. Companies in our sample have, as of year-end 2013, median total assets of $3.423 billion. The mean asset size for companies in our sample is $16.884 billion and the standard deviation is $62.278 billion. The median revenue of these companies in fiscal year 2013 is $2.549 billion with a standard deviation of $5.732 billion and a mean of $4.076 billion.

      The majority of companies in our sample are profitable, and the median net income for fiscal year 2013 for these companies is $96 million. To further explore the profitability of these companies, we calculated their return on assets (ROA). ROA measures profitability relative to total assets by dividing net income by total assets. The average ROA in fiscal year 2013 for the companies in our sample is 2.3%, and the median is 4.2%, which is below the average for companies in the Russell 2000 index. (145) In our sample, the median chief executive officer (CEO) compensation for the last fiscal year prior to adoption of an amendment is $6.354 million, and the average is $7.182 million.

      Institutional ownership is one factor that may influence the adoption of director compensation bylaws. Low institutional ownership may provide companies leeway to take actions which might be disfavored by shareholders. If institutional ownership is low, there is no concentrated shareholder base which can act effectively to overcome the free rider problem of dispersed shareholder ownership. On the other hand, companies with a high number of institutional owners may wish to adopt these bylaws because they can prevent a small number of shareholders from supporting a shareholder activist. Panel B sets forth information on institutional ownership in our sample.

      The median institutional ownership for the companies in our sample as of year-end 2013 is 69.94%, showing a high level of institutional involvement. The maximum level of institutional ownership for a firm in our sample is 92.6%. In the sample, the median ownership of Top 5 Institutional Ownership--the amount held by the top five institutional holders--is 27.06%. There are no controlling shareholders in our sample and the maximum amount held by a single institutional holder for the median firm is 8.39%, with a maximum of 14% of the company. Panel C of Table II shows other aspects of corporate governance.

      Companies incorporated in Delaware have been found to trade at a premium as compared to other companies. (148) This may be due to self-selection bias; that is, better companies may opt to incorporate in Delaware. But, it also may be that Delaware companies have historically traded at a premium due to the better laws and corporate governance apparatus that Delaware arguably provides. (149) Approximately 47% of our sample (fifteen companies) is incorporated in Delaware.

      A staggered board has been found to reduce firm value around the time of its adoption. (150) For these and other reasons, it has been opposed by corporate governance advocates and proxy advisory services, and may be an indicator of poor corporate governance practices. (151) In recent years this has led many S&P 500 companies to "destagger" and repeal their staggered boards. As of 2013, only 9.75% of S&P 500 companies had a staggered board. (152) However, this number rises to 50.66% of the Russell 2000 and 41.96% of the Russell 3000. (153) In our sample approximately 56%, or eighteen companies, have a staggered board which is above all three metrics.

      One spur to adopting director compensation bylaws may be an ongoing or recent encounter with a shareholder activist. (154) We measured activism by searching the Factset SharkRepellent database for any proposal by an institutional shareholder for substantive changes to corporate organization or operations brought after January 1, 2012, but before release of the Wachtell Memorandum, with respect to the companies in our sample. We found that approximately 12.5% of our sample had an activist event under these parameters. (155) This is significantly higher than the general rates of activism for the S&P 500, Russell 2000, and Russell 3000 companies as measured by Factset SharkRepellent. Table III sets forth more in-depth information concerning shareholder activism and the companies in our sample for the time period from May 8, 2012 (one year prior to the Wachtell Memorandum) through November 25, 2014 (approximately one year after the shareholder vote for Provident Financial). (156)

      Notably, while these companies experienced a high rate of shareholder activism before their adoption of the Wachtell Bylaw, they also experienced shareholder activism at the same rate thereafter. This may mean that companies adopted the Wachtell Bylaw in anticipation of future activism. Still, based on the SharkRepellent data, there appears to be no discernable pattern in the adoption of the Wachtell Bylaw.

   3. Price Effects

      For each of the thirty-two companies--listed in Tables I and III above--that had adopted the Wachtell Bylaw, we examined the wealth effects of firm-specific decisions to adopt and repeal a golden leash bylaw. Table IV sets forth our cross-sectional event studies for companies adopting and repealing the Wachtell Bylaw.

      A cross-sectional event study examines the effect of share price movements for a set of firms over a single period of time. We conducted the analysis by calculating the excess returns of each company on the date of their adoption of a director compensation bylaw over the three day period from the date of adoption against the Russell 2000 index. (157) The three-day period includes the trading days before and after the relevant date. Excess returns are the amount by which a company in our sample had returns that were different than the Russell 2000 Index. We selected this index because it most comports with the characteristics of the companies in our sample. We hypothesized that if golden leash compensation arrangements...