Can Institutional Investors Improve Corporate Governance Through Collective Action?

Transcription

1 Can Institutional Investors Improve Corporate Governance Through Collective Action? By Craig Doidge, Alexander Dyck, Hamed Mahmudi, and Aazam Virani May 2014 Abstract Can institutional investors generate sufficient power through collective action to drive improvements in governance? We use proprietary data on the private communications of a coalition of Canadian institutional investors to address this question. Firm-level regressions show that private engagements by the coalition influence firms adoption of shareholder democracy measures, say on pay advisory votes, and improve compensation structure and disclosure. Spillovers from engaged firms to non-engaged firms and activities to influence regulations and widely publicized governance scores, suggest a broader impact beyond the engaged firms. The coalition s collection of hard information on governance practices and soft information on firms concerns collected through private communications facilitated their successes. The coalition has less impact on controlled corporations and focuses on different governance mechanisms than activist hedge funds. Doidge, Dyck, and Virani are at the Rotman School of Management at the University of Toronto. Mahmudi is at the Price College of Business at the University of Oklahoma. We thank the Canadian Coalition for Good Governance for making their information available to us and Matt Fullbrook and the Shareholder Association for Research and Education (SHARE.ca) for providing data. David Beatty, Stephen Erlichman, Ed Iacobucci, and Wayne Kozun provided useful comments. Mary Lou, Ashley Newton, and Livia Nguyen provided excellent research assistance. We are grateful to the Rotman International Centre for Pension Management for financial support.

2 With more effective corporate governance, insiders commit to limit extraction of private benefits and in return, the firm has higher valuations and better access to external finance. Thus, it is important to understand which mechanisms play an important role in monitoring insiders and improving governance. Shleifer and Vishny (1997) argue that investors power relative to insiders is crucial and conclude that two approaches to governance dominate legal protections for investors and ownership by large investors. With this view, institutional investors are not important for corporate governance. The reason is that individual institutions typically hold small stakes and without coordination lack real power and face collective action disincentives to invest in efforts to improve firms governance. To address governance weaknesses and overcome the free rider problem, investors face numerous obstacles. Costs of collective action, which increase with group size (Olson (1965)), include the costs of gathering information, creating common knowledge, and coordinating actions. In addition, investors cite legal risk as an impediment to coordination (Black (1998) and McCahery, Sautner, and Starks (2013)). Consistent with the importance of these costs, recent theoretical research emphasizes the threat of exit by institutions as an alternative to monitoring (e.g., Admati and Pfleiderer (2009) and Edmans and Manso (2011)) and empirical evidence shows that the ease with which institutional investors can communicate and coordinate is an important determinant of their governance role (Bradley, Brav, Goldstein, and Jiang (2010) and Huang (2013)). However, little is known about the specific mechanisms institutional investors use to coordinate actions and overcome collective action costs or the extent to which coordinated investors can improve firms governance. One example of limited coordination, the Council of Institutional Investors Focus List of underperforming firms provides mixed results (e.g., Opler and Sokobin (1996) and Song and Szewczyk (2003)). In this paper we examine whether improved approaches to collective action can increase institutional investors power and allow them to play an important role in influencing the governance choices of a broad group of firms. We focus on Canada over the period from 2003 to 2012 and exploit the fact that legal changes facilitated communication amongst institutional investors, spawning the creation of the Canadian Coalition for Good Governance (CCGG) in The CCGG took a leading role in collecting information, creating common knowledge, and coordinating action to implement changes in governance in firms and in governance rules and institutions. To track its influence, we take advantage of access 1

3 provided to us of records of non-public communication from the CCGG to firms, including letter writing campaigns, phone calls, and engagements between the CCGG and firms. Prior studies recognize that much investor activism is done through private rather than public channels, and without this information it is difficult to assess the activists potential impact (e.g., Carleton, Nelson, and Weisbach (1998) and Becht, Franks, Mayer, and Rossi (2008)). There is considerable anecdotal support for the CCGG s importance. US activist and governance expert Ira Millstein (2006) called the CCGG an example for the rest of the world to emulate, and in 2008, the Global Proxy Watch described it as a powerhouse in Canada and global model of collective investor activism. Does the evidence match the rhetoric? We use proprietary data on the timing of the CCGG s private engagements with firms and the issues raised during these engagements to show that the CCGG plays an important role in influencing firms governance choices. Importantly, none of the governance changes advocated by the CCGG at the time of the engagements were required by corporate law, listing rules, or by securities regulators. During our sample period, the CCGG focused on three main governance issues that it considered important: majority voting for directors, say on pay, and compensation structure and disclosure. In general, improving internal governance creates value (Cunat, Gine, and Guadalupe (2012)). Prior research also finds that majority voting and say on pay can improve governance and are valued by shareholders (e.g., Ertimur, Ferri, and Oesch (2013), Cai and Walkling (2011), and Ferri and Maber (2013)) though this conclusion is not unanimous (e.g., Cai, Garner, and Walkling (2013) and Larcker, Ormazabal, and Taylor (2011)). Similarly, clawback provisions in compensation contracts are an effective governance mechanism and are associated with higher valuations (e.g., Iskandar-Datta and Jia (2013)). Using performance peer groups to filter out the common factor of firms performance measures to ensure that executives are not paid for luck has long been advocated (e.g., Holmstrom (1982)) and recent evidence shows that contracts with relative performance evaluation improve executives incentives (Bettis, Bizjak, Coles, and Young (2014)). In any case, the CCGG, whose members include pension funds and asset managers, is more concerned with broad, long-term improvements in governance and value compared to the more immediate price reactions often sought by other types of activist investors. 2

4 Our first set of tests focuses on majority voting, which the CCGG views as a critical component of basic governance principles (CCGG Second Annual Report). In , the CCGG initiated a campaign of active engagements and letter writing that targeted 42% of the firms in the S&P/TSX Composite index (the major stock index of the Toronto Stock Exchange). In this campaign, the CCGG primarily targeted firms in which its members held the most dollar stakes. After controlling for other factors that could influence firms adoption decisions, including other external pressures and firm characteristics, we find that CCGG engagement had a statistically significant and economically meaningful impact on the likelihood of subsequent adoption of majority voting. Including a full set of controls (limited controls), we find that CCGG engagement increased the probability of subsequent adoption during the next two years by 16% (30%). We next focus on the CCGG s engagements from 2008 to In these engagements, the CCGG took a new approach and in 104 cases (13% of firm-years), directly engaged firms in organized meetings between CCGG staff, leading CCGG members, and firms board members. Reflective of a desire to increase investor power, management was not invited and independent board members were rarely joined by management, legal counsel, or investor relations staff. In approximately half of these engagements (48/104) the CCGG informed firms of their preference for the firm to introduce say on pay, a mechanism that speaks both to shareholder democracy and a concern about compensation policies. The CCGG also raised concerns about compensation structure and disclosure, focusing on capping pension payouts, introducing clawback provisions, and using performance peer groups ( compensation policies ) in 46 of these engagements. In this campaign, the CCGG again targeted firms in which its members held larger stakes, but also focused on firms with worse performance and those not controlled by insiders. Adoption rates of say on pay for engaged firms are significantly higher than those for non-engaged firms (61% vs. 9%). Similarly, 70% of engaged firms adopted at least one of the compensation policies advocated by the CCGG compared to 13% of the non-engaged firms. Although these differences are large, attributing them to the influence of the CCGG faces two potential concerns: firms that adopted would have done so absent CCGG engagement and/or they did so because of external pressures other than the CCGG. As with majority voting, we estimate logit regressions that control for other external pressures and firm characteristics. After including these controls, we find strong evidence that CCGG engagements 3

5 are associated with subsequent adoptions. With a full set of controls (limited controls) engagement increases the probability of subsequent adoption of say on pay by 9% (25%). The specific timing of the CCGG s engagements and firms subsequent adoptions rules out reverse causality as an explanation for the results. Although we include a broad set of control variables, we cannot rule out the possibility that unobservable differences between engaged and non-engaged firms would lead engaged firms to adopt the governance changes, absent engagement by the CCGG. Of course, if this was the case, it is not clear why the CCGG would spend its limited resources engaging these firms in the first place. Nonetheless, we verify that our results hold after controlling for unobservable differences between firms engaged by the CCGG and non-engaged firms. Our evidence shows that the CCGG has a statistically and economically significant impact on the governance of engaged firms. However, the CCGG s goal is to influence a broad set of firms and in any year, engaged firms are a small fraction of all publicly-traded firms. Thus it is important to assess the extent to which the CCGG has a broader impact beyond the firms it directly engaged. One channel through which governance practices spill over across firms is through board interlocks (Bouwman (2011)). Each year, we identify interlocks between directors of engaged and non-engaged firms. We find that non-engaged firms with overlapping directors are more likely to adopt than other non-engaged firms and this effect is stronger when the overlapping directors are more influential. The CCGG employs two other principal strategies to more broadly influence firms governance practices. It participates in the evolution of public policy and it crafts and broadcasts good governance guidelines. Its greatest success in shaping public policy culminated in 2014 when the TSX made majority voting a listing requirement. The rule adopted by the TSX was similar to the model policy the CCGG encouraged firms to voluntarily adopt starting in The CCGG s strategy of shaping norms through the creation and dissemination of good governance guidelines and related activities is similar to an informal regulatory approach. The CCGG took active steps to encourage measurement of governance practices against principles and broadly disseminated information on the costs and benefits of adopting governance practices. The CCGG s executive director succinctly summarized this approach to the 4

6 CCGG s board: measurement + best practices + exposure = board behavioural change. 1 Consistent with this mechanism, we document an association between changes in CCGG principles and subsequent changes in publicly reported governance scores, which created potential reputational rewards for following CCGG best practices or penalties for not doing so. Taken together, the evidence from the CCGG s private engagements combined with its broader, more public efforts to influence public policy and develop good governance guidelines, suggests that the CCGG plays an important role in improving firms governance. Finally, we identify some limitations to improving firms governance through a collective action organization. The CCGG has less success in improving the governance practices of controlled corporations and regulatory approaches may be necessary to improve the governance of these firms. Compared to hedge fund activist investors, the CGGG focuses on structural and process reforms and avoids issues such as changes in business strategy or firm sales that lead to higher returns (e.g., Brav, Jiang, Partnoy, and Thomas (2008)). Firms also tend to respond more slowly to the CCGG compared to hedge funds. As part of its engagement strategy, the CCGG avoids publicly criticizing firms and generally does not pursue more aggressive actions such as shareholder proposals or proxy fights, which limits explicit penalties for firms that fail to respond. Our paper contributes to the literature on institutional investor activism and governance more broadly (see Gillan and Starks (2007) and Yermack (2010) for reviews)). Prior research that provides evidence of the monitoring role played by institutional investors (e.g., Hartzell and Starks (2003), Chen, Harford, and Li (2007), and Aggarwal, Erel, Ferreira, and Matos (2011)) typically focuses on total institutional ownership, ownership concentration, or ownership by different types of institutional investors, but cannot identify specific coordination mechanisms. Recent research by Huang (2013) and González and Calluzzo (2013) infers when institutional investors are more likely to coordinate or identifies cases when activists target the same firm. Both studies conclude that there are benefits to coordination. Our analysis of the CCGG, a collective action organization of investors designed to lower the costs of activism, shows that it is possible to improve governance at targeted firms and to promote market-wide adoption of such improvements. At the same time, there are limits to what such an 1 Report to the Board of Directors of CCGG,

7 organization can accomplish and there are comparative advantages of different types of mechanisms to facilitate investor power. Finally, the CCGG s preference for, and the apparent importance, of its private communications with firms is consistent with prior evidence from the US (Carleton, Nelson, and Weisbach (1998)) and the UK (Becht, Franks, Mayer, and Rossi (2008)). Overall, the results fit with a sophisticated law and finance view that emphasizes the power of legal traditions to create room for investors to protect themselves. The paper is organized as follows. Section 1 provides a framework and context for our empirical analysis. Section 2 describes the primary governance changes sought by the CCGG in their private engagements and the private data we exploit in our empirical tests. In Section 3 we present our main empirical results. In Section 4 we consider the CCGG s efforts to influence regulation and governance principles to improve governance for all firms. In Section 5 we provide a discussion on the limits of institutional investor power. We conclude in Section Challenges to Collective Action, the Formation of the CCGG, and its Governance Campaigns 1.1. Challenges to Collective Action by Investors Institutional investors have an interest in corporate governance actions that will increase value, but it is easy to understand why Shleifer and Vishny (1997) do not accord them a prominent place in their discussion of governance mechanisms. As a group, their holdings are large, but individually institutional investors typically have small holdings. Thus, there is a free rider problem. An individual investor bears all of the costs of governance engagements but reaps only a small percentage of the benefits, leading to limited engagement activity in equilibrium (e.g., Admati, Pfleiderer, and Zechner (1994)). As Olson (1965) states, The income of the corporation is a collective good to the stockholders, and the stockholder who holds only a minute percentage of the total stock, like any member of a latent group, has no incentive to work in the group interest. Specifically, he has no incentive to challenge the management of the company, however inept or corrupt it might be. A second challenge to collective action by investors is potential litigation risk. In the US, for example, when groups of shareholders act together and hold more than 5% of the shares, they are required to disclose their plans by filing form 13D with the SEC. This 6

8 filing makes their efforts public and raises the risk of a lawsuit on the completeness of their disclosure of their plans (Black (1998)) Creation of the CCGG as the Voice of the Shareholder The combination of the free rider problem and legal concerns help explain why prior to the 2000s there was no effective collective action organization for investors in Canada. Governance activity was undertaken by individual investors, or was guideline based with a comply-or-explain requirement, with the most notable example being the Dey Report (1994). 2 As in the US, important reasons for inaction by investors were concerns that meetings and communications between investors might be considered a solicitation or a joint action by securities regulators which would invoke public disclosure obligations and potentially costly public formal solicitation. The grounds for collective action by investors in Canada changed in 2001 when shareholder communication was facilitated by a change in proxy solicitation provisions in the Canada Business Corporation Act (CBCA) that removed doubts that communications between investors would constitute a solicitation. 3 The CCGG was formed shortly thereafter in May 2003, and the press release announcing its formation cited this regulatory change. The use of the CCGG as the vehicle to advocate for governance changes differs significantly from the approach of investors in the US and the UK. In Table 1A we list the largest institutional owners of Canadian firms included in the S&P/TSX index and identify which institutions are members of the CCGG. From the beginning, the CCGG membership included institutions from a broad cross section of investor types, ranging from defined benefit pension plans (public and corporate), to Canadian bankaffiliated asset managers, and mutual funds domiciled in Canada and focused on the Canadian equity market. It started with 16 members in 2003 and by the end of 2005 it had 47 full members with about $950 billion in assets under management. The Chairs of the CCGG during our sample period were both CEOs of leading pension plans and the largest institutions were represented on the CCGG s board from 2 The TSX required firms to disclose their compliance with 14 of the recommendations in this report. See Shiplov, Greve, and Rowley (2010) and Anand, Milne, and Purda (2012) for further details. Anand, Milne, and Purda (2012) report that these recommendations shifted governance practices prior to the CCGG activity analyzed in this paper. 3 These changes came into effect on November 24, 2001 through changes to the CBCA rules on proxies, solicitations, and shareholder proposals. Many of these reforms followed earlier changes introduced by the SEC in the US in

9 the beginning. The size of these organizations, their long investment horizons, and their significant liabilities denominated in Canadian dollars, make them likely candidates to invest in governance reform efforts. By contrast, in the US, Black (1998) states: Coordinated shareholder activism is rare. Instead, each institution acts as a lone wolf. The collective activity that does take place occurs via an industry association that represents one type of investor. For example, the Council of Institutional Investors largely represents the interests of pension plans. In the UK, Black and Coffee (1994) state that industry associations dominate. Activities of groups that cut across investor types are limited to public policy efforts rather than engagements with individual firms. The CCGG brings together many of the largest institutional owners of Canadian equities into one organization. Figure 1A shows CCGG members combined percentage stakes of Canadian firms in the S&P/TSX index in To produce this figure, we use data provided by the CCGG on its members holdings. CCGG members own an average stake of 15% in S&P/TSX firms. In 43% of the firms, CCGG members own 20% or more. Figure 1B shows that CCGG member stakes are slightly higher in the larger firms that make up S&P/TSX 60 index. The average ownership in these firms is 17%. Interestingly, there is no substantial difference in members average holdings of controlled corporations (corporations with multiple voting shares, or companies in which one individual or group controls 30% of the votes, which made up 35% of firms in the index in 2005). In both groups, the average CCGG stake is 14%, albeit with greater dispersion in holdings for controlled corporations. Overall, the CCGG can legitimately claim to represent a substantial ownership percentage of Canadian firms. While the CCGG speaks for many of the most important institutional investors in Canada, it does not speak for all institutional investors. Institutional investors with ownership stakes in S&P/TSX firms, but who are not CCGG members collectively own average stake of 27% in these firms. These institutions also have somewhat different investment preferences compared to CCGG members as they invest relatively more in cross-listed firms and relatively less in controlled corporations The CCGG s Efforts to Improve Governance The CCGG followed three strategies to improve governance. The first strategy, common to institutional investors in other settings, is to work in the policy arena and coordinate investors efforts to 8

10 influence rules and regulations. The second strategy is Broadcasting our guidelines to the people and public companies of Canada through our website, the media, and many professional organizations. This activity involved crafting governance guidelines and identifying best practices. The first CCGG guidelines focused on building high performance boards. These guidelines pushed for further reforms beyond those in the Dey report. 4 By 2004 the CCGG stated it had become a point of reference for companies wishing to appraise themselves using the self-appraisal tool, or for those who wanted to see best practices (Annual Report, , p. 7). The focus of our empirical analysis is based on the CCGG s third strategy: Engaging directly with company chairs, CEOs and corporate secretaries to promote good corporate governance policies and practices. This strategy included letter writing campaigns and direct engagements with firms chairs and independent board members. While letter writing is pursued by investor groups in other settings, it is rare that a collective action organization of investors engages directly with firms via meetings with their board members. This engagement policy was initially limited to behind-the-scenes discussion and did not include a commitment to coordinate and initiate public proxy fights. The CCGG signalled a first step toward using this approach in 2005 when it began a campaign to improve shareholder democracy. In 2003, in all Canadian firms (except one) directors were elected by a plurality of votes cast. In uncontested elections, directors could be elected with a single vote. There were no votes for individual directors, and no required publication of the vote totals. As the CCGG stated in its letters to issuers and in subsequent publicly posted board policies, the existing system does not permit shareholders to vote against an underperforming director and allows an entrenched board to continue to be in charge of the company, even if they are opposed by a majority of the owners of the company. The only option is to undertake a costly and confrontational proxy fight. 5 Instead, the CCGG suggested that to ensure effective shareholder democracy in the election of directors, the CCGG recommends that the board of every public company adopt a majority voting bylaw or board policy. It further specified this should include provisions for individual elections, disclosure of votes, use a majority standard 4 The Dey Report, commissioned by the TSX, recommended reforms to board structure, to separate the role of CEO and Chair, to ensure that independent directors sat on the audit and compensation committees, and to reform board processes to include director and board evaluations. 5 CCGG Majority Voting Policy, March

11 whereby board members that failed by this standard submitted their resignation which would be accepted by the board absent extraordinary circumstances. With these changes, investors could focus their discontent on specific directors, thereby raising reputational and real costs for directors and firms. The CCGG s second campaign focused on firm s compensation practices, with concerns about structure, disclosure, and accountability. The CCGG provided financial resources that supported the collection of data on current practices and began a dialogue with firms. In 2007 the CCGG issued its principles of executive compensation and provided user-friendly examples through best practices in compensation disclosure. As part of these efforts, the CCGG identified gaps in the mandated disclosures, in choices around uncapped pension payments, the absence of clawback provisions, and the fact that few firms used meaningful performance peers when awarding performance-based executive compensation. 6 A related, but distinct campaign became clear in April 2009 when the CCGG publicly advocated for firms to introduce an advisory say on pay vote, stating that it would give shareholders an opportunity to express directly to the board their satisfaction with the prior years compensation plan and actual awards. 7 Similar to its majority voting policy, after consulting with firms and legal experts, the CCGG issued a model form on Say on Pay for boards to consider introducing at their annual meetings in September Data on CCGG Engagements and Firms-Level Governance Changes 2.1. The CCGG s Private Engagements With Firms Private communications are the CCGG s preferred form of engaging firms. These private communications not only helped shape the CCGG s views, they resulted in some firms being directly exposed to pressure from the CCGG to make certain governance changes. To examine whether there is a differential impact on the firms engaged by the CGGG, we exploit data based on private communications between the CCGG and these firms, generously provided to us by the CCGG. 6 An issue perhaps notable for its absence was an initiative on anti-takeover devices or staggered boards. This reflects the fact that in Canada a number of features limit the impact of anti-takeover efforts. First, investors can call an Extraordinary General Meeting and replace directors, thus effectively eliminating staggered boards. Though they are rarely used, EGMs do make investors potentially more powerful. Second, in Canada investors can appeal to the Securities regulator to cease trade poison pills and this makes it difficult for directors to just say no to potential takeover offers. 7 CCGG Shareholder Engagement and Say on Pay Policy. 10

12 The private data consists of file records from the time period, including engagement letters with firms, ex post reports of those engagements, and assorted other information. With this data, we assembled a comprehensive record of the CCGG s engagements with firms, the specific items raised in the discussions, the identities of the meeting attendees (the specific CCGG members and the engaged firm s board members), the timing of such communications, and private information on CCGG members ownership stakes in the firms. For the majority voting campaign, we first identify the CCGG s efforts to convey its views to firms through letters and meetings. The 2005/2006 campaign had two related stages. It began in the fall of 2005 when the CCGG contacted the boards of the five major Canadian banks (RBC, TD Bank, Scotiabank, CIBC and BMO) and two major insurers (Manulife and Sun Life). All seven firms committed to adopt majority voting by the spring of 2006 and it was widely acknowledged that the CCGG was the driving force behind the adoptions. The CCGG then sent letters to 81 additional firms in which they identified majority voting as an issue of concern. In many cases, the letter stated the combined stakes of CCGG members in the firm and noted that the seven major financial institutions had agreed to adopt majority voting. Finally, the letter provided examples of ways of improvement. The engaged firms were sent a follow up letter, sometimes accompanied by a phone call, or other communication. The CCGG also published its policy guidelines on Majority Voting in August 2006 to provide firms with additional guidance on best practices. For the CCGG s say on pay initiative and its efforts to encourage changes in compensation policies, we focus on a second round of CCGG activity that started in 2008 and involved more intensive engagements between the CCGG and firms boards. It stated, the CCGG believes that institutional shareholders should have regular, constructive engagement with the boards and board compensation committees of public companies to explain their perspectives on governance, compensation and disclosure practices, and to provide detailed comment on the company s practices to the board. CCGG expects boards will welcome this direct constructive interaction with large shareholders, which will 11

13 normally be held without management or advisors, and that it will lead to a better alignment of the interests of shareholders with the interest of the board and management. 8 Over a period of four years from 2008 to 2011, the CCGG conducted 104 engagements with 65 different firms. Table 2 provides summary statistics on these engagements. The CCGG usually met with at least two independent directors (94% of meetings), the independent chair of the board or the lead independent director (79% of meetings), or the chair of the compensation committee and other independent directors. In 75% of the engagements, the meeting took place without any members of management, legal counsel, or investor relations staff. At most engagement meetings, one or two directors or senior executives from the CCGG s member institutions were present and accompanied by a senior CCGG staff member. In 64% of the engagements, at least one of the CCGG members was from a pension plan. In advance of most meetings, CCGG staff compiled a briefing report on the company to prepare for the meeting. More importantly for our purposes, after each meeting, CCGG staff prepared a meeting report that summarized the issues discussed and who was present. The meeting report was sent back to the independent directors of the engaged firm who checked it for accuracy before it was posted on the members only version of the web site. We read each report and identified the issues that were raised. We then identified four issues that were commonly raised in the engagements: say on pay, capped pensions, clawbacks, and performance peer groups. Table 2 provides summary statistics on the number of engagements by year as well as the issues raised in the engagements Data on Adoptions and Other Variables To identify whether firms adopted the governance practices that the CCGG raised in the engagements, we read through firms proxy circulars which were circulated in advance of annual shareholder meetings and are available on SEDAR. 9 In particular, we identified whether the proxy circulars disclosed that the firm had an internal policy requiring the resignation of a director receiving a majority of withheld votes during director elections (majority voting); had a policy of holding shareholder 8 CCGG Shareholder Engagement and Say on Pay Policy. 9 The System for Electronic Document Analysis and Retrieval (SEDAR) is a document system operated by the Canadian Securities Administrators which contains filings made by Canadian public companies. It is similar to EDGAR, operated by the Securities and Exchange Commission in the US. 12

14 advisory votes (say on pay); had a policy to allow for it recoup performance-based executive compensation in the event of a restatement of performance metrics (clawback); imposed a limit on the pension or retirement benefits that could be paid out as a part of executive compensation packages (capped pensions); evaluated performance relative to a specified group of peer firms when determining performance-based compensation (performance peer groups). We use several additional variables in our analysis. Table A.I provides complete definitions. Ownership by CCGG members was provided to us by the CCGG. Each year we compute the log of one plus the aggregate CCGG members dollar stakes ( Log(1+CCGG $ Ownership) ) and aggregate percentage stakes ( CCGG % Ownership) in each firm in the S&P/TSX Index between 2004 and When these data are not available, we use data from FactSet and Capital IQ, which are likely less precise because of limits to ownership disclosure. Ownership held by institutional investors that are not members of the CCGG is also from FactSet and Capital IQ ( Institutional % Ownership ex CCGG ). To identify corporations controlled by insiders we use the CCGG s definition. We set a dummy variable equal to one if a firm has multiple voting shares or if a shareholder (or affiliated shareholders) control 30% or more of a firm s votes ( Controlled Corporation ). To capture a firm s predisposition towards good governance, we use governance scores, reported by Canada s national business paper, The Globe and Mail. The governance practices of firms included in the S&P/TSX Composite index are rated based on four distinct categories (board composition, shareholding and compensation, shareholder rights, and disclosure). By construction, the scores also capture some of the governance differences between controlled corporations and widely-held firms. Because the criteria for these scores change over time, we set a dummy equal to one for firms ranked in the top third in a given year ( Good Governance ). Firm size ( Log(Assets) ) and the financing deficit ( Financing Deficit ) are from Worldscope. We follow Frank and Goyal (2003) and construct the financing deficit as the sum of dividends, investment and change in net-working capital less cash flow, deflated by total assets. We use data from Datastream to measure stock returns ( 1-year Stock Return ) and compute a proxy for liquidity ( Turnover ). Information on shareholder proposals is collected from SHARE.ca, the website of the Shareholder Association for Research and Education. We focus on shareholder proposals specifically related to the governance practices that the CCGG discussed with firms in their engagements ( Majority Voting 13

15 Proposal, Say on Pay Proposal, Compensation Policy Proposal ). We use a variety of sources to identify cross-listed firms and create a dummy that equals one for firms cross-listed on a US stock exchange in a given year ( Cross-Listed Firm ). For media exposure we use the log of one plus the number of media citations ( Log(1+Firm Media Cites) ) found in The Globe and Mail and The Financial Post during a given year. Finally, we use data on directors from the Clarkson Centre for Business Ethics and Board Effectiveness to construct a measure of board interlocks. We set a dummy variable equal to one for firms with a director who was also on the board of a firm that was engaged by the CCGG in the prior year ( Interlock ). 3. Empirical Results In this section, we present the results of tests that examine the CCGG s influence on the adoption of governance changes, including majority voting, say on pay, and compensation policies. For each governance policy, we first examine the determinants of CCGG engagements, i.e., which firms did the CCGG choose to engage? We estimate logit regressions, where the dependent variable equals one if a firm was engaged on a given policy by the CCGG in a given year and zero otherwise. The majority voting engagements and explanatory variables are from The logit regressions for say on pay and compensation policies use data on engagements and explanatory variables from 2008 to To test for an impact of CCGG engagements on firms adoptions of governance policies, we estimate logit regressions where the dependent variable is a dummy that equals one if a firm adopted a given governance policy within the subsequent two years ( Adoption ) and zero otherwise. The key explanatory variable of interest is the engagement dummy variable that equals one if the CCGG previously engaged the firm. The other explanatory variables are also lagged in the adoption logit regressions. The majority voting logit regressions include industry fixed effects based on one digit SIC codes and we report robust standard errors. Summary statistics for the variables used in the majority voting engagement and adoption logit regressions are in Table 3. The logit regressions for say on pay and compensation policies include industry and year fixed effects and the standard errors are clustered at the firm-level. Summary statistics for the variables used in the say on pay and compensation policy engagement and adoption logit regressions are in Table 5. 14

16 3.1. Majority Voting: Engagements and Adoptions The sample we use for majority voting engagements includes 212 firms in the S&P/TSX index with complete data for Of these, 88 firms were sent letters by the CCGG requesting that the firm adopt majority voting. The dependent variable is set to one for these firms ( Engaged on Majority Voting ). Table 4A presents the results of logit regressions that examine the determinants of the majority voting engagements. In this table we want to understand better the importance of the economic incentives of the CCGG to engage firms, captured by either CCGG members dollar or percentage stakes in a firm. The choice of firms to engage could also be influenced by other factors that could affect the success of the engagement. For example, ownership by other institutions could provide the CCGG with additional influence, firms with a reputation for good governance might be more receptive to the CCGG s engagements as might cross-listed firms, whereas controlled corporations could be more difficult to influence. We also use firm size as a catch-all measure for openness to external influences. Finally, the CCGG might also engage with firms that are underperforming. The coefficient on Log(1+CCGG $ Ownership) in model (1) is significant at the 1% level and the Pseudo R 2 is In model (2) CCGG % Ownership is statistically significant as is Institutional % Ownership ex CCGG, albeit at the 10% level. However, the Pseudo R 2 is less than half that of Log(1+CCGG $ Ownership) in model (1). The coefficients on the Good Governance and Cross-Listed Firm dummies and Log(Assets) are positive and significant in models (4), (6), and (7) but their explanatory power is also low compared to that of Log(1+CCGG $ Ownership). The coefficients on the Controlled Corporation dummy and 1-year Stock Return are not significant in models (3) and (5). In models (8) to (11), we include either Log(1+CCGG $ Ownership) or CCGG % Ownership and Institutional % Ownership ex CCGG plus either the Controlled Corporation dummy or the Good Governance dummy, along with the other three variables. In models (8) and (9), the only coefficient that is significant is that of Log(1+CCGG $ Ownership). In (10) and (11), the coefficients on CCGG % Ownership and Institutional % Ownership ex CCGG are both significant at the 5% level or better, as is Log(Assets). However, the Pseudo R 2 in models (10) and (11) is lower than that of model (1) which includes Log(1+CCGG $ Ownership) alone. Overall, the most important determinant of majority voting 15

17 engagements is Log(1+CCGG $ Ownership), as shown in model (1). This is consistent with statements made by the CCGG in its annual reports. The majority voting adoption results are in Table 4B. We focus on adoptions in the two years (2006 and 2007) subsequent to the engagements. The sample includes the same set of firms as in Table 4B but is smaller because 16 firms die in 2006 and These firms could not potentially adopt majority voting and are dropped. The coefficient on the Engaged on Majority Voting dummy is positive and significant in each model. In model (1), which includes the Engaged on Majority Voting dummy without any controls other than industry fixed effects, the marginal effect is 0.303, e.g., CCGG engagement increases the probability of adopting majority voting by 30%. We next move to specifications that include controls for other internal factors that could influence adoptions that are unrelated to CCGG engagements as well as other external influences. To capture two key internal factors, models (2) and (3) add the Controlled Corporation and Good Governance dummies, respectively. Controlled corporations are significantly less likely to adopt majority voting while firms with better governance are more likely to adopt majority voting. Models (4) and (5) are similar but also control for firm size as a catch-all measure for internal factors and openness to external influences. The coefficient on Log(Assets) is positive and significant, but the results are otherwise similar to those in models (2) and (3). In models (5) and (6) we consider specific factors that could influence adoption decisions. Higher percentage ownership by CCGG members and by other institutional investors could influence firms to adopt majority voting. With higher turnover, institutional investors can more easily express their displeasure by exiting, which could put pressure on firms to consider the CCGG s request. Firms with poor prior performance may be more readily influenced by investor pressures as might firms with a greater need to raise external capital. However, with the exception of Financing Deficit, which is significant at the 10% level in model (6) (but has a negative coefficient), none of these variables are significant. Other external factors that could lead to adoptions are media exposure, pressure from shareholder proposals, or pressure from the US markets. The coefficients on Log(1+Firm Media Cites), Majority Voting Proposal, and Cross-Listed Firm are all positive, but only that on Log(1+Firm Media Cites) is significant, and only at the 10% level. The Engaged on Majority Voting dummy is positive and 16

18 significant in both models (6) and (7). In these models that include a full set of controls, the marginal effects are 0.22 and 0.16, respectively. Overall, the variables that reliably explain the decision to adopt majority voting are whether or not the firm was engaged by the CCGG, whether or not it is controlled by insiders, and whether or not it already has good governance. Our catch-all variable, firm size, is statistically significant, but the specific proxies we use for internal factors and external pressures are not. Because we use prior engagements to explain subsequent adoptions, reverse causality cannot explain the results. However, we cannot rule out that there are unobservable differences between firms engaged by the CCGG and those not engaged and these differences cause firms to adopt majority voting rather than the CCGG engagements. Unfortunately, all majority voting engagements occurred in 2005 and there is little we can do to formally address this issue. We do note however, that as shown in Table 4A, the most important determinant of majority voting engagement is the aggregate dollar stake of CCGG members (approximately 40 institutions that controlled about $800 billion in assets in 2005) and that these members acquired their holdings independently from each other. Thus concerns about causality must stem from unobserved differences between firms in which CCGG members held larger dollar stakes and those in which CCGG members held smaller stakes. In the next sub-section that examines say on pay and compensation policies, we can more formally examine the impact of unobserved differences between engaged and non-engaged firms Say on Pay and Compensation Policies: Engagements and Adoptions In its second campaign, the CCGG focused on engagement meetings with firms boards to discuss governance issues. To facilitate the engagements, CCGG staff contacted the board and requested a meeting. Over the period from 2008 to 2011, it conducted 104 engagement meetings with 65 different firms. In approximately half of these engagements the CCGG informed firms of its preference for the firm to introduce say on pay. It also raised concerns about the firms policies related to compensation structure and disclosure (including capping pension payouts, introducing clawback provisions, and using performance peer groups) in 48 of these engagements. The sample includes firms in the S&P/TSX index during 2008 to 2011 with complete data. In the engagement logit regressions there are there are 782 firmyear observations. For say on pay (Table 6A) the dependent variable equals one for 48 observations ( Engaged on Say on Pay ) and for compensation policies (Table 7A), the dependent variable equals one 17

19 for 48 observations ( Engaged on Compensation Policies ). We focus on the same explanatory variables used for majority voting engagements in Table 4A, though we now have a four year panel. Inferences from models (1) to (6) are the same for Tables 6A and 7A. As with majority voting, in model (1), the coefficient on Log(1+CCGG $ Ownership) is statistically significant, though its explanatory power is lower. Surprisingly, the CCGG % Ownership in model (2) is not significant, although Institutional % Ownership ex CCGG is still significant at the 10% level. Models (3) and (4) show that controlled corporations are less likely to be engaged and firms with better governance are more likely to be engaged. In the majority voting campaign, the CCGG did not distinguish between controlled and widely-held corporations but in the campaign involving direct meetings, it did. Models (5), (6), and (7) show that the CCGG was more likely to engage firms with worse prior performance, cross-listed firms, and larger firms. Models (8) to (11) include either Log(1+CCGG $ Ownership) or CCGG % Ownership and Institutional % Ownership ex CCGG plus either the Controlled Corporation dummy or the Good Governance dummy, along with the other three variables. Inferences are similar to those in models (1) to (6) except that the coefficients on Institutional % Ownership ex CCGG and the Cross-Listed Firm dummy are less reliably significant in both tables. The Good Governance dummy is not significant in models (9) and (11) in Table 6A but retains its significance in Table 7A. Thus, similar to the majority voting engagements, the aggregate dollar stake of CCGG members is an important determinant of which firms are engaged by the CCGG. The main difference is that the CCGG was now also less likely to engage controlled corporations and more likely to engage firms with lower stock returns over the prior year. The logit regressions that examine the adoption of say on pay are in Table 6B. The sample construction is the same as that in Table 6A, but drops firms that die in the two years following a potential engagement. We focus on adoptions from 2009 to 2012 so that engagements and explanatory variables are lagged. We also drop firms in the years after they adopt say on pay as they no longer face a choice to adopt or not. This leaves a sample 632 firm-year observations. The coefficient on the Engaged on Say on Pay dummy is positive and significant in each model. The marginal effect in model (1) is 0.25, e.g., CCGG engagement increases the probability of adopting say on pay by 25%. Controlled corporations are less likely to adopt say on pay and firms with good governance 18

20 are more likely to adopt say on pay and these inferences hold no matter whether we control for internal factors and external pressures with firm size in models (4) and (5) or whether we use more explicit proxies in models (6) and (7). In these models, CCGG members percentage stakes and ownership by other institutions does not influence adoption. Nor does turnover, prior stock returns, or need to raise external finance. However, media attention, whether or not a firm received a say on pay proposal, and cross-listing status, are all positively related to adoptions. In these models, the marginal effects for the Engaged on Say on Pay dummy are 0.10 and 0.09, respectively. As with majority voting, the specific timing of the CCGG s engagements and firms subsequent adoptions allows us to rule out reverse causality as an explanation for the results. However, the potential concern that adoption decisions could be driven by unobservable differences between engaged and nonengaged firms remains. In its second campaign, the CCGG engaged firms in different years and on different issues. We exploit these differences to address endogeneity concerns and set a dummy variable that equals one for firms that were engaged by the CCGG on any issue in any year between 2008 and 2011 ( Engaged Firm ). For example, if a firm was engaged in 2010 on any issue, we set the dummy to one each year from 2009 to When we include the Engaged Firm dummy in the say on pay adoption regressions, it controls for unobservable differences between engaged and non-engaged firms and allows the Engaged on Say on Pay dummy to identify the specific impact of the say on pay engagements. The results are shown in models (8) and (9). Although the Engaged Firm dummy is positive and significant, the Engaged on Say on Pay dummy remains significant at the 1% level and our earlier inferences are unchanged. The results for adoption of compensation policies are in Table 7B. The sample construction is similar to that in Table 6B. The dependent variable is set to one if a firm adopted one or more of capped pensions, clawbacks, or performance peer groups within two years of the engagement and if the firm was engaged on that particular issue. A firm is dropped from the sample once it adopts all three of these policies. The final sample includes 732 firm-year observations. Inferences for models (1) to (4) are similar to those in Table 6B. The marginal effect on the Engaged on Compensation Policies dummy in model (1) is In models (5) and (6), the coefficient on Log(1+Firm Media Cites) is positive, but less reliably than in Table 6B. Compared to Table 6B, the coefficients on the Compensation Policy Proposal 19