x

A PHP Error was encountered

Severity: Notice

Message: Undefined variable: content_category

Filename: user/transcript.php

Line Number: 106

A PHP Error was encountered

Severity: Warning

Message: Invalid argument supplied for foreach()

Filename: user/transcript.php

Line Number: 106

Total Views  :   1441
Total Likes  :  0
Total Shares  :  0
Total Comments :  0
Total Downloads :  0

Add Comments
Presentation Slides

1) Passive Investors, Not Passive Owners1 Ian R. Appel2, Todd A. Gormley3, and Donald B. Keim4 September 4, 2015 Abstract Passive institutional investors are an increasingly important component of U.S. stock ownership, and their influence on firm-level governance is widely debated. To examine whether and by which mechanisms passive investors influence firms’ governance structures, we use an instrumental variable estimation and exploit variation in ownership by passively managed mutual funds that results from stocks being assigned to either the Russell 1000 or 2000 index. Our findings suggest that passive investors play a key role in influencing firms’ governance choices; ownership by passively managed funds is associated with more independent directors, the removal of takeover defenses, and more equal voting rights. Passive investors appear to exert influence through their large voting blocs, and consistent with the observed differences in governance having a positive influence on firm value and reducing the need for activism by other investors, we document a decline in activism by non-passive investors and improvements in firms’ longer-term performance. (JEL D22, G23, G30, G34, G35) Keywords: governance, institutional ownership, passive funds                                                                                                                         1 For helpful comments, we thank Bernard Black, Alon Brav, Alan Crane, Sevinc Cukurova, David Denis, Ran Duchin, Alex Edmans, Vyacheslav Fos, Erik Gilje, Stuart Gillan, Vincent Glode, Robert Jackson, Wei Jiang, Charles Jones, OÄŸuzhan Karakas, Borja Larrain, Doron Levit, Craig Lewis, Inessa Liskovich, Michelle Lowry, Nadya Malenko, Pedro Matos, David Matsa, Sebastien Michenaud, David Musto, Christian Opp, Ed Rice, Michael Roberts, Dick Roll, Luke Taylor, Paul Tetlock, Jeremy Tobacman, James Weston, and seminar participants at Boston College (Carroll), Columbia Business School, Florida Atlantic University, FTSE World Investment Forum, University of Pennsylvania (Wharton), University of Washington (Foster), U.S. Securities and Exchange Commission, NBER Summer Institute for Corporate Finance, Western Finance Association Meetings, ISB Summer Research Conference, Drexel Conference on Corporate Governance, Ohio State University Corporate Finance Conference, Society of Financial Studies Cavalcade, 9th International Finance UC Conference, and the NYU/Penn Conference on Law and Finance. We also thank Alon Brav for sharing his data on hedge fund activism, Louis Yang for his research assistance, and the Rodney L. White Center for Financial Research for financial support. 2 Carroll School of Management, Boston College, 140 Commonwealth Avenue Chestnut Hill, MA, 02467. Phone: (617) 552-1459. Fax: (617) 552-0431. E-mail: ian.appel@bc.edu 3 The Wharton School, University of Pennsylvania, 3620 Locust Walk, Suite 2400, Philadelphia, PA, 19104. Phone: (215) 746-0496. Fax: (215) 898-6200. E-mail: tgormley@wharton.upenn.edu 4 The Wharton School, University of Pennsylvania, 3620 Locust Walk, Suite 2400, Philadelphia, PA, 19104. Phone: (215) 898-7685. Fax: (215) 898-6200. E-mail: keim@wharton.upenn.edu

2) “We’re going to hold your stock when you hit your quarterly earnings target. And we’ll hold it when you don’t. We’re going to hold your stock if we like you. And if we don’t. We’re going to hold your stock when everyone else is piling in. And when everyone else is running for the exits. That is precisely why we care so much about good governance.” — F. William McNabb III, Chairman and CEO of Vanguard I. Introduction While there is considerable evidence that institutional investors influence the governance and policies of firms (e.g., Aghion, Van Reenen and Zingales (2013); Brav et al. (2008)), this evidence primarily focuses on the role of activists that accumulate shares and make demands upon managers or active fund managers that exit positions when managers perform poorly. Yet, such active investors represent only a subset of institutions. Many institutions are instead passive in that they do not actively buy or sell shares to influence managerial decisions. The investment objective of such institutions is to deliver the returns of a market index (e.g., S&P 500) or investment style (e.g., large-cap value) with low turnover, diversified portfolios, and minimal expenses. As shown in Figure 1, passive investors have grown significantly in recent years; the share of equity mutual fund assets held in passively managed funds tripled over the 1998-2014 period to 33.5%, and the share of total U.S. market capitalization held by passively managed funds quadrupled to more than 8%. However, the growth of passive investors raises questions about how effectively managers are being monitored. Many worry that passive investors lack the motives and resources to monitor their large, diverse portfolios, and that the increasing market share of such “lazy investors” weakens firm-level governance and hurts performance.1 Others counter that passive investing does not equate with passive ownership.2 In this paper, we examine whether passive institutional investors influence firms’ governance structures, and ultimately, performance. There are reasons to suspect that the growth of passive investors weakens the governance and performance of firms. First, such institutional investors may lack an incentive to monitor managers. In particular, passive funds seek to deliver the performance of the benchmark, and unlike actively managed funds, they have little motive to improve an individual stock’s performance. Second, passive investors may be less able to exert influence over managers. By seeking to minimize deviations from the underlying index weights, passive institutions tend to                                                                                                                         1 An example of this viewpoint was expressed in The Economist on Feb. 7, 2015. As it stated, “A rising chunk of the stock market sits in the hands of lazy investors. Index funds and exchange-traded funds mimic the market’s movements, and typically take little interest in how firms are run; conventional mutual funds and pension funds that oversee diversified portfolios dislike becoming deeply involved in firms’ management.” 2 For example, the title of this paper, “Passive Investors, Not Passive Owners,” was the title for an article written by Glenn Booraem, controller of Vanguard, in April 2013 highlighting the care Vanguard takes when voting proxies. See https://personal.vanguard.com/us/insights/article/proxy-commentary-042013. Similar views regarding the distinction between being a passive investor, but active owner, were espoused by Rakhi Kumar, head of corporate governance at State Street Global Advisors in The Financial Times on April 6, 2014 in an article titled, “Passive investment, active ownership,” and by David Booth, chairman and co-founder of Dimensional Fund Advisors, in the New York Times on March 16, 2013 in an article titled, “Challenging Management (but Not the Market)”. Passive Investors, Not Passive Owners -- Page 1

3) lack a traditional lever used by non-passive investors to influence managers—the ability to accumulate or exit positions. Third, given their diversified holdings, passive investors may have insufficient resources to research and monitor the corporate policies of each individual firm in their portfolio. And yet, there are reasons why the growth of passive investors may affect firms’ governance choices and performance. First, passive institutions may be motivated to monitor managers and improve overall market performance because this increases the value of their assets under management (Del Guercio and Hawkins (1999)). Moreover, because passive institutions are less willing to divest their positions in poorly performing stocks, they may be more motivated than other institutions to be engaged owners (Romano (1993), Carleton, Nelson, and Weisbach (1998)). Second, institutions that manage passive funds can use their sizable ownership stakes and voting ability to wield influence. All institutional investors have a fiduciary duty to vote their proxies in the best interest of shareholders, and managers may be more inclined to consider the views of passive investors over more active investors, which tend to exhibit higher turnover rates (Del Guercio and Hawkins (1999)). Finally, while passive institutions may lack the resources necessary to monitor the detailed policy choices of every firm in their large, diversified portfolios, they may be effective at engaging in widespread, but low-cost, monitoring of firms’ compliance with what they consider to be best governance practices (e.g., Black (1992), Black (1998)). Identifying the impact of passive investors on firms’ corporate governance and performance can be challenging. Correlations between passive investors and governance choices might not reflect a causal relation since ownership by passive investors might be correlated with factors—such as firms’ investment opportunities or ownership by active investors—that directly affect managerial decisions. To overcome this challenge and to assess whether passive investors affect firms’ governance and performance, we exploit variation in ownership by passive investors that occurs around the cutoff point used to construct two widely-used market benchmarks, the Russell 1000 and Russell 2000 indexes. The Russell 1000 comprises the largest 1,000 U.S. stocks, in terms of market capitalization, and the Russell 2000 comprises the next largest 2,000 stocks. Because portfolio weights assigned to each stock within an index are value-weighted, a stock’s index assignment has a significant impact on the extent of ownership by passive funds. For example, the 750th through 1,000th largest stocks will be included in the Russell 1000 and be given small portfolio weights because they represent the smallest firms in their index, while the 1,001st through 1,250th largest stocks will be included in the Russell 2000 and be given weights that are an order of magnitude larger because they represent the largest firms in their index. Therefore, for each dollar invested in a passive fund using the Russell 1000 as a benchmark, very little of it will be invested in stocks at the bottom of that index; while for each dollar invested in a passive fund using the Russell 2000 as a benchmark, a large proportion of it will be invested in stocks at the top of the index. This benchmarking by passive funds leads to a sharp difference in ownership by passive institutional investors for stocks at the top of the Russell 2000 relative to stocks at the bottom of the Russell 1000 even though these stocks are otherwise similar in terms of their overall market capitalization. We find that the ownership of passively managed mutual funds is, on average, about 66% higher for stocks at the top of the Russell 2000 index relative to stocks at the bottom of the Russell 1000 index. The difference in passive ownership matches what one would predict based on the amount of money estimated to be passively tracking the two indexes and Passive Investors, Not Passive Owners -- Page 2

4) corresponds with a significant shift in firms’ ownership structure. On average, the ownership stakes of three of the biggest passive investors, Vanguard, State Street, and Barclays Bank (which owned iShares during our sample), are a third higher among firms at the top of the Russell 2000, and each of these three institutions’ likelihood of owning more than 5% of a firm’s shares increases by two thirds on average, while their likelihood of being a top 5 shareholder is higher, on average, by 15%. We find no corresponding difference in ownership of stocks around the cutoff among actively managed mutual funds. Exploiting this variation in ownership around the Russell 1000/2000 cutoff in an instrumental variable (IV) estimation, we assess the effect of passive funds on firms’ governance structures and performance. Specifically, we instrument for ownership by passive funds with an indicator for being assigned to the Russell 2000 in a given year. Our IV estimation relies on the assumption that, after conditioning on stocks’ market capitalization, inclusion in the Russell 2000 index does not directly affect our outcomes of interest except through its impact on passive ownership. This assumption seems reasonable in our setting in that it is unclear why index inclusion would be directly related to governance and corporate performance after restricting the sample to stocks near the Russell 1000/2000 cutoff and after controlling for the factor that determines index inclusion—stocks’ end-of-May market capitalization. The three broad governance outcomes we analyze reflect those highlighted in a recent speech by the CEO of Vanguard (see McNabb (2014)) and the historical proxy-voting policies of the largest passive institutional investors. While passive institutions do vary their voting strategy across firms on governance issues (e.g., see Davis and Kim (2007)), three common themes of their historical proxy voting policies were (1) to support greater board independence, (2) oppose antitakeover provisions, and (3) oppose unequal voting rights, as occurs when firms maintain a dual class share structure (e.g., see McNabb (2014) and the Appendix for more details on voting guidelines of three prominent passive institutional investors). We also analyze vote outcomes, such as the average support for management and governance-related shareholder proposals, which could be directly related to a potential mechanism by which passive investors may exert influence—their ability to exercise “voice”. Using our IV approach, we find that passive investors have a significant impact on each of the three aspects of governance. First, an increase in ownership by passive funds is associated with an increase in board independence. A one standard deviation increase in ownership by passive funds is associated with about a 0.7 standard deviation increase in the share of directors on a firm’s board that are independent. Second, passive ownership is associated with the removal of takeover defenses. A one standard deviation increase in ownership by passive funds is associated with 3.5 percentage point increase in the likelihood of removing a poison pill and a 2.5 percentage point increase in the likelihood of reducing restrictions on shareholders’ ability to call special meetings. These findings are economically large given that, on average, only 4% of firms remove a poison pill and 0.6% of firms eliminate restrictions on special meetings each year during our sample period. We find less evidence that passive ownership is associated with differences in other takeover defenses, including classified boards or supermajority voting requirements. Finally, an increase in passive ownership is associated with firms being less likely to have unequal voting rights, as captured by having a dual class share structure. A one standard deviation increase in ownership by passive funds is associated with about a one standard deviation decrease in likelihood of having a dual class share structure. Our evidence suggests that a key mechanism by which passive investors exert their influence is through the power of their large voting blocs (i.e., voice). Passive ownership is Passive Investors, Not Passive Owners -- Page 3

5) associated with a decline in the share of votes in support of management proposals, suggesting managers face a more contentious and attentive shareholder base, and an increase in support for governance-related shareholder proposals. A one standard deviation increase in ownership by passive funds is associated with about a 0.75 standard deviation decline in support for management proposals and about a 0.5 standard deviation increase in support for governance proposals. We find little evidence that these differences in support are driven by a change in the either the number or type of proposals brought to a vote. Because the size and concentration of passive investors’ ownership stakes may make it easier for activist investors to rally support for their demands (Brav et al. (2008), Bradley et al. (2010)), an alternative mechanism by which passive investors might influence governance outcomes is by facilitating the activist efforts of other investors. However, we find no evidence of a positive association between ownership by passive funds and the likelihood of a firm experiencing a hedge fund activism event or takeover event. Instead, we find evidence that a larger ownership stake by passive funds is associated with a decline in hedge fund activism, which is consistent with the engagement of passive investors reducing the need for activism by other investors. However, these findings do not exclude the possibility that passive investors’ ownership stakes increase the threat of activism by others, and that this perceived threat increases the power of passive investors’ voice. For example, companies may be responsive to the governance views of passive investors so as to lessen the likelihood that these investors later lend support to an activist campaign initiated by others. Consistent with the documented changes in governance having a positive influence on firm value we find that, on average, an increase in passive ownership is associated with an improvement in firms’ future performance. While we find no evidence of an association between passive ownership and measures of performance in our main IV specification, we find evidence that longer-term ownership by passive investors is associated with significant improvements in firms’ return on assets (ROA) and Tobin’s Q. On average, a one standard deviation increase in ownership by passive funds is associated with about a third of standard deviation increase in ROA. We find little evidence, however, that passive ownership is associated with differences in the level or composition of managerial pay (a fourth issue that is prominently discussed by passive investors during our sample period) or firms’ capital structure or investments. Overall, the findings are consistent with passive investors improving firm value by insisting on basic governance-related changes, as these changes appear to improve firm value but require a low level of costly monitoring, while avoiding more costly and firm-specific interventions related to managerial pay and firms’ investment or capital structure choices. Our findings are robust to various specification choices. For example, varying the number of stocks we investigate around the cutoff between the two indexes or varying the functional form we use to control for firms’ end-of-May market cap, which is the key factor determining stocks’ index assignment each year, does not affect our findings. The findings are also robust to adding various controls, including (1) firms’ float-adjusted market cap, which is a proprietary measure used by Russell to determine a stock’s ranking within indexes, (2) firms’ industry, (3) firms’ past stock returns, and (4) whether firms recently switched indexes. Moreover, the findings are robust to using alternative definitions of passive ownership as the key explanatory variable, including the institution-level (13F) ownership stake of the three largest passive institutions or institution-level measure of “quasi-index” ownership, as defined by Bushee (2001). Finally, we find no effect of passive ownership in placebo tests that assume differences in passive ownership at alternative market cap thresholds (i.e., instead of the Russell Passive Investors, Not Passive Owners -- Page 4

6) 1000/2000 cutoff). Overall, our findings contribute to the broad literature that studies the effects of institutional ownership of common stock. One strand of this literature analyzes institutional investors’ impact on various aspects of corporate governance, including governance indices (Aggarwal et al. (2011), Chung and Zhang (2011)), CEO pay sensitivity (Hartzell and Starks (2003)), and shareholder proposals (Gillan and Starks (2000)), while another strand studies the effects of institutional investors on corporate policies, including leverage (Michaely, Popadak, and Vincent (2014)), dividends (Grinstein and Michaely (2005)) and R&D (Bushee (1998), Aghion, Van Reenen, and Zingales (2013)). A number of recent papers also highlight the role of specific types of institutional investors, such as activist hedge funds (Brav et al. (2008); Klein and Zur (2009)). We contribute to this literature by focusing on passive institutions—a less studied, but increasingly important set of institutional investors. In this regard, our paper is related to studies of several pension funds that follow passive investment strategies but successfully engage in activism (e.g., Carleton, Nelson, and Weisbach (1998), Del Guercio and Hawkins (1999)). Some argue that activism by pension fund managers is at least partially motivated by politics rather than wealth maximization (Romano (1993)). Given that such pressures are largely confined to public pension funds, it is not clear that this success extends to passive investors more generally. However, consistent with these case studies and contrary to survey evidence that passive investors might lack the willingness and ability to monitor managers (Useem et al. (1993)), our evidence suggests that passive investors are not passive owners. In particular, we find evidence that passive investors successfully influence firms’ governance choices and improve long-term, firm-level performance. The results of this paper also provide new insights into the determinants of firms’ governance structures and how large shareholders influence managerial decisions. Typically, institutional investors, such as blockholders, are thought to influence governance through a combination of “voice” and “exit” (e.g., Edmans (2014) and Levit (2013)). Voice refers to direct intervention by shareholders through either formal (e.g., proxy voting) or informal (e.g., letters to the board) channels (Harris and Raviv (2010); Levit and Malenko (2011); Maug (1998); Shleifer and Vishny (1986)), while exit refers to the threat or actual selling of shares (Admati and Pfleiderer (2009); Edmans (2009); Edmans and Manso (2011)). However, because passive funds maintain portfolio weights that are often closely aligned with the weights in their chosen benchmark, their ability to influence managers is primarily limited to voice, which is thought to constrain their ability to influence corporate outcomes. Our paper finds otherwise.3 Finally, our work is related to recent papers that use the Russell 1000/2000 cutoff to analyze the price effects of additions and deletions from a market index (Chang, Hong, and Liskovich (2014)), the importance of institutional investors’ portfolio weights for monitoring incentives (Fich, Harford, and Tran (2015)), and the association between institution-level (13F) ownership and payouts, investment, CEO pay, management disclosure, acquisitions, and CEO                                                                                                                         3 In this regard, our findings complement those of Iliev and Lowry (2015), who analyze the determinants of mutual funds’ reliance on proxy advisory service companies like Institutional Shareholder Services (ISS). While not the focus of the paper, Section 4.3 of Iliev and Lowry presents evidence that index funds are more likely to “actively vote” their shares (as measured by being less likely to follow ISS vote recommendations on non-binding shareholder proposals). Choi, Fisch, and Kahan (2013) find similar evidence that the voting decisions of Vanguard, and other large fund families, vary substantially from ISS vote recommendations. Our findings demonstrate that the active monitoring of passive investors results in actual differences in firms’ governance structures and performance. Passive Investors, Not Passive Owners -- Page 5

7) power (Boone and White (2014); Crane, Michenaud, and Weston (2014); Lu (2013); Mullins (2014); Schmidt (2012)). In contrast to these papers, we use the Russell 1000/2000 cutoff and fund-level data to isolate variation in ownership by passively managed mutual funds, and we analyze the impact of such investors on governance outcomes they explicitly mention as being important (e.g., independent directors, fewer takeover defenses, and equal voting rights), and the mechanisms by which passive investors might influence such governance outcomes (e.g., proxy voting, shareholder proposals, and facilitating activism by others).4 II. Sample, data sources, and descriptive statistics In this paper, we merge stock-level data on mutual fund ownership and Russell equity index membership with firm-level governance, proxy voting, accounting, and executive compensation data. We now briefly describe each data source and our sample. A. Mutual fund holdings and Russell 1000/2000 index membership We use the S12 mutual fund holdings data compiled by Thomson Reuters and available from Wharton Research Data Services (WRDS) to compute mutual fund holdings in a stock as a percent of its market capitalization. Since May 2004, all mutual funds holding stocks traded on U.S. exchanges are required to report those holdings every quarter to the SEC using Forms NCSR and N-Q. Reported securities include all NYSE/AMEX/NASDAQ, Toronto and Montreal common stocks. Before May 2004, funds were required to report holdings only twice a year using Form N-30D, but many voluntarily reported holdings in the other two quarters. To adjust for any missing/unreported holdings between report dates prior to May 2004, we populate missing holdings by assuming that the holdings from the earlier date stay constant and use monthly data from CRSP (prices, adjustment factors) to compute imputed dollar values of these holdings.5 We exclude observations where the total mutual fund holdings exceed a firm’s market capitalization. We calculate the total market cap of each stock using the CRSP monthly file as the sum of shares outstanding multiplied by price for each class of common stock associated with a firm (i.e., we sum across all PERMNOs associated with each PERMCO). To classify a mutual fund as either passively managed or actively managed, we use a method similar to that of Busse and Tong (2012) and Iliev and Lowry (2015). Specifically, we obtain fund names by merging the Thomson Reuters data with the CRSP Mutual Fund data using the MFLINKS table available on WRDS. We then flag a fund as passively managed if its fund name includes a string that identifies it as an index fund or if the CRSP Mutual Fund Database classifies the fund as an index fund.6 We classify all other mutual funds that can be matched to the CRSP mutual fund data as actively managed, and funds that cannot be matched are left                                                                                                                         4 Beyond our focus on passively managed mutual funds and their impact on governance outcomes and corporate performance, our empirical methodology also differs from previous and contemporaneous papers that use the Russell cutoff as a source of identification. Using our IV methodology, we do not find differences in ownership by active institutions around the Russell 1000/2000 threshold, nor do we find that index assignment affects other corporate outcomes, including capital structure, investments, the composition of managerial pay, CEO turnover, and acquisitions, as found in some of these previous studies. We discuss the tradeoffs of the different methodologies used in this identification setting on our website, which can be found at http://ssrn.com/abstract=2641548. 5 WRDS estimates that approximately 60% of funds additionally report holdings every quarter before 2004. Thanks to Denys Glushkov at WRDS for assistance with S12 holdings. 6 The strings we use to identify index funds are: Index, Idx, Indx, Ind_ (where _ indicates a space), Russell, S & P, S and P, S&P, SandP, SP, DOW, Dow, DJ, MSCI, Bloomberg, KBW, NASDAQ, NYSE, STOXX, FTSE, Wilshire, Morningstar, 100, 400, 500, 600, 900, 1000, 1500, 2000, and 5000. Passive Investors, Not Passive Owners -- Page 6

8) unclassified. To generate variables for mutual fund ownership disaggregated into these three categories, we compute the percentage of each stock’s market capitalization that is owned by passive, active, and unclassified mutual funds at the end of each quarter.7 Our subsequent analysis is restricted to the sample of stocks found in the Russell 1000 and 2000 indexes between 1998 and 2006. We obtain data for the Russell 1000 and 2000 indexes from Russell, and we start the sample at 1998 because this is the first year Russell provides us with its proprietary, float-adjusted market capitalization, which is used to determine the rank (i.e., portfolio weight) of each security within an index. We end the sample prior to 2007, which is when Russell implemented a new methodology to construct the two indexes such that they no longer necessarily reflect the 1,000 and next 2,000 largest stocks by market capitalization. Russell also provided us with their proprietary end-of-May total market capitalization values for each year from 2002 to 2006. The importance of the end-of-May market capitalizations and of ending the sample prior to 2007 is described in Section III. B. Governance, voting, accounting, and compensation data Governance and voting data are largely obtained from Riskmetrics (ISS), which provides information on several aspects of corporate governance for firms in the S&P 1500. Following Riskmetrics’ classification of a director’s independence, which excludes linked directors (e.g., those with business ties to the firm), we calculate the percentage of independent directors on the boards of each firm for each year in the sample from the director dataset. The governance dataset from Riskmetrics is used to create indicator variables for whether a firm removes a takeover defense or has a dual class share structure in a given year. The governance database is available for alternating years in the sample, except for 1998 when there is a three-year lag. We also construct several variables related to shareholder proposals and voting. We use the voting results database from Riskmetrics to calculate the average percentage of shares that vote in support of management proposals at annual meetings and in support of shareholder-initiated governance proposals for each firm between reconstitutions of the Russell indexes (i.e., between July of year t and June of year t+1). Our data on poison pills are obtained from Shark Repellent (FactSet). Shark Repellent provides historical information on firms’ most recent poison pill, such as when the poison pill was renewed, withdrawn, or allowed to expire. The database covers all firms in the Russell 3000 beginning in 2001. We define our variable for poison pill removal as an indicator equal to 1 if a firm’s poison pill is either withdrawn or allowed to expire at time t, and zero otherwise. Because Shark Repellent only reports information on a firm’s most recent poison pill, our indicator only flags firms that removed a poison pill during our sample period and did not reinstate a poison pill subsequently. Annual accounting data are from Compustat, and we use executive compensation data from Execucomp. Accounting variables are winsorized at the 1% and 99% levels. Definitions for all our key variables are provided in Appendix Table 1. C. Sample and descriptive statistics                                                                                                                         7 In a previous version of this paper, we instead used total institutional ownership, as reported in the Thomson Reuters Institutional Holdings (13F) Database, and classified institutions as either passive or active using the quasi-index, transient, and dedicated categories of Bushee (2001). A discussion of these findings and the various tradeoffs inherent in using either the S12 or 13F holdings data can be found in Section VII.B of the paper. Passive Investors, Not Passive Owners -- Page 7

9) For our main analysis, we restrict our sample to stocks in the 250 bandwidth around the cutoff, as determined using the end-of-June Russell-assigned portfolio weights for stocks within each index. This sample spans an economically important set of midcap and small cap stocks, and as discussed in Section VII.A, our subsequent findings are robust to using both wider and narrower bandwidths. Table 1 reports summary statistics for our main sample. The average level of mutual fund ownership (as a percentage of shares outstanding) is around 25%. Actively managed funds are the largest category (approximately 19% of shares outstanding), with passive and unclassified funds each accounting for about 3% of shares outstanding. Support for management proposals is high (85%), consistent with the notion that many of the issues addressed by these proposals are routine in nature, while support for shareholder-initiated governance proposals is considerably lower (36%). Independent directors make up 65% of the total number of directors for firms in the sample. The table also shows that poison pill removals and the lessening of restrictions on shareholders’ ability to call a special meeting are relatively uncommon events in our sample, occurring in just 4% and 0.6% of firm-year observations, respectively. About 13% of firms have a dual class share structure. Finally, firms’ ROA averages about 0.03. III. Empirical framework Identifying the impact of passive investors on corporate governance and performance poses an empirical challenge. Cross-sectional correlations between passive ownership, governance, and performance might not reflect a causal relation since ownership by passive investors might be correlated with factors—such as firms’ access to capital, investment opportunities, or ownership by active investors—that directly affect corporate outcomes. Failure to control for such factors could introduce an omitted variable bias that confounds inferences. To overcome this challenge and to determine the importance of passive investors, we use stocks’ assignment to the top of the Russell 2000 index as an exogenous shock to ownership by passive investors. We now describe our identification strategy. A. Russell index construction and passive institutional investors Passive funds attempt to match the performance of a market index by holding a basket of representative securities in the particular market index being tracked in proportion to their weights in the index. The most visible types of passive funds are index funds, which hold nearly all stocks in the market index rather than a representative sample. Two market indexes widely used as benchmarks are the Russell 1000 and Russell 2000. The Russell 1000 comprises the largest 1000 U.S. stocks in terms of market capitalization, while the Russell 2000 comprises the next largest 2000 stocks. An example index fund that uses the Russell 1000 as a benchmark is the Vanguard Russell 1000 Index Fund (VRNIX), while the Vanguard Russell 2000 Index Fund (VRTIX) uses the Russell 2000 as a benchmark. To account for changes in stocks’ ranking by market cap, the Russell indexes are reconstituted each year at the end of June. On the last Friday of June, Russell Investments determines which stocks will be included in the two indexes for the following twelve months using market capitalization as of the last trading day in May of that year.8 In other words, the                                                                                                                         8 However, when the last Friday of June falls on the 29th or 30th, the two indexes are reconstituted on the preceding Friday. During the following twelve months, stocks are only deleted from the indexes due to Chapter 7 bankruptcy filings, delistings, and corporate actions (takeovers), while IPOs are added quarterly to the indexes on the basis of the market capitalization breaks established during the most recent Passive Investors, Not Passive Owners -- Page 8

10) 1000 largest stocks at the end of the last trading day in May will be included in the Russell 1000, while the next 2000 largest stocks will be included in the Russell 2000.9 Each stock’s weight in the index is then determined using its end-of-June float-adjusted market cap. The float-adjusted market capitalization is different than the market capitalization used to determine index membership in that it only includes the value of shares that are available to the public. For example, shares held by another company or individual that exceed 10% of shares outstanding, by another member of a Russell index, by an employee stock ownership plan (ESOP), or by a government will be removed when calculating a firm’s float-adjusted market capitalization, as will unlisted share classes. Therefore, a stock that was the 1,000th largest stock in total market capitalization need not be the stock with the smallest portfolio weight in the Russell 1000 index. Because the Russell indexes are value-weighted, index assignment has a significant effect on portfolio weights; the 1000th largest stock at the end of May will be included in the Russell 1000 and be given a very small portfolio weight within its index, while the 1001th largest stock will be included in the Russell 2000 and be given a much larger weight in its index. For example, between 1998 and 2006, the average portfolio weight of the bottom 250 stocks in the Russell 1000 was 0.012%, while the average portfolio weight of the top 250 stocks in the Russell 2000 was an order of magnitude larger at 0.127%. The difference in portfolio weights persists over a wide range around the cutoff. This is seen in Figure 2, where we plot the end-of-June portfolio weights of the 500 smallest float-adjusted stocks in the Russell 1000 and the 500 largest floatadjusted stocks in the Russell 2000 for the year 2006. These differences in portfolio weights can have a significant impact on the extent of a stock’s ownership by passive investors. Because passive funds weight their holdings based on the portfolio weights of the underlying index in an attempt to minimize tracking error, it is more important that they match the weights of the stocks at the top of the index than of stocks at the bottom of the index. In other words, for each dollar invested in a passive fund benchmarked to the Russell 1000, very little of it will be invested in stocks at the bottom of that index, while for each dollar invested in a passive fund benchmarked to the Russell 2000, a large proportion of it will be invested in stocks at the top of the index. Because of the considerable amount of money passively tracking the two Russell indexes (Chang, Hong and Liskovich (2014)), the portfolio decisions of passive institutions can lead to large ownership differences in stocks around the Russell 1000/2000 threshold. The importance of index assignment for ownership by passive investors is illustrated in Figure 3, where we sort stocks using their end-of-May CRSP market capitalization and plot the average share of firms in the Russell 2000 and average end-of-September ownership by                                                                                                                                                                                                                                                                                                                                                                                               reconstitution. For more details regarding the reconstitution process and eligibility for inclusion in the Russell indexes, see Russell Investments (2013). 9 Beginning in 2007, Russell implemented a “banding” policy where firms within a certain range of the cutoff would not switch indexes. For example, a firm that was in the Russell 2000 index last year but was among the 1000 largest firms this year would only move to the Russell 1000 index if its market capitalization exceeded a certain threshold. Since our identification strategy relies on controlling for the factors that determine a firm’s index assignment each year, we restrict our attention to years prior to the implementation of this banding policy where only the end-of-May market capitalization calculated by Russell is used to determine firms’ index assignment. For a press release regarding the implementation of this banding policy by Russell, see https://www.russell.com/us/news/press-release.aspx?link=pressreleases/2007/PR20070403.htm, and for more details on how the banding thresholds are determined each year, see Russell Investments (2013). Passive Investors, Not Passive Owners -- Page 9

11) passively managed funds. The sample in Figure 3 contains the top 500 stocks of the Russell 2000 and bottom 500 stocks of the Russell 1000 for each year between 1998 and 2006, as determined using the end-of-June Russell-assigned portfolio weights within each index. By construction, the top-left panel of Figure 3 shows no break in size between the 500th and 501st largest stocks in this sample, but as shown in the middle-left panel, there is a rather large jump in the probability of being assigned to the Russell 2000 index around this break. The end-of-May market cap reported by CRSP does not perfectly predict a stock’s index assignment because Russell makes a number of adjustments when calculating its proprietary market capitalization values such that these values, which are used to determine a stock’s index membership, do not perfectly match market capitalizations reported in sources such as CRSP. And consistent with index assignment having an important impact on ownership, the bottom-left panel of Figure 3 demonstrates a distinct jump in the ownership of passive funds around this midway point. During our sample period, the total ownership stake of passive funds is, on average, 66% higher for a stock among the top 250 stocks of the Russell 2000 relative to a stock among the bottom 250 stocks of the Russell 1000 (p-value of difference < 0.001). The magnitude of the observed difference in passive ownership corresponds well to the magnitude one would predict using estimates of the total amount of passive assets tracking each of the two indexes. While the Russell 1000 is more than 10 times larger in total market cap than the Russell 2000 during our sample period, there is only about 2 to 3 times more dollars passively tracking the Russell 1000 relative to the Russell 2000 (see Table 1, Panel A of Chang, Hong, and Liskovich (2014)).10 Using their estimates for 2004, $38.9 billion in assets were passively tracking the Russell 2000, which accounts for about 3.14% of the index’s total market cap of $1,237 billion, while there was only $84.9 billion of assets passively tracking the Russell 1000, accounting for just 0.71% of the index’s total market cap of $12,002 billion. Based on these estimates, assignment to the Russell 2000 in that year would increase a stock’s passive institutional ownership by about 2.5 percentage points, which is similar to the 2.1 percentage point increase we detect in 2004 using our measure of passive ownership. In practice, the realized differences in passive ownership we detect will be slightly smaller around the cutoff than predicted by this simple back-of-the-envelope calculation because passive investments by some institutions, like pension funds, are not reported in the S12 mutual fund database.11 The importance of index assignment for passive ownership is further highlighted by looking at the ownership stake of three of the largest passive institutions during our sample period—Vanguard, State Street, and Barclays Bank (which owned iShares during our sample)— using the Thomson Reuters Institutional Holdings (13F) Database. On average, the ownership stake of each of these three institutions for the 250 firms at the top of the Russell 2000 relative to the bottom 250 firms of the Russell 1000 is a third higher, while the likelihood of each institution                                                                                                                         10 The disproportionate amount of money passively tracking the Russell 2000 occurs because the Russell 2000 is the most widely used market index for small cap stocks. The Russell 1000, which spans both large and midcap stocks, is less widely used as a benchmark because it faces more competition from other large cap and midcap market indexes, including the S&P 500 (which is the most popular market index), the CRSP U.S. midcap index, and the S&P 400 midcap index. 11 While using mutual fund holdings allows us to more precisely identify passive and active ownership stakes, a disadvantage is that it does not capture all institutional passive holdings. However, in Section VII.B, we demonstrate that our findings are not sensitive to instead using all institutional holdings, as reported in the Thomson Reuters Institutional Holdings (13F) Database, and to using alternative proxies for passive ownership, as was done in a previous version of this paper. Passive Investors, Not Passive Owners -- Page 10

12) owning more than 5% of a firm’s shares is two thirds higher and the likelihood of being a top 5 shareholder is 15% higher. We find no evidence that index assignment is related to ownership by actively managed funds. This is shown in the remaining two panels of Figure 3 where we plot the percent ownership for actively managed funds and mutual funds we are unable to classify. For each panel, including passive ownership, we scale the vertical axis to span a standard deviation on each side of the sample mean. As seen in those panels, there is no corresponding difference in either active or unclassified mutual fund ownership; we formally test and demonstrate this lack of a difference in other types of ownership in Section III.C. B. Identification strategy and empirical specification The construction of the Russell 1000 and 2000 indexes thus provides a source of exogenous variation in ownership by passive investors. Stocks at the top of the Russell 2000 exhibit greater ownership by passive investors because of their inclusion at the top of their index, while stocks at the bottom of the Russell 1000 do not. Because index assignment is determined by an arbitrary rule surrounding the market capitalization of the 1000th largest firm, this variation in ownership is plausibly exogenous after conditioning on firms’ market capitalization. We use an instrumental variable strategy to identify the effect of ownership by passive investors on firms’ corporate governance and corporate performance; in particular, we use inclusion in the Russell 2000 as an instrument for ownership by passive funds. Because index assignment is determined by a stock’s market capitalization, and because market capitalization may directly affect a stock’s institutional ownership for reasons separate from index assignment, we also include a robust set of controls for stocks’ end-of-May market capitalization in our estimation. Specifically, we estimate the following: N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit ) ) + γ Ln( Floatit ) + δ t + ε it n (1) n=1 where Yit is the outcome of interest for firm i in reconstitution year t (i.e. from end-of-June year t to end-of-June of year t+1) scaled by its sample standard deviation; Passive%it is the percent of a firm’s shares held by passively managed mutual funds at the end of the first quarter of the reconstitution year t (i.e., end of September) scaled by its sample standard deviation; Mktcapit is the end-of-May CRSP market capitalization of stock i in year t; and Floatit is the float-adjusted market capitalization calculated by Russell when initially setting the portfolio weights during the end-of-June reconstitution. We scale both Yit and Passive%it by their sample standard deviations such that the point estimate β can be interpreted as the standard deviation difference in Yit for a one standard deviation increase in Passive%it. We control for float-adjusted market capitalization because it is used by Russell to compute portfolio weights within each index and could be related to a firm’s stock liquidity, which may affect firms’ governance and other corporate outcomes (Back, Li, and Ljungqvist (2014); Edmans, Fang, and Zur (2013)). We also include reconstitution year fixed effects, δ t , to ensure that our estimates are identified using within-year variation in ownership and are not driven by the aggregate upward trend in ownership by passive investors (see Figure 1). Finally, we cluster the standard errors, ε , at the firm level.12                                                                                                                         12 We do not include firm fixed effects in our estimation since only a small fraction of our sample firms switch indexes at some point during the sample and because many of the governance and corporate outcomes we study are likely to be affected by sustained rather than transitory variation in passive ownership. Since firm fixed effects will remove this sustained variation, they will likely not capture the Passive Investors, Not Passive Owners -- Page 11

13) To account for the possibility that ownership by passive funds, as measured using Passive%, might be correlated with the error term, ε , because of the omitted variable issues discussed above, we instrument for ownership by passive funds using index assignment. Specifically, we instrument Passive% in the above estimation using R2000it, which is an indicator equal to one if stock i is part of the Russell 2000 index in reconstitution year t. As shown in Figure 3, being assigned to the Russell 2000 is associated with a significant jump in ownership by passive funds for stocks at the top of Russell 2000 relative to stocks at the bottom of the Russell 1000. Our IV estimation relies on the assumption that, after conditioning on stocks’ market capitalization, inclusion in the Russell 2000 index is associated with an increase in Passive% (relevance condition) but does not directly affect our outcomes of interest except through its impact on ownership by passive investors (exclusion restriction). We verify the relevance condition below in our first stage estimations, and the exclusion restriction seems reasonable in that it is unclear why index inclusion would be directly related to our outcomes of interest after robustly controlling for the factor that determines index inclusion—firms’ end-of-May market capitalization, as calculated by Russell. To control for firms’ market capitalization, we restrict our sample to the 250 stocks at the bottom of the Russell 1000 and top 250 stocks of the Russell 2000 and include a robust set of controls for firms’ log market capitalization, Ln(Mktcap), as measured using CRSP data, by varying the polynomial order N we use to control for end-of-May market capitalization.13 In later robustness tests, we also show robustness to varying the number of firms we include around the cutoff between the two indexes and to instead using end-of-May market caps to rank stocks and select our sample each year. The use of R2000it as an instrument allows us to isolate an exogenous source of variation in passive ownership. While non-index funds that passively seek to deliver the performance of a benchmark portfolio have discretion over which stocks to hold within the benchmark, the instrumental variable never uses such endogenous variation in passive ownership; the IV estimation only uses variation in ownership that is driven by a stock’s index assignment and the reshuffling of holdings by passively managed mutual funds seeking to minimize their tracking error. We do not use the actual portfolio weight or ranks of stocks as our instrument because this would introduce a potentially serious endogeneity concern. In particular, this is problematic because after Russell assigns stocks to an index, it determines actual weights using various endogenous factors, including liquidity and inside ownership. For the most illiquid, highest inside ownership stocks, Russell assigns a smaller portfolio weight than would be justified based purely on their end-of-May market capitalization so as to minimize the costs of institutions attempting to track the index. Because weights are related to a stock’s liquidity and inside ownership, it would be problematic to use them as instruments because both factors could directly affect the governance structures of firms.14                                                                                                                                                                                                                                                                                                                                                                                               relevant variation and thus potentially provide misleading inferences (e.g., see McKinnish (2008); Gormley and Matsa (2014)). 13 At some level, our estimation can be viewed as one that makes use of a threshold event in a non-RD estimation, as discussed in Bakke and Whited (2012). 14 This issue of why the actual weights or rankings should not be used as instruments or as part of a regression discontinuity is also discussed in Chang, Hong, and Liskovich (forthcoming) and Mullins (2014). Passive Investors, Not Passive Owners -- Page 12

14) C. First stage estimation In this section, we report estimates of our first-stage regression of passive mutual fund holdings on membership in the Russell 2000 index plus additional controls. Specifically, we estimate N Passive% it = η + λ R2000it + ∑ χ n ( Ln( Mktcapit ) ) + σ Ln( Floatit ) + δ t + uit n (2) n=1 where R2000it is a dummy variable equal to 1 if stock i is in the Russell 2000 Index for reconstitution year t (i.e., from end-of-June of year t to end of June year t+1). In our initial tests, we also analyze other outcome measures, including the percentage of shares outstanding owned by all mutual funds; the percentage of shares outstanding owned by actively managed funds; and the percentage of shares outstanding owned by unclassified mutual funds. The model is estimated over the 1998-2006 period, and uses a bandwidth of 250 firms and a third-order polynomial. The results, reported in Table 2, confirm that mutual fund ownership is related to membership in the Russell, particularly for passive mutual funds. So that the point estimates in Table 2 align with the observed differences in ownership shown in Figure 3, we do not scale the ownership variables by their sample standard deviations in these initial estimates. The first column shows that aggregate mutual fund ownership is significantly higher (at the 10% level) for the 250 stocks at the top of the Russell 2000 than for 250 stocks at the bottom of the Russell 1000. As expected, this relation appears to be driven entirely by passive funds: the estimated coefficient is positive and significant at the 1% level for the passive funds (column 2), but insignificant for actively managed and unclassified funds (columns 3 and 4). In Table 3 we demonstrate that the estimated relation between passive ownership and Russell 2000 membership is robust to using lower order polynomials, and to better quantify the economic magnitude of the observed difference in ownership, we scale Passive% by its sample standard deviation. Using a bandwidth of 250 firms and varying the polynomial order of controls for market cap, we consistently find an increase in ownership by passive funds of about a half of a sample standard deviation (Table 3, columns 1–3). In all cases, the increase is statistically significant at the 1% level.15 The lack of a difference in ownership for actively managed and unclassified mutual funds is also robust to varying the polynomial order of controls for Mktcap. This can be seen in Appendix Table 2. Consistent with actively managed funds being unaffected by a stock’s index assignment, we find no evidence of a difference in ownership by either actively managed funds or unclassified funds and the point estimates are economically small (between 1% to 5% of a their sample standard deviations). We also do not find evidence that membership in the Russell 2000 is associated with an increase in the visibility of a stock and subsequent analyst coverage, which is another mechanism by which index assignment might improve firms’ governance. In particular, if we re-estimate Equation (2) instead using the number of analysts as the dependent variable, we find no evidence that assignment to the top of the Russell 2000 is associated with greater analyst coverage; if anything, we find evidence that inclusion in the Russell 2000 is associated with less analyst                                                                                                                         15 Because our IV model is just-identified, the IV estimation is median-unbiased and weak instruments are unlikely to be a concern in our setting, especially given the strong first stage estimates (Angrist and Pischke, 2009). Additionally, the Kleibergen-Paap F stat on the excluded instrument exceeds 10, providing further confidence that a weak instrument is unlikely to be a concern (see Stock, Wright, and Yogo (2002) and Angrist and Pischke (2009)). Passive Investors, Not Passive Owners -- Page 13

15) coverage but the estimates are not robust to wider bandwidths. Likewise, Crane, Michenaud, and Weston (2014) find no evidence of an increase in media coverage among firms at the top of the Russell 2000. The lack of an increase in either analyst or media coverage among firms at the top of the Russell 2000 bolsters our assumption that index assignment in our setting will only affects firms’ governance structure through its effect on passive ownership. D. Why index assignment may matter A question that naturally arises is why index assignment might matter at all for firms’ passive ownership. If the increased ownership stake that comes with a stock being assigned to the Russell 2000 index allows passive investors to exert additional influence and correct a governance structure they deem suboptimal (as our findings below suggest), why would passive investors not also increase their ownership stake among stocks at the bottom of the Russell 1000 so as to exert more influence among those companies as well? In other words, what would prevent passive institutions from being more active, and hence, undoing the potential importance of index assignment? There are two likely explanations for why index assignment may matter for firms’ governance structures. First, passive institutions are simply more focused on minimizing expenses and tracking errors than on affecting governance. While increasing an ownership stake for one stock at the bottom of the Russell 1000 might not significantly affect a fund’s tracking errors relative to a Russell 1000 benchmark, a similar increase for a number of other stocks would. Moreover, such targeted activism would likely increase fund expenses since the passive investor would need to research which stocks to target. Combined these two effects would likely result in lost market share to competitors with lower costs and lower tracking errors. Second, index assignment may create a coordinated increase in ownership by passive institutions that might otherwise be hard to replicate. Achieving the same total increase in ownership stake may be prohibitively large for any one passive institution to achieve alone, and coordinating a combined ownership increase among multiple passive institutions may either be too costly or impose additional regulatory disclosure requirements these institutions wish to avoid. Overall, our finding that index assignment corresponds with a shift in passive ownership suggests that passive institutions and the funds they manage are not active in the traditional sense of trying to accumulate or exit positions since such actions would undo the importance of index assignment. We now turn to analyzing whether passive ownership and index assignment affect firms’ governance structures and the potential mechanisms by which passive investors may exert influence. IV. How passive investors affect firms’ corporate governance To select the governance outcomes for our analysis, we start from a 2014 speech given by the Chairman and CEO of Vanguard, Bill McNabb, that summarizes the broad governance issues on which Vanguard focuses. These issues include “Independent oversight” (i.e., board independence), “Annual director elections and minimal anti-takeover devices”, “Shareholder voting rights consistent with economic interest” (i.e., no dual class share structures that provide disparate voting rights to different groups of shareholders), and “Sensible compensation tied to performance”.16 We then compare these issues to the proxy voting policies of Vanguard and                                                                                                                         16 This speech, which can be found at http://www.lerner.udel.edu/sites/default/files/WCCG/PDFs/events/Transcript%20_UDel%20Corp%20Governance%2010%2030%202014_%20FINAL%20for%20UD%20websi te.pdf, is based off the governance principles Vanguard states on its website at Passive Investors, Not Passive Owners -- Page 14

16) other large passive investors during our earlier sample period by obtaining the initial proxy voting policies provided to the SEC by Vanguard, State Street, and Barclays Bank when the filing of such policies was first required beginning on July 1, 2003. From these proxy-voting policies, it is clear that these four broad governance issues were also a focus of passive investors during our sample period. In particular, the largest passive investors (1) supported greater board independence, (2) opposed takeover defenses, (3) opposed unequal voting rights, as occurs when firms maintain a dual class share structure, and (4) supported compensation plans that align management’s interests with shareholders and avoid excessive awards (see Appendix).17 But, do passive investors, whose impact tends to be limited to “voice,” have an effect on these aspects of corporate governance? In this section, we investigate these questions using the identification strategy and instrumental variable estimation described in Section III to analyze their impact on three of these issues: board independence, takeover defenses, and equal voting rights. We will analyze their impact on the fourth issue, the level and structure of executive compensation, in Section VI.B. A. Independent directors We first assess whether passive institutions exert influence on board independence. Increasing the percent of independent directors was a specific concern of many passive investors during our sample period (see Appendix) and is one dimension of governance where passive investors have a direct say via their proxy votes in director elections. Passive investor support for independent directors likely stems from the belief that independent directors are more likely to be effective monitors (Fama and Jensen (1983), Weisbach (1988)). Table 4 reports results for our IV estimation using percentage of independent director scaled by its sample standard deviation as the dependent variable. We find that passive investors do indeed have a significant impact on this key dimension of corporate governance. We find a statistically significant positive relation (at the 1% level) between Passive% and the percentage of independent directors that is robust to various polynomial order controls for market capitalization. The economic magnitude of the relation is sizable. A one standard deviation increase in ownership by passive funds is associated with a 0.65 to 0.76 standard deviation increase in the share of independent directors on a firm’s board (Table 4, columns 1–3). In unreported analysis, we find this increase in director independence is not driven by an increase in board size; to the contrary, greater ownership by passive funds is associated with smaller boards.18                                                                                                                                                                                                                                                                                                                                                                                               https://about.vanguard.com/vanguard-proxy-voting/corporate-governance/index.html. Two other governance issues that were discussed in this speech, but are not as easily tested, are “Accountability” (of both the board and management) and “Shareholder engagement”. 17 These historical proxy voting guidelines can be found on the SEC website and are summarized in the Appendix. Other popular governance issues, like splitting the positions of CEO and Chairman of the Board, however, are not mentioned in either the Vanguard speech or the proxy voting guidelines of the largest passive investors. At some level, this particular exclusion is not surprising since some passive institutions (e.g. Vanguard) have the same individual act as both CEO and Chairman, and consistent with passive investors not holding a view on this issue, we find no association between passive ownership and whether a company’s CEO serves as Chairman of the Board. 18 Because Riskmetrics only covers firms in the S&P 1500, the sample size in Table 4 is about a third smaller than the first stage estimates reported in Table 3. However, this reduced sample size does not pose a problem for our estimation. The first stage estimates in the smaller sample of observations with non-missing director data remain statistically significant at the 1% level. This can be seen in Appendix Passive Investors, Not Passive Owners -- Page 15

17) The impact of passive investors on board independence is even larger prior to changes regarding board independence requirements at the NYSE and Nasdaq exchanges. In late 2002, both exchanges proposed changes to require that all firms listed on the exchange have a majority of independent directors, and the SEC approved the proposed changes in 2003. Consistent with passive investors having more of an influence on board independence prior to 2003, we find that a one standard deviation increase in ownership by passive funds is associated with a 1.3 to 1.4 standard deviation increase in share of independent directors on a firm’s board prior to 2003 (Table 5, columns 1–3) but only a 0.26 to 0.35 standard deviation increase after 2002 (columns 4–6). The differences in the estimates across time period are statistically significant at the 1 percent confidence level.19 B. Takeover defenses We now consider the association between passive investors and takeover defenses. Opposition to takeover defenses, including poison pills and restrictions on shareholders’ ability to call special meetings, were a common theme of passive investors’ proxy voting guidelines during our sample (see Appendix). While poison pills may be in shareholders’ interests under some circumstances, they are often seen as a mechanism used to shelter managers from the disciplining effects of hostile takeovers. Specifically, poison pills (formally known as “shareholder rights plans”) effectively bar any single shareholder from acquiring more than a pre-defined percentage of shares (often between 10% and 15%) without significantly diluting their holdings (Bebchuk, Cohen, Ferrell (2009)). While Coates (2000) notes that essentially every firm has a “shadow pill” in place because a pill may be implemented by a board at any time without shareholder approval, having a poison pill in place is still thought to provide managers with advantages in fighting off hostile bids and unwanted activists.20 Moreover, institutional investors widely call for the redemption of poison pills and support efforts to subject them to shareholder votes in order to improve the accountability of managers and boards. We find evidence that ownership by passive funds is associated with an increase in the removal of poison pills. To determine the influence of passive institutions on the removal of poison pills, we estimate equation (1) with an indicator variable equal to one if the firm’s poison pill is either withdrawn or allowed to expire and zero otherwise. These estimates are reported in Table 6. The estimated coefficient is positive and statistically significant (at the 1% level). A one standard deviation increase in Passive% is associated with a 0.18–0.20 standard deviation (i.e., 3.3–3.8 percentage point) increase in the likelihood of a poison pill being removed (Table 6, columns 1–3). The estimate is economically sizable given that, on average, only 4% of firms remove a poison pill each year. We next analyze whether ownership by passive investors is associated with a greater ability for shareholders to call a special meeting, another important aspect of governance (Daines and Klausner (2001); Cremers and Nair (2005)). Similar to poison pills, restrictions on                                                                                                                                                                                                                                                                                                                                                                                               Table 3A. The first stage estimates for our later estimates in Tables 6, 7, and 8 can be found in Appendix Tables 3A-3B. We do not separately report first stage estimates for Tables 9 and 10 since their samples are comparable to that used in Table 3. 19 While the proposed exchange listing requirements did not become effective until 2004, many firms began complying in 2003. Given this, we follow Chhaochharia and Grinstein (2009) and use the year 2003 as the potential breaking point; see Chhaochharia and Grinstein (2009) for more details. 20 As noted by Bebchuk, Cohen, and Ferrell (2009), “having a pill in place saves the need to install it in ‘the heat of battle’… [and] signals to hostile bidders that the board ‘will not go easy’.” Passive Investors, Not Passive Owners -- Page 16

18) shareholders’ ability to call special meetings can represent a potential impediment to effective governance by delaying dissident shareholders’ ability to remove directors, and such restrictions, especially if combined with a poison pill, are also seen as an effective takeover defense for entrenched managers (Daines and Klausner (2001)). To assess the ability of passive institutions to reduce restrictions on shareholders’ ability to call special meetings, we estimate equation (1) with an indicator variable equal to one if the firm eliminates such restrictions, and zero otherwise. These estimates are reported in columns 4–6 of Table 6. We find evidence that ownership by passive funds is associated with the removal of restrictions on shareholders’ ability to call special meetings. The estimated coefficient is positive and statistically significant (at the 1% level) in all of the estimations; in particular, a one standard deviation increase in passive ownership is associated with about a 0.30–0.34 standard deviation (i.e., 2.4–2.7 percentage point) increase in the likelihood that a firm eliminates restrictions on shareholders’ ability to call special meetings. Relative to the average share of firms that lift restrictions each year in our sample, which is about 0.6%, the estimated magnitude is sizable. In unreported analysis, we also analyzed the impact of passive ownership on whether firms have annual director elections. Staggered director elections and classified boards are another type of takeover defense that passive institutions typically oppose (see McNabb (2014) and Appendix). We find suggestive evidence that passive ownership is also associated with firms being less likely to have a classified board, but the estimates are not statistically significant at conventional levels. The statistically weaker results for classified boards may partially be an artifact of the time period of our sample; Guo, Kruse, and Nohel (2008) note that shareholder efforts to de-classify boards intensified significantly in 2003 following the passage of SarbanesOxley. Consistent with this possibility, we find stronger evidence that passive ownership is associated with firms being less likely to have a classified board after 2003, but the estimates are only statistically significant in bandwidths wider than the 250 stocks around the threshold we use in our main analysis.21 C. Equal voting rights and dual class share structures Finally, we analyze whether ownership by passive investors is associated with the voting rights of shareholders. Passive institutions uniformly oppose dual class share structures and any other form of unequal voting rights and often threaten to withhold support for managers or directors of any company that does not provide equal voting rights to all shareholders (see McNabb (2014) and the Appendix for examples). Passive institutions also state they will refuse to support any attempts by companies to implement a dual class share structure (as might occur during a merger). Moreover, by concentrating voting power among insiders, Klausner (2012) argues that dual class share structures are one of the most powerful takeover defenses, providing yet another reason passive investors oppose them, and Gompers, Ishii, and Metrick (2010) find evidence that dual class share structures can negatively impact firm value. To assess whether ownership by passive investors is associated with a firm being less likely to have unequal voting rights, we construct an indicator that equals one if the firm has a dual class share structure, and zero otherwise, as determined by Riskmetrics. These estimates                                                                                                                         21 In unreported analysis, we also analyzed the impact of passive ownership on super majority vote requirements, a fourth antitakeover device specifically mentioned in Vanguard and State Street’s historical proxy voting policies. We find a negative association between passive ownership and the likelihood a firm has supermajority voting requirements, but the point estimates are neither statistically significant nor economically large. Passive Investors, Not Passive Owners -- Page 17

19) are reported in Table 7. We find evidence that ownership by passive funds is associated with firms being less likely to have a dual class share structure. The estimated coefficient is negative and statistically significant (at the 1% level) in all of the estimations; a one standard deviation increase in Passive% is associated with about a one standard deviation decrease in the likelihood that a firm has a dual class share structure.22 Another voting rights issue that is discussed in the proxy voting guidelines of passive investors is their opposition to cumulative voting. As Vanguard’s states in its proxy voting guidelines, cumulative voting can allow “shareholders a voice in director elections that is disproportionate to their economic investment in the corporation.” However, in unreported analysis, we do not find an association between passive ownership and whether firms have cumulative voting for directors. V. Possible mechanisms by which passive investors influence governance A key mechanism by which passive investors might influence a firm’s governance structure is via their voice. In particular, passive investors may use their ownership stake and ability to vote in order to monitor firms and ensure conformity with their views on governance structures. Alternatively, it is also possible the passive investors’ influence is not the result of them being active owners. Instead, passive investors’ concentrated ownership may facilitate activism by others, such as hedge funds, by lowering the costs for other activists attempting to coordinate votes against management (Brav et al. (2008), Bradley et al. (2010)). In this section, we investigate these two possible channels. A. The power of passive investors’ “voice” To address whether passive investors’ exercise voice and influence firms’ governance through their large voting blocs, we first analyze support for management proposals. Shareholder voting at annual meetings is a fundamental duty of shareholders, and votes against management proposals can be a proxy for increased monitoring by shareholders (Easterbrook and Fischel (1983)). To assess whether passive institutions influence voting outcomes, we estimate equation (1) with the dependent variable defined as the average percentage of shares that vote in support of management proposals. Consistent with an increased monitoring of managers and with passive investors exercising voice, we find that greater ownership by passive funds is associated with less support for management proposals (Table 8, columns 1–3). The estimated coefficients are negative and statistically significant (at the 1% level). A one standard deviation increase in ownership by passive funds is associated, on average, with about a 0.75 standard deviation decline in support for management proposals. Consistent with passive investors being active in monitoring managers, management appears to be confronted with a more contentious shareholder base when passive funds, which are less able to vote with their feet, make up a larger percentage of the ownership.                                                                                                                         22 Because adding a dual class share structure is typically not allowed by stock exchanges after a firm’s initial IPO, the observed difference in dual class structures is most likely driven by firms removing a dual class share structure rather than failing to add one. Consistent with this, in unreported estimates we find that passive ownership is positively associated with the removal of dual class shares, but unlike our findings for poison pills and restrictions on shareholder meetings, the estimates are not statistically significant at conventional levels. This is likely attributed to the relatively small number of companies that make such changes following their initial public offering; on average, only about 0.9 percent of firms remove a dual class share structure each year in our sample. Passive Investors, Not Passive Owners -- Page 18

20) The decline in support for management proposals does not originate from a shift in the number or type of management proposals put to a vote. In unreported analysis, we find that greater ownership by passive funds is not associated with a change in the total number of management proposals, and we find little evidence of an association with the composition of proposals. The lack of difference in the composition of proposals suggests the lower support for management proposals is not driven by managers submitting a greater number of lessshareholder-friendly proposals. We next analyze support for governance-related shareholder proposals. While these proposals are non-binding, they potentially increase pressure on boards to make changes to their firms’ governance structure (Del Guercio and Hawkins (1999)). If passive investors use such votes to exercise voice and influence, we might expect to observe an increase in support for such proposals. In further support of passive investors exercising voice via their votes, we find evidence that ownership by passive funds is associated with an overall increase in support for governancerelated shareholder proposals. On average, a one standard deviation increase in ownership by passive funds is associated with a 0.49–0.65 standard deviation increase in support for governance proposals (Table 8, columns 4–6). While the increase in support is only statistically significant at the 10% level when adding second- or third-order polynomial controls (p-values 0.062 and 0.064, respectively), the implied magnitudes are economically large. The lower statistical significance likely reflects the relatively small number of such governance proposals. Similar to management proposals, we find no systematic relation between ownership by passive funds and differences in the types of shareholder proposals voted on. B. No increased activism by others An alternative mechanism by which passive ownership might influence firms’ governance structure is by facilitating activism by other, non-passive investors. In particular, the size and concentration of passive investors’ ownership stakes may increase activist investors’ ability to rally support for their demands (Brav et al. (2008), Bradley et al. (2010)). Such added pressure from activist investors might also explain a number of the governance differences we observe. In other words, is it possible that the observed differences in governance are not driven by passive investors being engaged owners, but rather, driven by their ownership stake making it easier for others to engage in activism? We find no evidence, however, that greater ownership by passive investors is associated with more activism by non-passive institutions; instead, we find evidence of less activism by non-passive institutions, consistent with passive investors monitoring managers and reducing the need for activism by other investors. To demonstrate this, we estimate equation (1) with an indicator variable equal to one if the firm experiences a hedge fund activism event, as defined in Brav, Jiang, Partnoy, and Thomas (2008) and Brav, Jiang, and Kim (2010), and zero otherwise.23 These estimates are reported in Table 9. The point estimates are negative and statistically significant. We find that a one standard deviation increase in passive fund ownership is associated with a 0.13–0.16 standard deviation (i.e., 1.6–2.0 percentage point) decline in the                                                                                                                         23 We thank Alon Brav for making these data on hedge fund activism events available to us. The database is an updated sample [1994-2011] using the same data collection procedure and estimation methods as in Brav, Jiang, Partnoy, and Thomas (2008) and Brav, Jiang, and Kim (2010). For more information on how the database is constructed, please see https://faculty.fuqua.duke.edu/~brav/HFactivism_SEPTEMBER_2013.pdf. Passive Investors, Not Passive Owners -- Page 19

21) likelihood of hedge fund activism. This magnitude is large given that a firm’s likelihood of an activism event in a given year in our sample is, on average, only 1.6%.24 While the observed decline in activism by non-passive institutions is consistent with passive investors successfully affecting governance outcomes and reducing the need for activism by others, it does not negate the possibility that the concentration of passive institutions’ ownership stakes increases the threat of activism by others, or that this threat increases the influence of passive investors’ “voice”. Concerned about an increased threat of activism, managers may be particularly responsive to the views of passive investors and be taking actions to preempt an actual activist campaign. Anecdotal evidence suggests that informal discussions between passive institutions and managers, backed up with the threat of voice, are often used to exert influence.25 VI. Do passive investors affect performance, compensation, or other corporate policies? Ownership by passive investors might also be associated with differences in firm performance, managerial compensation, or corporate policies. Overall performance or corporate policies might differ if the observed differences in governance associated with passive investors help mitigate managerial agency conflicts or if managers adjust corporate policies so as to preempt hedge fund activism campaigns that rely on the support of passive investors. We now explore this possibility. A. Overall performance There is considerable debate about the value implications of various governance structures or whether the potential influence of passive investors will necessarily improve firm performance. Because greater board independence, fewer takeover defenses, and equal voting rights arguably increase shareholder rights, one might expect that passive ownership mitigates agency conflicts and is associated with improved performance. However, theory suggests that board independence might be a result rather than a cause of performance (Hermalin and Weisbach (1998)), and the empirical evidence regarding the performance implications of board independence is mixed (e.g., Bhagat and Black (2002); Hermalin and Weisbach (2003)). Likewise, the value implication of removing poison pills and other takeover defenses is debatable (e.g., Stein (1988), Coates (2000)). More broadly, one might also argue that the optimal governance structure may vary considerably across firms (e.g., Coles, Daniel, Naveen (2008); Duchin, Matsusaka, Ozbas (2010)), and hence, the potential “one-size-fits-all” governance view of passive investors may not always represent an improvement for individual firms.                                                                                                                         24 In further support that the observed differences in governance are not driven an increase in activism by other, non-passive investors, we find no evidence that firms with greater passive ownership are more likely to be the target of a takeover attempt, another mechanism by which activists might exercise influence. 25 Glenn Booraem, controller of Vanguard funds, notes that engagement with directors and management of companies is a key component of Vanguard’s governance program, and that Vanguard has “found through hundreds of discussion every year” that it is “frequently able to accomplish as much—or more— through dialogue” as through voting (see Booream (2013)). And in a speech from October 2014, the CEO and Chairman of the Vanguard group, F. William McNabb, noted that Vanguard sent out 923 letters to firms in 2013, 358 of which requested specific changes in governance, and that 80 of these companies had adopted substantive changes without having to go through a shareholder proposal (see McNabb (2014)). Earlier findings regarding activism by TIAA-CREF also confirm the importance and impact of such private negotiations (Carleton, Nelson, and Weisbach (1998)). Passive Investors, Not Passive Owners -- Page 20

22) Consistent with the governance structure promoted by passive investors having a positive impact on performance for the average firm, we find evidence that ownership by passive funds is related to an overall improvement in firms’ future performance, as measured using firms’ ROA. Although passive fund ownership is not associated with significant differences in firms’ overall ROA in our main specification (Table 10, columns 1-3), it is positively associated with firms’ ROA after adding controls for whether a firm switched indexes that particular year (columns 46). This is likely because improvements in performance may take time to manifest, and one would not expect to find a relation between changes in passive fund ownership and performance for firms that just switched indexes. Consistent with this, we find that adding controls for such recent movers reveals a positive and statistically significant association between passive ownership and ROA. On average, a one standard deviation increase in passive fund ownership is associated with about a 0.31–0.41 standard deviation increase in long-term ROA. In unreported estimates, we also find that passive fund ownership is positively associated with Tobin’s Q, another commonly used measure of firm performance.26 B. Executive compensation There has been much debate regarding managerial pay and whether its growth reflects an efficient market outcome or an agency conflict, and passive investors commonly discuss the importance of the need to properly reward and incentivize managers while avoiding “excessive” pay (see McNabb (2014) and the Appendix). To assess whether passive fund ownership affects CEO compensation structure, we examine total CEO pay, its composition, and the sensitivity of CEO pay to stock price movements. We find less evidence that ownership by passive funds is associated with a difference in overall managerial pay or its composition. In unreported analysis, we find that while Passive% is negatively associated with total pay, the estimates are not statistically significant except in wider bandwidths. We also find little evidence that passive fund ownership is associated with differences in the composition of managerial pay (salary, bonuses, and grants of restricted stock, each scaled by total pay) or the sensitivity of pay to stock price movements (as measured using the delta or vega of the manager’s stock portfolio; see Gormley, Matsa, and Milbourn (2013) for variable definitions). Thus, for our sample at least, passive institutions appear to have a relatively little impact on executive compensation. However, it is important to note that our sample predates the implementation of “Say on Pay” by the Dodd-Frank Act in 2010. This provision, which requires nonbinding votes on executive pay packages, potentially provides an added mechanism for passive investors to influence compensation decisions. C. Cash, dividend, financing, and investment policies There is an extensive literature addressing the relation between corporate ownership structure and corporate policies; for example, agency theories suggest that better monitoring by shareholders might lead to changes in leverage, acquisitions, cash levels, and payout policies (Jensen (1986), La Porta et al. (2000)). To examine whether ownership by passive investors is associated with differences in these other corporate policies, we estimate equation (1) with standard measures of financing, investment, cash, and dividend policies as the outcome variable. We find relatively little evidence that ownership by passive funds is associated with                                                                                                                         26 Similar to ROA, we find a positive association between passive ownership and Tobin’s Q only after controlling for whether a firm switched indexes that year. Importantly, our earlier estimates for governance and vote outcomes are unaffected by the inclusion of the additional controls for whether a firm switched indexes that year. These robustness tests are discussed in Section VII.A. Passive Investors, Not Passive Owners -- Page 21

23) corporate policies related to investment, capital structure, or cash holdings. In unreported results, we find no difference in firms’ leverage, capital expenditures, R&D expenses, cash-toasset ratio, or acquisitions. These findings are consistent with anecdotal evidence that passive investors lack the resources necessary to research and influence corporate policies that are inherently more firm-specific. We do, however, find weak evidence that passive ownership is associated with higher dividends. In unreported analysis, we find that a one standard deviation increase in Passive% is associated with about a 0.15 standard deviation increase in firms’ dividend yield (significant at 10% level in some specifications). We find qualitatively similar results if we instead use a payout ratio and scale firms’ annual dividends by their net income, but the estimates are not statistically significant at conventional levels. VII. Additional robustness checks and choice of specification In this section, we discuss the robustness of our IV estimates. In particular, we demonstrate that our findings are not sensitive to how we measure end-of-May market caps, to adding additional controls, to varying the sample bandwidth around the threshold, to using alternative definitions of passive institutional ownership as our key explanatory variable, or to using end-of-May market cap rankings to select our sample of stocks each year. We also address the possibility of a selection bias around the Russell 1000/2000 threshold, particularly for the subsample of observations covered by Riskmetrics. A. Robustness to choice of controls, choice of bandwidth, and placebo tests The assumption of our identification strategy is that after limiting the sample to stocks close to the Russell 1000/2000 threshold and controlling for the one factor that determines index membership (i.e., end-of-May market cap), index membership does not directly affect our outcomes except through its effect on passive ownership. This is the exclusion restriction of the IV estimation. However, because Russell Investments uses a proprietary method to calculate firms’ total market caps, we are only able to imperfectly control for the underlying market cap used to determine index assignment.27 Our findings, however, are robust to using alternative ways to measure firms’ end-ofMay market cap. In particular, using the noisy end-of-May market caps obtained directly from Russell to measure Mktcap does not affect our findings. This is shown in Appendix Table 4, where we re-estimate our main IV regressions for the period 1998-2006 using the 250 bandwidth with third-order polynomial controls for Ln(Mktcap) after replacing the CRSP market cap with the Russell-provided market cap for the years 2002-2006. The estimates are nearly the same as before. Our findings are also robust to instead using the Compustat security monthly file to determine end-of-May market cap (see Appendix Table 5). Our findings are also robust to including various controls. Adding 2-digit SIC industry fixed effects to the specification does not affect our findings (see Appendix Table 6). Our findings are also largely unaffected if we add controls to account for firms that moved from the                                                                                                                         27 According to Russell’s documentation, their proprietary calculation of market capitalization includes some ownership stakes, like common stock, non-restricted exchangeable shares, and partnership units, but excludes other forms of shares, such as preferred stock or redeemable shares (Russell 2013). The share price chosen by Russell to compute market capitalization can also vary for firms that have multiple share classes or did not trade on the last day of May. Similar to Mullins (2014), we contacted Russell Investments and were only able to obtain a noisy measure their proprietary measure of market capitalizations for the years 2002 through 2006. Russell does not have the data prior to 2002. See Mullins (2014) for more details regarding the likely sources for this noise. Passive Investors, Not Passive Owners -- Page 22

24) Russell 1000 to the Russell 2000 that year, and vice versa. If such switchers differ in other dimensions and represent a disproportionate share of either index, this could affect our earlier estimates. However, all of the findings are robust to the inclusion of these controls (see Appendix Table 7). In unreported analysis, we find that our estimates are unaffected by the inclusion of additional controls for a stock’s liquidity, such as the Amihud measure of illiquidity or a stock’s average bid-ask spread. Our estimates are also robust to our choice of bandwidth around the Russell 1000/2000 threshold. This is shown in Appendix Figure 1, where we plot the point estimates and 95th percentile confidence intervals when varying the bandwidth between 100 and 500 firms and using a first-order polynomial control for Ln(Mktcap); estimates are reported for both the first stage and IV specifications of Tables 3-10. The estimates are relatively similar across the various bandwidths, and there is no evidence to indicate that our findings are sensitive to the choice of bandwidth. Finally, in further support that our findings are not driven by omitted variables that may be correlated with firms’ end-of-May market cap, we do not find an association between passive ownership and our outcomes of interest in placebo IV or reduced form tests that use alternative thresholds. For example, if we restrict the sample to the top 500 firms of the Russell 2000, and replace our R2000 indicator with an indicator for the bottom 250 firms of this subsample, our IV estimation does not detect an effect of passive ownership on any of our outcomes, nor do we find any of our findings in a reduced form estimation of the outcomes onto R2000. Likewise, we do not find an effect of passive ownership in a similar placebo test that uses the bottom 500 firms of the Russell 1000. B. Robustness to alternative definitions of passive ownership For our analysis above, we measure the ownership stake of passive investors by summing up the ownership of mutual funds we classify as passively managed. A key advantage of using the Thomson Reuters Mutual Fund Holdings Database is that it allows for a precise measure of passive ownership. A disadvantage of the fund-level data, however, is that it misses the holdings of passive institutional investors that do not manage mutual funds or ETFs. The Thomson Reuters S12 mutual fund data we use exclude holdings by banks, insurance companies, and pension funds, some of which might also adopt passive investment strategies. While the exclusion of these passive institutions does not affect the validity of our IV estimation, it does mean one must be more careful in interpreting the IV point estimates. In particular, attempting to back out the implied change in governance structure for a given percentage change in passive ownership might lead to an overestimation of the actual economic magnitude of interest.28 To avoid this potential concern, we scale our measure of Passive% by its sample standard deviation so that point estimates instead reflect the observed difference in governance for a one standard deviation                                                                                                                         28 For example, if the reduced form estimation of board independence detects a 4.87 percentage point increase in the share of directors classified as independent for stocks in the Russell 2000 and the first stage estimation detects a 0.94 percentage point increase in holdings by passively managed mutual funds, then the IV estimate for board independence will equal 4.87/0.94 = 5.18. In other words, the IV estimate will indicate that a one percentage point change in passive ownership causes a 5.18 percentage point increase in board independence. But if the true increase in passive ownership for stocks assigned to the Russell 2000, after accounting for passive investors not accounted for in the mutual fund data, is instead 2.1 percentage points, then the true effect of a one percentage point increase in passive ownership on board independence would be 4.87/2.1 = 2.32 percentage points. Passive Investors, Not Passive Owners -- Page 23

25) difference in passive fund ownership. Under the assumption that the standard deviation change in passive fund ownership for stocks assigned to the Russell 2000 would be similar with the inclusion of any passive investors not captured by the S12 data, the point estimates we obtain in the scaled regression will accurately reflect the economic magnitude of interest. Using a broader measure of passive ownership, however, has no effect on our findings. To illustrate this, we obtain data on institutional holdings from the Thomson Reuters Institutional Holdings (13F) Database. Any financial institution exercising discretionary management of investment portfolios over $100 million in qualified securities is required to report its aggregate holdings quarterly to the SEC using Form 13F, and consistent with this capturing a larger share of institutional ownership than the S12 data, we find that institutional holdings account for about 70% of market capitalization compared to the 25% of market capitalization accounted for by mutual funds in the S12 data. We then classify institutions as either passive or active using Bushee’s (2001) classification of institutions. In particular, we classify “quasi-index” institutions as passive and “transient” or “dedicated” institutions as active.29 Using this alternative measure of passive and active ownership, we repeat our first stage and IV estimations. These estimates are reported in Tables 11 and 12. As further evidence that only passive investors adjust their holdings to index assignment, our first stage estimates only detect an increase in “quasi-index” ownership (which includes some of the largest passive investors, like Vanguard, State Street, and Barclay’s Bank) and no increase in “transient” or “dedicated” ownership. Moreover, as shown in Table 12, our IV estimates when using Quasi-index% remain qualitatively similar.30 Our findings are also robust to using alternative definitions of passive investors. In particular, if we instead measure passive ownership as just the sum of holdings by Barclays Bank, State Street, and Vanguard, we get similar findings. In unreported first stage estimates, we find that being assigned to the Russell 2000 is associated with a very large and statistically significant increase in the combined holdings of these three passive institutions; they account for more than half of the increase in Quasi-index ownership shown in Table 11. Moreover, our IV estimations remain large and statistically significant when we use the combined ownership of these three firms as the explanatory variable instead of all quasi-index ownership. This can be seen in Appendix Table 8. These findings provide additional confidence that our earlier estimates are capturing the influence of passive investors and that the IV estimation is not sensitive to how we measure passive ownership. C. Robustness to alternative sampling choices                                                                                                                         29 To avoid changes in the classification of an institution over time, we use Bushee’s “permanent” classification. Our findings, however, are similar if we do not use Bushee’s “permanent” classification and instead use the time-varying classifications provided by Bushee or restrict the measure of passive ownership to institutions that are classified as a quasi-indexer in every year of our sample period. 30 Similar to before, one must be cautious in interpreting the economic magnitudes of these estimates. Because the Bushee (2001) “quasi-index” classification includes some active investors and actively managed mutual funds, the first stage estimates for the implied standard deviation change in Quasiindex% likely understate the true standard deviation change in passive holdings. In particular, the 2.3 percentage point increase in quasi-index holdings found in Table 11, Column (2) corresponds to about a 0.14 standard deviation change in Quasi-index%, which is considerably smaller than the 0.5 standard deviation change detected when using a more precise measure of passive holdings. This smaller first stage estimate will cause the IV estimates to be inflated when using Quasi-index% scaled by its sample standard deviation as the explanatory variable to be instrumented. Passive Investors, Not Passive Owners -- Page 24

26) In our main analysis, we select our sample to be the 250 stocks with the smallest portfolio weights in the Russell 1000 and the 250 stocks with the largest portfolio weights in the Russell 2000. Our findings, however, are not sensitive to instead using end-of-May market caps to determine the sample of stocks each year. In particular, we can instead rank stocks based on their end-of-May market cap, as calculated using CRSP, and select the sample for each year of the sample using firms ranked 750th through 1250th that year. An advantage of this latter approach is that it eliminates the risk that Russell’s float-adjusted reweighting of stocks within an index affects our findings. A disadvantage of this approach, however, is that we are no longer necessarily comparing the very bottom firms of the Russell 1000 against the very top firms of the Russell 2000, which is where we would expect to find the biggest difference in passive ownership (and hence, outcomes) to occur. This sampling choice, however, has little impact on our IV estimates. While the first stage estimates are expectedly smaller in magnitude when we use end-of-May market caps to rank stocks and select our sample each year (coefficient = 0.383, t-stat = 9.54), the IV estimations are largely unchanged (see Appendix Table 9). D. Ruling out potential sample selection biases One potential concern with our analysis is the possibility of systematic differences (beyond market capitalization, which we control for) in the type of stocks on the two sides of the Russell 1000/2000 threshold. For example, one might worry that “fallen angels” (stocks experiencing significant declines in stock price) are more likely to appear in the Russell 2000, while “rising stars” (stocks experiencing significant increases in stock price) are more likely to appear in the Russell 1000. Additionally, this concern could be particularly relevant for our analysis that uses the Riskmetrics data, which is largely limited to firms in the S&P 1500, if there are systematic differences in the likelihood of fallen angels or rising stars from the Russell 2000 sample being included in the S&P 1500 (and hence, Riskmetrics) relative to the likelihood of fallen angels or rising stars from the Russell 1000 sample being included in the S&P 1500. If present, such difference could cause a violation of the exclusion restriction. However, we do not find significant evidence of differences in the lagged returns (prior to reconstitution) between the stocks on different sides of the Russell 1000/2000 threshold. Specifically, we find no evidence that “fallen angels” or “rising stars” are disproportionately represented in either index, including the subsample covered by Riskmetrics. To show this, we create indicator variables for “rising stars” and “fallen angels” that equals 1 if a stock’s return is in the top or bottom 5% of the sample for the 12 months before reconstitution (i.e., end-of-May in year t-1 to end-of-May in year t), respectively. We do not find statistically significant differences in the proportion of fallen angels and rising stars in the Russell 1000 versus the Russell 2000, either in the full sample or in the sample of observations covered by Riskmetrics. Moreover, we do not find evidence of a significant difference in either the lagged stock return or lagged change in stocks’ end-of-May market cap ranking between the stocks on either side of the Russell 1000/2000 threshold for the sample observations covered by Riskmetrics. While there is evidence in the full sample that stocks at the top of the Russell 2000 have a higher average lagged stock return, our main second-stage results are robust to controlling for past stock price returns (see Appendix Table 10) or other indicators for a large change in market cap over the past year. Finally, any selection into the Riskmetrics database based on dual class share structures cannot easily explain our findings regarding dual class structures. Since S&P requires firms to have a public float of at least 50% of the stock in order to be added to the S&P 1500, companies with dual class structures might be less likely to be included in the S&P 1500. However, because Passive Investors, Not Passive Owners -- Page 25

27) our analysis is limited to the subsample of firms covered by Riskmetrics, our estimates indicate that among firms covered by Riskmetrics, more passive ownership is associated with a firm being less likely to have a dual class share structure. In other words, any type of selection (if it existed) would not be a problem for our analysis. VIII. Conclusion Passive institutions, like Vanguard and State Street, are an increasingly important component of U.S. stock ownership, and the impact of their growth on firm-level governance is widely debated. Despite arguments that they may be lazy investors that lack both the motivation and resources to monitor managers, there are multiple reasons why passive investors may have a vested interest in affecting firms’ governance structures and performance and why their large ownership stakes might make them an influential voice in decisions pertaining to firms’ governance structures. To examine whether passively managed mutual funds affect firms’ governance, and if so, by which mechanisms, we exploit variation in passive institutional ownership that occurs around the cutoff used to construct the Russell 1000 and Russell 2000 indexes. Benchmarking to these indexes leads to about a 66% difference in passive ownership for stocks at the top of the Russell 2000 relative to stocks at the bottom of the Russell 1000. Thus, we instrument passive institutional ownership with an indicator for being assigned to the Russell 2000 in a given year and analyze the influence of passive investors in an economically important sample of large U.S. publicly listed firms. Our instrumental variable estimation relies on the assumption that after conditioning on firms’ market capitalization, which determines index assignment, inclusion in the Russell 2000 index does not directly affect our governance or corporate outcomes except through its impact on ownership by passive investors. Our findings suggest that while passive institutional investors are not active owners in the traditional sense of accumulating or selling shares in a target company with the express purpose of influencing management, they are not passive owners either. In particular, we find that ownership by passively managed mutual funds is associated with more independent directors on a board, fewer takeover defenses, and more equal voting rights, as captured by a firm being less likely to have a dual class share structure. The observed differences in actual governance structures suggest that passive institutions are attentive to firms’ corporate governance, and that they use their large voting blocs to exercise voice and exert influence. For example, we find that higher passive ownership is associated with less support for management proposals and a greater support for shareholder-initiated governance proposals. Engagement by passive investors also appears to reduce the need for activism by other, non-passive investors; we find that companies with greater passive ownership exhibit improvements in long-term performance and are less likely to be targeted for activism by a hedge fund. Our findings, however, do not resolve the ongoing debate regarding the value implications of various governance structures, including board independence, takeover defenses, and equal voting rights for shareholders, and whether the optimal governance structure may vary across firms in ways that do not always conform to the proxy-voting guidelines of the largest passive institutions. The findings also do not address whether passive investors attempt to determine the individual governance needs of each company in their large portfolios or instead follow a “check the box” approach to governance. While some large passive investors do vary their voting strategies across firms in ways that are not consistent with such a one-size-fits-all approach to governance (e.g., see Davis and Kim, (2007)), additional analysis regarding these questions would seem to be a promising direction for further research. Passive Investors, Not Passive Owners -- Page 26

28) References Admati, Anat R., and Paul Pfleiderer, 2009, The “Wall Street Walk” and shareholder activism: Exit as a form of voice, Review of Financial Studies 22, 2645–2685. Aggarwal, Reena, Isil Erel, Miguel Ferreira, and Pedro Matos, 2011, Does governance travel around the world? Evidence from institutional investors, Journal of Financial Economics 100, 154–181. Aghion, Philippe, John Van Reenen, and Luigi Zingales, 2013, Innovation and Institutional Ownership, American Economic Review 103, 277–304. Angrist, Joshua D., and Jörn-Steffen Pischke, 2009, Mostly harmless econometrics: An empiricist’s companion, Princeton University Press, Princeton. Back, Kerry, Tao Li, and Alexander Ljungqvist, 2014, Liquidity and governance, NBER working paper 19669. Bakke, Tor-Erik, and Toni M. Whited, 2012, Threshold events and identification: A study of cash shortfalls, Journal of Finance 67, 1083–1111. Bebchuk, Lucian, Alma Cohen, and Allen Ferrell, 2009, What matters in corporate governance? Review of Financial Studies 22, 783–827. Bhagat, Sanjai, and Bernard S. Black, 2002, The Non-Correlation Between Board Independence and Long-Term Firm Performance, Journal of Corporation Law 27, 231–274. Black, Bernard S., 1992, Agents watching agents: The promise of institutional investor voice, UCLA Law Review 39, 811–893. Black, Bernard S., 1998, Shareholder activism and corporate governance in the United States, in Peter Newman, ed.: The New Palgrave Dictionary of Economics and the Law (Palgrave Macmillan: New York, NY). Boone, Audra L., and Joshua T. White, 2014, The effect of institutional ownership on firm transparency and information production, forthcoming in Journal of Financial Economics. Bradley, Michael, Alon Brav, Itay Goldstein, and Wei Jiang, 2010, Activist arbitrage: A study of open-ending attempts of closed-end funds, Journal of Financial Economics 95, 1–19. Brav, Alon, Wei Jiang, and Hyunseob Kim, 2010, Hedge fund activism: A review, Foundations and Trends in Finance 4, 1–66. Brav, Alon, Wei Jiang, Frank Partnoy, and Randall Thomas, 2008, Hedge fund activism, corporate governance, and firm performance, Journal of Finance 63, 1729–1775. Bushee, Brian J., 1998, The influence of institutional investors on myopic R&D investment behavior, Accounting Review 73, 305–333. Bushee, Brian J., 2001, Do institutional investors prefer near-term earnings over long-run value? Contemporary Accounting Research 18, 207–246. Busse, Jeffrey A., and Qing Tong, 2012, Mutual fund industry selection and persistence, Review of Asset Pricing Studies 2, 245–274. Carleton, William T., James M. Nelson, and Michael S. Weisbach, 1998, The influence of institutions on corporate governance through private negotiations: Evidence from TIAACREF, Journal,of Finance 53, 1335–1362. Chang, Yen-Cheng, Harrison Hong, and Inessa Liskovich, 2014, Regression discontinuity and the price effects of stock market indexing, Review of Financial Studies, forthcoming. Chhaochharia, Vidhi, and Yaniv Grinstein, 2009, CEO compensation and board structure, The Journal of Finance 64, 231–261. Passive Investors, Not Passive Owners -- Page 27

29) Choi, Stephen, Jill Fisch, and Marcel Kahan, 2013, Who calls the shots? How mutual funds vote on director elections, Harvard Business Law Review 3, 35–81. Chung, Kee H., and Hao Zhang, 2011, Corporate governance and institutional ownership, Journal of Financial and Quantitative Analysis 46, 247–273. Coates IV, John C., 2000, Takeover defenses in the shadow of the pill: A critique of the scientific evidence, Texas Law Review 79, 271–382. Coles, Jeffrey L., Naveen D. Daniel, and Lalitha Naveen, 2008, Boards: Does one size fit all? Journal of Financial Economics 87, 329–356. Crane, Alan D., Sébastien Michenaud, and James P. Weston, 2014, The effect of institutional ownership on payout policy: Evidence from index thresholds, working paper. Cremers, K.J. Martijn, and Vinay B. Nair, 2005, Governance mechanisms and equity prices, Journal of Finance 60, 2859–894. Daines, Robert, and Michael Klausner, 2001, Do IPO charters maximize firm value? Antitakeover protection in IPOs, The Journal of Law, Economics, & Organization 17, 83–120. Davis, Gerald F., and E. Han Kim, 2007, Business ties and proxy voting by mutual funds, Journal of Financial Economics 85, 552–570. Del Guercio, Diane, and Jennifer Hawkins, 1999, The motivation and impact of pension fund activism, Journal of Financial Economics 52, 293–340. Duchin, Ran, John G. Matsusaka, and Oguzhan Ozbas, 2010, When are outside directors effective? Journal of Financial Economics 96, 195–214. Easterbrook, Frank H., and Daniel R. Fischel, 1983, Voting in corporate law, Journal of Law and Economics 26, 395–427. Edmans, Alex, 2009, Blockholder trading, market efficiency, and managerial myopia, Journal of Finance 64, 2481–2513. Edmans, Alex, 2014, Blockholders and corporate governance, prepared for Annual Review of Financial Economics. Edmans, Alex, Vivian W. Fang, and Emmanuel Zur, 2013, The effect of liquidity on governance, Review of Financial Studies 26, 1443–1482. Edmans, Alex, and Gustavo Manso, 2011, Governance through trading and intervention: A theory of multiple blockholders, Review of Financial Studies 24, 2395–2428. Fama, Eugene F., and Michael C. Jensen, 1983, Separation of Ownership and Control, Journal of Law and Economics 26, 301–325. Fich, Eliezer M., Jarrad Harford, and Anh L. Tran, 2015, Motivated monitors: the importance of institutional investors’ portfolio weights, Journal of Financial Economics, forthcoming. Gillan, Stuart L., and Laura T. Starks, 2000, Corporate governance proposals and shareholder activism: the role of institutional investors, Journal of Financial Economics 57, 275–305. Gompers, Paul A., and Joy Ishii, and Andrew Metrick, 2010, Extreme governance: An analysis of dual-class firms in the United States, Review of Financial Studies 23, 1051–1088. Gormley, Todd A. and David A. Matsa, 2014, Common errors: How to (and not to) control for unobserved heterogeneity, Review of Financial Studies 27(2), 617–661. Gormley, Todd A., David A. Matsa, and Todd Milbourn, 2013, CEO compensation and corporate risk: Evidence from a natural experiment, Journal of Account and Economics 56, 79–101. Grinstein, Yaniv, and Roni Michaely, 2005, Institutional holdings and payout policy, Journal of Finance 60, 1389–1426. Passive Investors, Not Passive Owners -- Page 28

30) Guo, Re-Jin, Timothy A. Kruse, and Tom Nohel, 2008, Undoing the powerful anti-takeover force of staggered boards, Journal of Corporate Finance 14, 274 – 288. Harris, Milton, and Arthur Raviv, 2010, Control of corporate decisions: Shareholders vs. management, Review of Financial Studies 23, 4115–4157. Hartzell, Jay C., and Laura T. Starks, 2003, Institutional investors and executive compensation, Journal of Finance 58, 2351–2374. Hermalin, Benjamin E., and Michael S. Weisbach, 1998, Endogenously chosen boards of directors and their monitoring of the CEO, American Economic Review 88, 96–118. Hermalin, Benjamin E., and Michael S. Weisbach, 2003, Boards of directors as an endogenously determined institution: A survey of the economic literature, Economic Policy Review 9, 7–26. Iliev, Peter, and Michelle Lowry, 2015, Are mutual funds active voters? The Review of Financial Studies 28, 446-485. Investment Technology Group, 2008, 2008 Russell Reconstitution Recap. Jensen, Michael C., 1986, Agency costs of free cash flow, corporate finance and takeovers, American Economic Review 76, 323–329. Klausner, Michael, 2012, Fact and fiction in corporate law and governance, Stanford Law Review 65, 1325–1370. Klein, April and Emmanuel Zur, 2009, Entrepreneurial shareholder activism: Hedge Funds and other private investors, Journal of Finance 64, 187–229. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert W. Vishny, 2000, Agency problems and dividend policies around the world, Journal of Finance 55, 1–33. Levit, Doron, 2013, Soft shareholder activism, working paper. Levit, Doron, and Nadya Malenko, 2011, Nonbinding voting and shareholder proposals, Journal of Finance 66, 1579–1614. Lu, Ruichang, 2013, How does institutional ownership affect bank loan pricing: Evidence from a regression discontinuity design, working paper. Maug, Ernst, 1998, Large shareholders as monitors: Is there a trade-off between liquidity and control? Journal of Finance 53, 65–98. McKinnish, T. 2008, Panel data models and transitory fluctuations in the explanatory variable, In Modeling and evaluating treatment effect in econometrics, eds. Daniel L. Millimet, Jeffrey A. Smith, and Edward J. Vytlacil, 334–58. Amsterdam: Elsevier. McNabb, F. William III, 2014, Getting to know you: Sharing practical governance viewpoints, University of Delaware, John Weinberg Center for Corporate Governance, speech on October 30, 2014. Michaely, Roni, Jillian Popadak, and Christopher Vincent, 2014, The deleveraging of U.S. firms and institutional investors’ role, working paper. Mullins, William, 2014, The governance impact of index funds: Evidence from a regression discontinuity, working paper. Romano, Roberta, 1993, Public pension fund activism in corporate governance reconsidered, Columbia Law Review 93, 795–853. Russell Investments, 2013, Russell U.S. Equity Indexes Construction and Methodology. Schmidt, Cornelius, 2012, Shareholding monitoring incentives and corporate policies, working paper. Shliefer, Andrei, and Robert W. Vishny, 1986, Large shareholders and corporate control, Journal of Political Economy 94, 461–488. Passive Investors, Not Passive Owners -- Page 29

31) Sommer, Jeff, 2013, Challenging Management (but Not the Market), The New York Times, March 16. Stein, Jeremy C., 1988, Takeover Threats and Managerial Myopia, Journal of Political Economy 96, 61-80. Stock, James H., Jonathan H. Wright, and Motohiro Yogo, 2002, A survey of weak instruments and weak identification in generalized method of moments, Journal of Business & Economics Statistics 20, 518–529. The Economist, 2015, “Capitalism’s unlikely heroes: Why activist investors are good for the public company,” February 7th. Useem, Michael, Edward H. Bowman, Jennifer Myatt, and Craig W. Irvine, 1993, US institutional investors look at corporate governance in the 1990s, European Management Journal 11, 175–89. Vanguard, 2014, Our views on corporate governance, https://about.vanguard.com/vanguardproxy-voting/corporate-governance/. Weisbach, Michael S., 1988, Outside directors and CEO turnover, Journal of Financial Economics 20, 431–460. Passive Investors, Not Passive Owners -- Page 30

32) 35%   30%   %  of  equity  mutual  fund  assets  that  are  passively  managed   25%   20%   15%   10%   2014   2013   2012   2011   2010   2009   2008   2007   2006   2005   2004   2003   2002   2001   2000   1999   1998   5%   10.0%   8.0%   %  of  total  market  cap  held  by  passively  managed  funds   6.0%   4.0%   2.0%   2014   2013   2012   2011   2010   2009   2008   2007   2006   2005   2004   2003   2002   2001   2000   1999   1998   0.0%   Figure 1 Growth of passive investors, 1998-2014 This figure plots the estimated percent of all U.S. equity mutual fund assets under management between 1998 and 2014 that are held in passively managed funds and the estimated percent of total U.S. market capitalization held by passively managed mutual funds. We construct the figure by matching the S12 mutual fund holdings data compiled in the Thomson Reuters Mutual Fund Holdings Database to market caps reported in CRSP and fund names in the CRSP Mutual Fund data. We use a name-parsing procedure along with the index fund identifier from the CRSP Mutual Fund File to classify mutual funds as passively managed. Our procedure is described in the text. Holdings and market cap are calculated each year at the end of the third quarter. Passive Investors, Not Passive Owners -- Page 31

33) Por$olio  weight  in  index  %   0.2   0.2   0.18   0.18   0.16   0.16   0.14   0.14   0.12   0.12   0.1   0.1   0.08   0.08   Bo2om  500  ï¬rms  within   Russell  1000   0.06   0.04   0.06   0.04   0.02   0   500   Top  500  ï¬rms  within   Russell  2000   0.02   0   600   700   800   Ranking  within  index   900   1000   0   100   200   300   400   500   Ranking  within  index   Figure 2 Portfolio weights in the Russell 1000 and 2000 indices by within-index ranking for the year 2006 This figure plots the portfolio weights of the bottom 500 firms in the Russell 1000 index and the top 500 firms in the Russell 2000 index for the end-of-June 2006. Observations are ordered by their within-index ranking such that rankings of 1 and 1000 represent the firms with the largest and 1000th largest portfolio weight in the index, respectively. The portfolio weights are given as a percent. Passive Investors, Not Passive Owners -- Page 32

34) Ac0ve  ownership  %   Ln(Market  capitaliza0on)   24   22   21   19   1   0.5   0   Passive  ownership  %   0   100   200   300   400   500   600   700   800   900   1000   21   15   9   100   200   300   400   500   600   700   800   900   1000   0   100   200   300   400   500   600   700   800   900   1000   0   Unclassified  ownership  %   Frac0on  in  Russell  2000   0   27   100   200   300   400   500   600   700   800   900   1000   6   4   2   0   Figure 3 Market capitalization, index assignment, and ownership by market capitalization rankings for the bottom 500 firms of Russell 1000 and top 500 firms of Russell 2000 4   3   2   1   0   100   200   300   400   500   600   700   800   900   1000   Ranking  based  on  end-­â€of-­â€May  market  capitaliza0on   This figure plots the average end-of-May Ln(market capitalization), fraction of firm-year observations in the Russell 2000, and passive, active, and unclassified mutual fund ownership (%) by ranking, where ranking is determined using end-of-May market capitalization, as reported in CRSP. The sample includes the bottom 500 firms of the Russell 1000 and the top 500 firms of the Russell 2000, as determined using end-of-June Russell-assigned portfolio weights for each index. Mutual fund ownership is calculated as of September each year, and all averages are calculated using bins of 10 firms and data from 1998-2006. For the ownership panels, we scale the vertical axis to report a standard deviation on each side of the sample mean. Passive Investors, Not Passive Owners -- Page 33

35) Table 1 Summary statistics This table reports summary statistics of our key variables for our main sample: firms in the 250 bandwidth around the cutoff between the Russell 1000 and 2000 indexes from 1998-2006. Definitions for all variables are provided in Appendix Table 1. Accounting variables are winsorized at the 1% level, and we delete observations where either mutual fund ownership is missing or total mutual fund holdings exceed a stock's market capitalization. Obs. Total mutual fund ownership % Passive ownership % Active ownership % Unclassified ownership % Independent director % Poison pill removal Greater ability to call special meeting Indicator for dual class shares Mngt. proposal support % Shareholder gov. proposal support % Indicator for hedge fund activism ROA Mean Median 4,415 4,415 4,415 4,415 2,871 2,957 1,858 1,858 1,288 202 4,415 4,291 25.2 3.0 18.9 3.2 65.1 0.04 0.006 0.13 84.7 36.3 0.016 0.03 25.0 2.6 18.1 2.5 66.7 0 0 0 87.2 31.5 0 0.04 SD 12.9 2.3 10.9 2.9 18.1 0.19 0.08 0.33 11.9 22.8 0.12 0.11 Passive Investors, Not Passive Owners -- Page 34

36) Table 2 Impact of index assignment on mutual fund ownership This table reports estimates of a regression of mutual fund holdings on an indicator for membership in the Russell 2000 index plus additional controls. Specifically, we estimate N Ownership% it = η + λ R2000it + ∑ χ n ( Ln( Mktcapit )) + σ Ln( Float )it + δ t + u it n n=1 where R2000it is a dummy variable equal to 1 if stock i is in the Russell 2000 Index at end of June in year t, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, N is the polynomial order we use to control for Ln(Mktcapit), Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. Ownership%it measures mutual fund ownership (in percent) for stock i at the end of September in year t. In this table we use four different definitions for Ownership% for stock i: (1) the percentage of shares outstanding owned by all mutual funds (from S12 filings); (2) the percentage of shares outstanding owned by "passive" funds; (3) the percentage of shares outstanding owned by “active” mutual funds; and (4) the percentage of shares outstanding owned by “unclassified” mutual funds. The mutual fund classifications are defined in the text. The sample consists of the top 250 firms in the Russell 2000 index and bottom 250 firms of the Russell 1000 index (i.e., bandwidth = 250) for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a polynomial order control for Ln(Mktcap) of N = 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols * and *** indicate significance at the 10% and 1% levels, respectively. Percent of firm's common shares held by: Dependent variable = Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations R-squared Passive Active Unclassified (1) R2000 All mutual funds (2) (3) (4) 1.216* (0.662) 1.086*** (0.067) 0.118 (0.604) 0.012 (0.135) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,654 4,415 0.21 1,654 4,415 0.62 1,654 4,415 0.12 1,654 4,415 0.09 Passive Investors, Not Passive Owners -- Page 35

37) Table 3 First stage estimation for ownership by passively managed funds This table reports estimates of our first-stage regression of passive ownership onto an indicator for membership in the Russell 2000 index plus additional controls. Specifically, we estimate N Passive% it = η + λ R2000it + ∑ χ n ( Ln( Mktcapit )) + σ Ln( Float )it + δ t + u it n n=1 where R2000it is a dummy variable equal to 1 if stock i is in the Russell 2000 index at end of June in year t, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. Passive%it is the percentage of shares outstanding owned by passively managed mutual funds, as defined in the text, for stock i at the end of September in year t scaled by its sample standard deviation. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Passive % scaled by its sample standard deviation Dependent variable = (1) R2000 Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations R-squared (2) (3) 0.505*** (0.028) 0.512*** (0.028) 0.473*** (0.029) 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,654 4,415 0.61 1,654 4,415 0.62 1,654 4,415 0.62 Passive Investors, Not Passive Owners -- Page 36

38) Table 4 Ownership by passive investors and board independence This table reports estimates of our instrumental variable estimation used to identify the effect of ownership by passive investors on board independence. Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is the percentage of independent directors on the board of firm i in year t (from Riskmetrics) scaled by its sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Dependent variable = Independent director % (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) 0.729*** (0.160) 0.762*** (0.162) 0.654*** (0.159) 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,082 2,871 1,082 2,871 1,082 2,871 Passive Investors, Not Passive Owners -- Page 37

39) Table 5 Passive ownership and board independence, pre- versus post-2002 rule change This table reports estimates of the second-stage regression of our instrumental variable estimation used to identify the effect of passive investors on the percentage of independent board directors both before and after the 2002 change in exchange-listing requirements regarding board independence. The estimation is the same as in Table 4, except we now separately estimate the model over the 1998-2002 and 2003-2006 time periods using a bandwidth of 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Both the dependent variable and Passive% are scaled by their sample standard deviations. Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Independent director % Sample years = 1998-2002 (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) 1.314*** 1.461*** 1.257*** (0.298) (0.303) (0.297) Sample years = 2003-2006 (4) (5) 0.354*** 0.324** (0.136) (0.137) (6) 0.264* (0.160) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes 882 1,682 882 1,682 882 1,682 549 1,189 549 1,189 549 1,189 Passive Investors, Not Passive Owners -- Page 38

40) Table 6 Ownership by passive investors and takeover defenses This table reports estimates of our instrumental variable estimation used to identify the effect of institutional ownership by passive investors on takeover defense outcomes. Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is the governance variable for firm i in year t scaled by its sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed mutual funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. The governance variables investigated in this table, from Shark Repellent (Factset) and Riskmetrics, are: an indicator for either the withdrawal or expiration (without renewal) of a poison pill in year t, and an indicator for there being fewer restrictions on shareholders' ability to call a special meeting in year t. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 19982006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and first, second, and third polynomial order controls for Ln(Mktcap). Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% levels. Dependent variable = (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations Greater ability to call special meeting Poison pill removal (2) (3) 0.176*** 0.181*** 0.203*** (0.0647) (0.0650) (0.0741) (4) (5) (6) 0.304*** 0.310*** 0.341*** (0.0999) (0.108) (0.114) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,164 2,957 1,164 2,957 1,164 2,957 1,050 1,858 1,050 1,858 1,050 1,858 Passive Investors, Not Passive Owners -- Page 39

41) Table 7 Ownership by passive investors and dual class share structures This table reports estimates of our instrumental variable estimation used to identify the effect of passive investors on the likelihood of dual class shares. Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is an indicator equal to 1 if firm i has dual class shares in year t according to Riskmetrics scaled by its sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed mutual funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Dependent variable = Indicator for dual class shares (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) -0.886*** (0.179) -1.031*** (0.167) -1.005*** (0.181) 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,050 1,858 1,050 1,858 1,050 1,858 Passive Investors, Not Passive Owners -- Page 40

42) Table 8 Ownership by passive investors and shareholder support for proposals This table reports estimates of our instrumental variable estimation to identify the effect of passive investors on shareholder support for management proposals and shareholder-initiated governance proposals. Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is either the average percentage of shareholders that vote along with management proposals at annual meetings for i in year t (from Riskmetrics) or the average percentage of shareholders that vote in support of a shareholder-initiated governance proposal for firm i in year t (from Riskmetrics) each scaled by their sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed mutual funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Management proposal support % (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) -0.783***-0.745***-0.734*** (0.180) (0.179) (0.231) Governance proposal support % (4) (5) (6) 0.492** (0.247) 0.649* (0.348) 0.622* (0.336) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes 775 1,288 775 1,288 775 1,288 127 202 127 202 127 202 Passive Investors, Not Passive Owners -- Page 41

43) Table 9 Ownership by passive investors and hedge fund activism This table reports estimates of our instrumental variable estimation used to identify the effect of ownership by passive investors on the likelihood of hedge fund activism. Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is an indicator equal to 1 if firm i experiences a hedge fund activism event in year t, as defined in Brav, Jiang, Partnoy, and Thomas (2008) and Brav, Jiang, and Kim (2010), scaled by its sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed mutual funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold, and polynomial order controls for Ln(Mktcap) of N = 1, 2, and 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols * and ** indicate significance at the 10% and 5% levels, respectively. Dependent variable = Indicator for hedge fund activism event (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) -0.131* (0.0721) -0.130* (0.0718) -0.162** (0.0805) 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,654 4,415 1,654 4,415 1,654 4,415 Passive Investors, Not Passive Owners -- Page 42

44) Table 10 Ownership by passive investors and firms' return on assets This table reports estimates of our instrumental variable estimation used to identify the effect of ownership by passive institutional investors on firms' performance, as measured using firms' return on assets (ROA). Specifically, we estimate N Yit = α + β Passive% it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where Yit is the ROA for firm i in year t scaled by its sample standard deviation, Passive%it is the percentage of shares outstanding owned by passively managed mutual funds (as defined in the text) for stock i at the end of September in year t scaled by its sample standard deviation, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. We instrument Passive% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The specification in columns (1)-(3) are the same as in earlier tables, but in columns (4)-(6), we add two additional controls to the specification: an indicator that equals one for firms that are in the Russell 2000 index in year t but were in the Russell 1000 in year t-1, and an indicator that equals one for firms that are in the Russell 1000 index in year t but were in the Russell 2000 index in year t-1. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Mutual Fund Holdings Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and first, second, and third polynomial order controls for Ln(Mktcap). Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Dependent variable = ROA (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects Controls for movers # of firms Observations (2) (3) -0.028 -0.015 0.035 (0.098) (0.093) (0.105) (4) (5) (6) 0.304*** 0.310*** 0.414*** (0.111) (0.106) (0.121) 250 1 yes yes no 250 2 yes yes no 250 3 yes yes no 250 1 yes yes yes 250 2 yes yes yes 250 3 yes yes yes 1,600 4,291 1,600 4,291 1,600 4,291 1,600 4,291 1,600 4,291 1,600 4,291 Passive Investors, Not Passive Owners -- Page 43

45) Table 11 Impact of index assignment on institution-level (13F) stock ownership This table reports estimates of our first-stage regression of institutional holdings on an indicator for membership in the Russell 2000 index plus additional controls. Specifically, we estimate N Ownership% it = η + λ R2000it + ∑ χ n ( Ln( Mktcapit )) + σ Ln( Float )it + δ t + u it n n=1 where R2000it is a dummy variable equal to 1 if stock i is in the Russell 2000 index at end of June in year t, Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t, N is the polynomial order we use to control for Ln(Mktcapit), Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. Ownership%it measures institution-level (13F) ownership (in percent) for stock i at the end of September in year t. In this table we use four different definitions for Ownership% for stock i: (1) the percentage of shares outstanding owned by all institutional investors; (2) the percentage of shares outstanding owned by "quasi-index" institutions, as classified by Bushee (2001); (3) the percentage of shares outstanding owned by "dedicated" institutions as classified by Bushee; and (4) the percentage of shares outstanding owned by “transient” institutions as classified by Bushee. The Bushee classifications are defined in the text. The sample consists of the top 250 firms in the Russell 2000 index and bottom 250 firms of the Russell 1000 index (i.e., bandwidth = 250) for which we obtain holdings data from Thomson Reuters Institutional Holdings (13F) Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a polynomial order control for Ln(Mktcap) of N = 3. Standard errors, ε, are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Percent of firm's common shares held by: Dependent variable = Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations R-squared Quasiindex Dedicated Transient (1) R2000 All Institutions (2) (3) (4) 1.354 (1.517) 2.381*** (0.748) -0.539 (0.891) -0.445 (0.845) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,633 4,357 0.24 1,633 4,357 0.26 1,633 4,357 0.02 1,633 4,357 0.09 Passive Investors, Not Passive Owners -- Page 44

46) Table 12 Robustness of IV estimates to using passive indicator based on institution-level (13F) stock ownership This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables where passive ownership is measured using the percentage of stock held by "quasi-index" institutions, as classified by Bushee (2001) and defined in the text. The estimation and outcomes are the same as in Tables 4-10, except Passive% is replaced by Quasi-index%, the share of market cap held by quasi-index institutions scaled by its sample standard deviation. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Quasi-index % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations Poison pill removal (1) Dep. variable = Ind. directors % Ability to Mngt. Ind. for dual call special proposal class shares meeting support % Gov. proposal support % HF activism event ROA (2) (3) (4) (5) (6) (7) (8) 1.197*** (0.388) 0.885* (0.479) 0.958** (0.473) -2.866** (1.170) -1.148** (0.516) 1.297* (0.680) -0.580* (0.336) 1.803** (0.899) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,073 2,840 1,160 2,940 1,047 1,847 1,047 1,847 768 1,279 125 200 1,633 4,357 1,586 4,246 Passive Investors, Not Passive Owners -- Page 45

47) Appendix for “Passive Investors, Not Passive Owners”

48) IX. Appendix – Excerpts from Fund Governance/Voting Policies In this appendix, we provide excerpts from the historical voting policies of various institutional investors that offer index-related investment products. These proxy-voting policies were all obtained from fund prospectuses issued in late 2003 and early 2004 shortly after the SEC first required the filing of such voting policies on July 1, 2003. Four common themes of these governance/voting policies are (1) to withhold support or vote against boards that are not sufficiently independent, (2) broadly oppose takeover defenses, like poison pills, restrictions on shareholders’ ability to call a special meeting, and classified boards, (3) oppose unequal voting rights (i.e., dual class share structures), and (4) push for executive compensation that is tied to performance but not excessive. A. iShares: Proxy Voting Guidelines31 “The Company has adopted as its proxy voting policies the proxy voting guidelines of BGFA [Barclays Global Fund Advisors], the investment advisor to each Fund. The Company has delegated to BGFA the responsibility for voting proxies on the portfolio securities held by each Fund. Therefore, the remainder of this section discusses BGFA’s proxy voting guidelines… When voting proxies, BGFA attempts to ensure that companies follow practices that advance their economic value and allow the market to place a proper value on their assets. With respect to certain specific issues: • • • BGFA generally supports management in the election of directors and generally supports proposals that strengthen the independence of boards of directors; BGFA generally does not support proposals on social issues that lack a demonstrable economic benefit to the issuer and the Fund investing in such issuer; and BGFA generally votes against anti-takeover proposals and proposals which would create additional barriers or costs to corporate transactions.” B. State Street Global Advisors: Proxy Voting Policies and Procedures32 “For most issues and in most circumstances, we abide by the following general guidelines… FM votes in support of management on the following ballot items… • • • • • • • Elimination of cumulative voting… Capitalization changes which eliminate other classes of stock and voting rights… Elimination of pre-emptive rights for share issuance of less than a given percentage (country specific - ranging from 5% to 20%) of the outstanding shares Elimination of "poison pill" rights… Stock option plans which are incentive based and not excessive Other stock-based plans which are appropriately structured Reductions in super-majority vote requirements…                                                                                                                         31 32 http://www.sec.gov/Archives/edgar/data/930667/000119312503100400/d485bpos.txt http://www.sec.gov/Archives/edgar/data/826686/000104746904004745/a2128691z485bpos.txt Appendix for Passive Investors, Not Passive Owners -- Page 1

49) FM votes against management on the following items… • • • • • • • • Capitalization changes that add "blank check" classes of stock or classes that dilute the voting interests of existing shareholders… Anti-takeover and related provisions that serve to prevent the majority of shareholders from exercising their rights or effectively deter appropriate tender offers and other offers Amendments to bylaws which would require super-majority shareholder votes to pass or repeal certain provisions Elimination of Shareholders' Right to Call Special Meetings Establishment of classified boards of directors… Shareholder rights plans that allow the board of directors to block appropriate offers to shareholders or which trigger provisions preventing legitimate offers from proceeding Excessive compensation… Proposals requesting re-election of insiders or affiliated directors who serve on audit, compensation, and nominating committees… FM votes in support of shareholders on the following ballot items… • • • • • Establishment of an annual election of the board of directors Mandates requiring a majority of independent directors on the Board of Directors and the audit, nominating, and compensation committees… Mandates that shareholder-rights plans be put to a vote or repealed… Repeals of various anti-takeover related provisions Reduction or elimination of super-majority vote requirements…” C. Vanguard: Proxy Voting Guidelines33 “The Board of Trustees (the Board) of each Vanguard fund that invests in stocks has adopted proxy voting procedures and guidelines to govern proxy voting by the fund. The Board has delegated day-to-day oversight of proxy voting to the Proxy Oversight Committee (the Committee), comprised of senior Vanguard officers and subject to the operating procedures and guidelines described below… I. THE BOARD OF DIRECTORS A. ELECTION OF DIRECTORS We believe that good governance starts with a majority-independent board, whose key committees are comprised entirely of independent directors. As such, companies should attest to the independence of directors who serve on the Compensation, Nominating and Audit committees…                                                                                                                         33 http://www.sec.gov/Archives/edgar/data/105563/000093247104000415/wellington485b032004.txt   Appendix for Passive Investors, Not Passive Owners -- Page 2

50) We will generally support proposals to declassify existing boards (whether proposed by management or shareholders), and will block efforts by companies to adopt classified board structures, in which only part of the board is elected each year. II. APPROVAL OF INDEPENDENT AUDITORS We believe that the relationship between the company and its auditors should be limited primarily to the audit, although it may include certain closely related activities that do not, in the aggregate, raise any appearance of impaired independence. We will generally support management's recommendation for the ratification of the auditor, except in instances where audit and audit-related fees make up less than 50% of the total fees paid by the company to the audit firm. We will evaluate on a case-by-case basis instances in which the audit firm has a substantial non-audit relationship with the company (regardless of its size relative to the audit fee) to determine whether we believe independence has been compromised. III. COMPENSATION ISSUES A. STOCK-BASED COMPENSATION PLANS We believe that appropriately designed stock-based compensation plans, administered by an independent committee of the board and approved by shareholders, can be an effective way to align the interests of long-term shareholders and the interests of management, employees, and directors. Conversely, we oppose plans that substantially dilute our ownership interest in the company, provide participants with excessive awards, or have inherently objectionable structural features… IV. CORPORATE STRUCTURE AND SHAREHOLDER RIGHTS We believe the exercise of shareholder rights, in proportion to economic ownership, to be a fundamental privilege of stock ownership that should not be unnecessarily limited. Such limits may be placed on shareholders' ability to act by corporate charter or by-law provisions, or by the adoption of certain takeover provisions. We believe that, in general, the market for corporate control should be allowed to function without undue interference from these artificial barriers. Our positions on a number of the most commonly presented issues in this area are as follows: A. SHAREHOLDER RIGHTS PLANS (POISON PILLS) A company's adoption of a so-called poison pill effectively limits a potential acquirer's ability to buy a controlling interest without the approval of the target's board of directors. Such a plan, in conjunction with other takeover defenses, may serve to entrench incumbent management and directors… B. CUMULATIVE VOTING Appendix for Passive Investors, Not Passive Owners -- Page 3

51) We are generally opposed to cumulative voting under the premise that it allows shareholders a voice in director elections that is disproportionate to their economic investment in the corporation. C. SUPERMAJORITY VOTE REQUIREMENTS We support shareholders' ability to approve or reject matters presented for a vote based on a simple majority. Accordingly, we will support proposals to remove supermajority requirements and oppose proposals to impose them. D. RIGHT TO CALL MEETINGS AND ACT BY WRITTEN CONSENT We support shareholders' right to call special meetings of the board (for good cause and with ample representation) and to act by written consent. We will generally vote for proposals to grant these rights to shareholders and against proposals to abridge them. E. CONFIDENTIAL VOTING We believe that the integrity of the voting process is enhanced substantially when shareholders (both institutions and individuals) can vote without fear of coercion or retribution based on their votes. As such, we support proposals to provide confidential voting. F. DUAL CLASSES OF STOCK We are opposed to dual class capitalization structures that provide disparate voting rights to different groups of shareholders with similar economic investments. As such, we will oppose the creation of separate classes with different voting rights and will support the dissolution of such classes. V. CORPORATE AND SOCIAL POLICY ISSUES Proposals in this category, initiated primarily by shareholders, typically request that the company disclose or amend certain business practices. We generally believe that these are "ordinary business matters" that are primarily the responsibility of management and should be evaluated and approved solely by the corporation's board of directors. Often, proposals may address concerns with which we philosophically agree, but absent a compelling economic impact on shareholder value (e.g., proposals to require expensing of stock options), we will typically abstain from voting on these proposals. This reflects our belief that regardless of our philosophical perspective on the issue, these decisions should be the province of company management unless they have a significant, tangible impact on the value of our investment and, we don't view management as responsive to the matter.” Appendix for Passive Investors, Not Passive Owners -- Page 4

52) 0.7   First  Stage   1.6   0.6   1.2   0.5   0.8   0.4   0.4   0.3   Independent  Directors  %   0   1   Greater  Ability  to  Call  Spec.  Meet.   0   0.6   Poison  Pill  Removal   BW 0.4   0.2   0   Dual  Class  Shares  Indicator   0   0.75   -­â€0.5   -­â€0.5   0.5   -­â€1   -­â€1   0.25   -­â€1.5   -­â€1.5   0   -­â€2   Management  Prop.  Support  %   -­â€2   Governance  Proposal  Support  %   Hedge  Fund  AcOvism  Indicator   2   0.2   1   1.5   0   0.75   -­â€0.2   0.5   -­â€0.4   ROA   0.25   1   0.5   0   -­â€0.5   500   450   400   350   300   250   200   150   100   -­â€0.6   500   450   400   350   300   250   200   150   100   0   500   450   400   350   300   250   200   150   100   Appendix Figure 1 First stage and IV point estimates in the 100 through 500 bandwidths around Russell 1000/2000 threshold This figure plots the point estimate and 95th percentile confidence intervals by bandwidth choice for the outcomes reported in Tables 3-10. The first stage and IV estimations are the same as in Tables 3-10 except the bandwidth is varied between 100 and 500 firms around the Russell 1000/2000 threshold. A third-order polynomial control for Ln(Mktcap) is included in all estimations. Appendix for Passive Investors, Not Passive Owners -- Page 5 500 499 498 497 496 495 494 493 492 491 490 489 488 487 486 485 484 483 482 481 480 479 478 477 476 475 474 473 472 471

53) Appendix Table 1 Variable definitions Variable Name Source Definition R2000 Mutual fund ownership % Passive % Active % Unclassified % Independent director % Poison pill removal Greater ability to call spec. meet. Indicator for dual class shares Mngt. proposal support % Shareholder gov. prop. support % Indicator for hedge fund activism ROA Institutional ownership % Quasi-index % Dedicated % Transient % Russell Investments Thomson Reuters S12 files Thomson Reuters S12 files Thomson Reuters S12 files Thomson Reuters S12 files Riskmetrics (Directors) Shark Repellent (FactSet) Riskmetrics (Governance) Riskmetrics (Governance) Riskmetrics (Voting Results) Riskmetrics (Voting Results) Brav, Jiang, and Kim (2010) Compustat Thomson Reuters 13F files Brian Bushee website Brian Bushee website Brian Bushee website Indicator equal to 1 if firm is in the Russell 2000 % of shares outstanding held by mutual funds in September of year t % of shares outstanding held in September of year t by passively managed funds % of shares outstanding held in September of year t by actively managed funds % of shares outstanding held in September of year t by unclassified funds % of board seats held by directors classified as independent by Riskmetrics Indicator equal to 1 if poison pill is withdrawn or allowed to expire at time t Indicator equal to 1 if shareholders better able to call a special meeting at time t Indicator equal to 1 if a firm has dual class shares at time t Percentage of 'Yes" votes for management proposals Percentage of 'Yes" votes for sharehold governance proposals Indicator equal to 1 if a firm has an activism event at time t Net income (ni) / total assets (at) % of shares outstanding held by institutional investors in September of year t % of shares outstanding held by quasi-indexer institutions in September of year t % of shares outstanding held by dedicated insitutions in September of year t % of shares outstanding held by transient insitutions in September of year t Appendix for Passive Investors, Not Passive Owners -- Page 6

54) Appendix Table 2 First stage estimation for ownership by actively managed and unclassified mutual funds This table reports estimates of our first-stage regression of ownership by actively managed and unclassified mutual funds onto an indicator for membership in the Russell 2000 index plus additional controls over the 1998-2006 sample period. The specification is the same as in Table 3, except that the dependent variable in columns (1)-(3) is now Active%it, which is the percentage of shares outstanding owned by actively managed mutual funds for stock i at the end of September in year t scaled by its sample standard deviation, and the dependent variable in columns (4)-(6) is now Unclassified%it, which is the percentage of shares outstanding owned by unclassified mutual funds for stock i at the end of September in year t scaled by its sample standard deviation. Both Active% and Unclassified% are defined in the text. Standard errors are clustered at the firm level and reported in parentheses. Dependent variable = Active % scaled by its sample standard deviation Unclassified % scaled by its sample standard deviation (1) R2000 Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations R-squared (2) (3) (4) (5) (6) 0.055 (0.055) 0.049 (0.054) 0.011 (0.056) 0.028 (0.047) 0.020 (0.046) 0.004 (0.047) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes 1,654 4,415 0.12 1,654 4,415 0.12 1,654 4,415 0.12 1,654 4,415 0.08 1,654 4,415 0.09 1,654 4,415 0.09 Appendix for Passive Investors, Not Passive Owners -- Page 7

55) Appendix Table 3A First-stage estimation for Tables 4, 6, and 7 This table reports estimates of our first-stage regression of passive ownership onto an indicator for membership in the Russell 2000 index plus additional controls over the 1998-2006 sample period. The specification is the same as in Table 3, but we now restrict our sample to the smaller subsample of observations with non-missing Riskmetrics (Directors) data on board independence, non-missing Shark Repellent (FactSet) data on poison pills, or non-missing Riskmetrics (Governance) data on shareholders' ability to call special meetings and dual class share structures. Specifically, these are the first-stage estimates for the IV estimates reported in Tables 4, 6, and 7. Standard errors are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Dependent variable = Passive % scaled by its sample standard deviation (1) R2000 Bandwidth Polynomial order, N Float control Year fixed effects 1st stage estimate for… # of firms Observations R-squared (2) (3) (1) (2) (3) (4) (5) (6) 0.423*** (0.031) 0.428*** (0.031) 0.412*** (0.030) 0.624*** (0.040) 0.628*** (0.041) 0.596*** (0.046) 0.551*** (0.046) 0.536*** (0.046) 0.510*** (0.043) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes Table 4, Col. (1) Table 4, Col. (2) Table 4, Col. (3) Table 6, Col. (1) Table 6, Col. (2) Table 6, Col. (3) 1,082 2,871 0.74 1,082 2,871 0.74 1,082 2,871 0.74 1,164 2,957 0.50 1,164 2,957 0.50 1,164 2,957 0.50 Table 6, Table 6, Table 6, Col. (4) & Col. (5) & Col. (6) & Table 7, Table 7, Table 7, Col. (1) Col. (2) Col. (3) 1,050 1,858 0.67 1,050 1,858 0.67 1,050 1,858 0.67 Appendix for Passive Investors, Not Passive Owners -- Page 8

56) Appendix Table 3B First-stage estimation for Table 8 This table reports estimates of our first-stage regression of passive ownership onto an indicator for membership in the Russell 2000 index plus additional controls over the 1998-2006 sample period. The specification is the same as in Table 3, but we now restrict our sample to the smaller subsample of observations with non-missing Riskmetrics (Voting Results) data on % support for management proposals and shareholder-intitiated governance proposals. Specifically, these are the first-stage estimates for the IV estimates reported in Table 8. Standard errors are clustered at the firm level and reported in parentheses. *** indicates significance at the 1% level. Dependent variable = Passive % scaled by its sample standard deviation (1) R2000 Bandwidth Polynomial order, N Float control Year fixed effects 1st stage estimate for… # of firms Observations R-squared (2) (3) (4) (5) (6) 0.420*** (0.036) 0.415*** (0.036) 0.373*** (0.041) 0.774*** (0.198) 0.696*** (0.231) 0.754*** (0.232) 250 1 yes yes 250 2 yes yes 250 3 yes yes 250 1 yes yes 250 2 yes yes 250 3 yes yes Table 8, Table 8, Table 8, Table 8, Table 8, Table 8, Column (1) Column (2) Column (3) Column (4) Column (5) Column (6) 775 1,288 0.74 775 1,287 0.74 775 1,287 0.74 127 202 0.59 127 202 0.60 127 202 0.61 Appendix for Passive Investors, Not Passive Owners -- Page 9

57) Appendix Table 4 Robustness of findings to using Russell-provided market capitalization This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we instead measure end-of-May market caps using Russell-provided market caps for the years 2002-2006. The estimation and outcomes are the same as in Tables 4-10, except Mktcapit is the Russell-provided end-of-May market cap of stock i in year t, except when it is missing (i.e., years 1998-2001), in which case, we use the CRSP market value of equity of stock i measured at May 31 in year t. We instrument Passive% using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations Poison pill removal (1) Dep. variable = Ind. directors % Ability to Mngt. Ind. for dual call special proposal class shares meeting support % Gov. proposal support % HF activism event ROA (2) (3) (4) (5) (6) (7) (8) 0.487*** (0.147) 0.194** (0.095) 0.387*** (0.135) -0.482** (0.187) -0.694*** (0.237) 0.287 (0.340) -0.156* (0.0891) 0.540*** (0.133) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,082 2,871 1,164 2,957 1,050 1,858 1,050 1,858 775 1,288 127 202 1,654 4,415 1,600 4,291 Appendix for Passive Investors, Not Passive Owners -- Page 10

58) Appendix Table 5 Robustness of findings to using Compustat market capitalization This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we instead measure end-of-May market caps using Compustat The estimation and outcomes are the same as in Tables 4-10, except that Mktcapit is the the Compustat market value of equity of stock i measured at May 31 in year t. We instrument Passive% using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Ind. directors % Ability to Poison pill call special removal meeting Ind. for dual class shares (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations (2) (3) (4) 0.650*** (0.169) 0.176** (0.073) 0.299*** (0.112) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,024 2,739 1,085 2,791 992 1,764 992 1,764 Mngt. Gov. HF activism proposal proposal event support % support % (5) ROA (6) (7) (8) 0.565* (0.294) -0.134* (0.081) 0.417*** (0.124) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 733 1,225 119 191 1,549 4,171 1,496 4,052 -1.139*** -0.796*** (0.183) (0.243) Appendix for Passive Investors, Not Passive Owners -- Page 11

59) Appendix Table 6 Robustness of findings to including industry fixed effects This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we add 2-digit SIC industry fixed effects. The data, outcome variables, and specification are the same as in Tables 4-10 except that we now also include 2-digit SIC industry fixed effects in the specification. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Ind. directors % Ability to Poison pill call special removal meeting Ind. for dual class shares (1) Passive % Bandwidth Polynomial order, N Float control Year fixed effects 2-digit industry FE # of firms Observations (2) (3) (4) 0.537*** (0.151) 0.182** (0.0795) 0.349*** (0.116) 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 1,080 2,871 1,164 2,957 1,050 1,858 1,050 1,858 Mngt. Gov. HF activism proposal proposal event support % support % (5) ROA (6) (7) (8) 0.478 (0.336) -0.161* (0.0840) 0.453*** (0.122) 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 775 1,288 127 202 1,654 4,415 1,600 4,291 -0.781*** -0.660*** (0.178) (0.230) Appendix for Passive Investors, Not Passive Owners -- Page 12

60) Appendix Table 7 Robustness of findings to including controls for firms that recently switched indexes This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we add controls to account for firms that recently switched indexes. Specifically, the data, outcome variables, and specification are the same as in Tables 4-10 except that we now two additional controls to the specification: an indicator that equals one for firms that are in the Russell 2000 index in year t but were in the Russell 1000 in year t-1, and an indicator that equals one for firms that are in the Russell 1000 index in year t but were in the Russell 2000 index in year t-1. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). Standard errors, ε, are clustered at the firm level. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Ind. directors % Ability to Ind. for Mngt. Gov. Poison pill call special dual class proposal proposal removal meeting shares support % support % Bandwidth Polynomial order, N Float control Year fixed effects Controls for movers # of firms Observations (4) (5) (7) (8) (2) 0.514*** (0.170) 0.201*** (0.077) 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 1,080 2,871 1,164 2,957 1,050 1,858 1,050 1,858 775 1,288 127 202 1,654 4,415 1,600 4,291 0.296*** -1.090*** -0.953*** (0.113) (0.193) (0.250) (6) ROA (1) Passive % (3) HF activism event 0.663** (0.324) -0.287*** 0.414*** (0.097) (0.121) Appendix for Passive Investors, Not Passive Owners -- Page 13

61) Appendix Table 8 Robustness of findings to using only ownership of Barclays Bank, State Street, and Vanguard This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of aggregate institutional ownership by Vanguard, State Street, and Barclays Bank on our governance and corporate outcome variables. Specifically, we estimate N Yit = α + β BSV % it + ∑ θ n ( Ln( Mktcapit )) + γ Ln( Float )it + δ t + ε it n n=1 where: Yit is the outcome variable for firm i in year t; BSV%it is the percentage of shares outstanding owned by Barclays Bank, State Street, and Vanguard of stock i at the end of September in year t; Mktcapit is the CRSP market value of equity of stock i measured at May 31 in year t; and Floatit is the float-adjusted market value of equity (provided by Russell) at June 30 in year t, and δt are year fixed effects. The outcome variables investigated in this table are the same as in earlier tables, and we instrument BSV% in the above estimation using R2000it, an indicator equal to one if firm i is part of the Russell 2000 index in year t. The data consist of firms in the two Russell indexes for which we obtain holdings data from Thomson Reuters Institutional Holdings (13F) Database and which we match with data from the monthly CRSP file. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. BSV % Bandwidth Polynomial order, N Float control Year fixed effects # of firms Observations Ind. directors % Poison pill removal (1) Dependent variable = Ability to Mngt. Ind. for dual call special proposal class shares meeting support % Gov. proposal support % HF activism event ROA (2) (3) (4) (5) (6) (7) (8) 0.673*** (0.159) 0.249*** (0.0920) 0.432*** (0.152) -1.293*** (0.262) -0.792*** (0.265) 0.838* (0.496) -0.204** (0.0967) 0.596*** (0.184) 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 250 3 yes yes 1,073 2,840 1,160 2,940 1,047 1,847 1,047 1,847 768 1,279 125 200 1,633 4,357 1,586 4,246 Appendix for Passive Investors, Not Passive Owners -- Page 14

62) Appendix Table 9 Robustness of findings to using end-of-May market cap rankings to select sample This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we instead choose our sample using end-of-May market cap rankings. The data, outcome variables, and specification are the same as in Tables 4-10 except that we choose our sample by ranking stocks within a year using their end-of-May CRSP market cap and selecting our sample to only include stocks ranked 750th largest through 1250th largest each year. The model is estimated over the 1998-2006 period using a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level and reported in parentheses. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Passive % Bandwidth Polynomial order, N Float control Year fixed effects 2-digit industry FE # of firms Observations Ind. directors % Poison pill removal (1) Dependent variable = Ability to Mngt. Ind. for dual call special proposal class shares meeting support % Gov. proposal support % HF activism event ROA (2) (3) (4) (5) (6) (7) (8) 0.448** (0.202) 0.186 (0.123) 0.378** (0.186) -1.800*** (0.366) -0.748** (0.333) 0.352 (0.304) -0.249** (0.114) 0.353** (0.175) 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 1,073 2,949 1,157 2,966 1,053 1,886 1,053 1,886 772 1,294 119 195 1,626 4,431 1,593 4,325 Appendix for Passive Investors, Not Passive Owners -- Page 15

63) Appendix Table 10 Robustness of findings to including control for a firm's lagged stock return This table reports estimates of the second-stage regression of our instrumental variable estimation to identify the effect of institutional ownership by passive investors on our governance and corporate outcome variables when we add a control to account for a firm's stock return in the previous reconstitution year. Specifically, the data, outcome variables, and specification are the same as in Tables 410 except that we add one control to the specification: the stock return in the year prior to the determination of a stock's index assignment [i.e., from end-of-May in year t-1 to end-of-May in year t]. The model is estimated over the 1998-2006 period using a bandwidth of 250 firms around the Russell 1000/2000 threshold and a third polynomial order control for Ln(Mktcap). To demonstrate the robustness of the association between passive ownership and longer-term performance, we include the additional controls for recent movers, used in columns 4-6 of Table 10, when analyzing ROA (column 8). Standard errors, ε, are clustered at the firm level. The symbols *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Dependent variable = Ind. Ability to Ind. for Mngt. Gov. Poison pill directors call special dual class proposal proposal removal % meeting shares support % support % Bandwidth Polynomial order, N Float control Year fixed effects Control for lagged stock return # of firms Observations (4) (5) ROA (6) (7) (8) 0.604* (0.326) -0.160* (0.0840) 0.466*** (0.128) (1) Passive % (3) HF activism event (2) 0.675*** (0.163) 0.209** (0.0846) 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 250 3 yes yes yes 1,075 2,856 1,111 2,855 1,046 1,850 1,046 1,850 746 1,252 126 201 1,537 4,185 1,483 4,068 0.355*** -1.021*** -0.711*** (0.119) (0.184) (0.268) Appendix for Passive Investors, Not Passive Owners -- Page 16

64) Biographical Sketch of Authors Ian Appel is an assistant professor of finance at the Carroll School of Management at Boston College. His research interests are in the areas of corporate finance, law and finance, and institutional investors. His recent research examines the effects of shareholder litigation and passive institutional investors. Ian received his Ph.D. in finance from The Wharton School at the University of Pennsylvania. Todd Gormley is an Assistant Professor of Finance at The Wharton School at the University of Pennsylvania, where he teaches advanced corporate finance in the MBA program and empirical methodologies in the PhD program. Professor Gormley’s research focuses on how financial sector competition affects the local economy, how external governance affects managerial choices, and how individuals and managers respond to risk. His recent research examines whether passive institutional investors are passive owners and whether managers have an underlying preference to “play it safe” when external governance is weakened. Professor Gormley received his Ph.D. in Economics from the Massachusetts Institution of Technology. Donald Keim is the John B. Neff Professor of Finance and Director of the Rodney L. White Center for Financial Research at the Wharton School, University of Pennsylvania. His research has dealt with: the relation between stock returns and predetermined variables (dividend yields, market cap, earnings/price ratios, and calendar turning points); tests of asset pricing models; the behavior of institutional investors with respect to trading, stock holdings, and corporate governance; the investment choices of participants in defined contribution pension plans; and the risks and returns of stock market-based real estate investments. He received an MBA and PhD from the University of Chicago.