Greed Is Good - UW Tacoma Digital Commons

Greed Is Good - UW Tacoma Digital Commons

Greed is Good: The Agency Costs of Blockholder Philanthropy This Draft: May 14, 2013 Roger White* School of Accountancy Robinson College of Business...

226KB Sizes 0 Downloads 5 Views

Greed is Good: The Agency Costs of Blockholder Philanthropy

This Draft: May 14, 2013

Roger White* School of Accountancy Robinson College of Business Georgia State University

*[email protected]

I thank Nerissa Brown, Xiaohua Fang, Steve Fuller, and Lynn Hannan for comments that greatly improved this manuscript.

Greed is Good: The Agency Costs of Blockholder Philanthropy

Abstract Recent experimental research in corporate social responsibility suggests that principal philanthropy offers benefits to the firm. I test this finding using archival data in a natural experiment. In publically traded firms, I find that charitable pledges by blockholders create agency problems that overwhelm any benefits and destroy shareholder value. This effect is stronger when the blockholder has, beyond his economic incentives, a fiduciary duty (as a director or fund manager) to monitor the firm and its managers. I attribute these findings to small investors relying on the self-interest of major shareholders to monitor their shared investment. A charitable pledge lessens the market’s expectation of the philanthropic blockholder’s self-interest, which reduces the ability of small investors to rely on him (and his preference for wealth-maximization) to monitor the firm.

Introduction Corporate social responsibility is an increasingly popular topic in business research (see Moser & Martin, 2012 and Taneja, Taneja, & Gupta, 2011 for overviews from the accounting and management/ethics standpoints, respectively). Most of these studies focus on managers taking charitable actions with owner resources. However, of the approximately $300 billion given to charity in the US in 2011, corporations were responsible for only about 5%, whereas individuals were responsible for 73% (Nichols, 2012).

1

The charitable actions of individuals have received, to the best of my knowledge, no serious examination from either academic or practitioner researchers in business. This is unfortunate, as many of these individuals are shareholders (principals) of publically traded corporations whose charity could affect the value of the firms they own and monitor. I attempt to address this gap in the literature and examine the effect of individual shareholder charity on firm value. Examining individual philanthropy is complicated by data issues. While corporations must disclose financial statements to the public for reasons of transparency, individuals have no such obligation. Given this matter of data availability, I am forced to rely on a natural experiment in which individuals publically announce their philanthropy. While it is popular for wealthy individuals to announce and celebrate their charitable gifts, I focus on a specific philanthropic drive, The Giving Pledge, for the purposes of this paper. The Giving Pledge is a charitable drive instigated by Bill Gates and Warren Buffett in 2010 (Banjo & Guth, 2010). The charity drive seeks to secure pledges from super wealthy individuals who undertake to donate or bequeath to charity at least half of their wealth. SEC 1

Giving USA’s Annual Report on Philanthropy only includes charitable donations. It is difficult to account for the amount firms spend in other CSR activities (e.g. green investing).

1

filings reveal that many of the signatories to The Giving Pledge are blockholders (own greater than 5% of shares) of publically traded corporations. After identifying the subset of The Giving Pledge participants who own blocks of US corporations, I use event studies to identify investors’ reaction to the news that a blockholder has pledged a substantial portion of their wealth to charity. Behavioral theories based in psychology generally suggest that the preferences of a firm’s stakeholders for altruism and charity will result in positive abnormal returns to a firm’s socially responsible actions, such as philanthropic pledges by blockholders (see Jones, 1995 for a review of stakeholder theory). Agency theory, however, suggests that small shareholders rely on the self-interest of blockholders to monitor managers. If a major charitable pledge by a blockholder signals weakening self-interest (and increasing selflessness), then smaller shareholders may fear that the philanthropic blockholder 2

has incentives misaligned with their own, and subsequently question his ability to act as an effective monitor. If blockholder charity creates this agency problem, and enough small shareholders lose trust in the ability of the philanthropic blockholder to monitor their investment, firm value could suffer. I weigh these effects, the stakeholders’ preferences for altruism versus the agency problem of selfless blockholders. I find that the philanthropic pledges of blockholders are met with a resoundingly negative market reaction of about -4.6% (significant at p < 0.01 level), which corresponds to the destruction of about $20 million ($405 million) in shareholder value for the median (mean) firm in my sample. These losses are more severe when the charitable blockholder has formal monitoring responsibilities (serving on the board of directors or acting as the manager of a hedge fund’s stake). In sum, these results suggest that socially responsible 2

I use male pronouns in this paper with apologies to Shelby White, the single female in my sample of 32 philanthropic blockholders.

2

actions by an owner may bring more costs than benefits to the firm in multi-owner settings where the owners rely on one another’s self-interested preferences for wealth maximization (as is typical with publically traded, widely held corporations). Hypothesis Development and Theory Acting in a socially responsible manner brings many benefits to the firm. For example, the literatures in management and accounting have drawn links from CSR to improved firm financial performance (for a review, see Orlitzky, Schmidt, & Rynes, 2003), reporting quality (Y. Kim, Park, & Wier, 2012), information environment (Schadewitz & Niskala, 2010), and employee effort (Porter & Kramer, 2002). The vast majority of this research focuses on costly, socially responsible decisions that managers make with owner resources. Whether managers should take such actions is a debated subject (see Jensen, 2001 for an overview of the advantages of having managers focus on value maximization relative to stakeholder preferences and social responsibility), but what is not is that CSR occurs in various forms across the market, and its popularity seems to be growing (Lim & Tsutsui, 2012). While most archival CSR research focuses on the actions of managers in firms, more money is donated to charity by individuals. The philanthropic activities of the super wealthy are particularly newsworthy, and it is not uncommon for the media to note their positions of corporate authority and ownership when discussing their charity (e.g. Willett, 2012). On an annual basis, individuals in the US donate to charity about fifteen times the sum of all corporate gifts. While CSR can take forms other than direct giving, this ratio suggests that the socially responsible actions of owners and insiders of large corporations are perhaps economically significant.

3

Analytical and experimental results in CSR research can lend some insight as to the effect of owner social responsibility, as CSR models and experiments rarely distinguish between owners and managers (generally, these studies examine a simple principal-agent model). For example, employee altruism that results from the philanthropy of their employer is unlikely to distinguish between giving initiated by managers and giving initiated by owners (Aguilera, Rupp, Williams, & Ganapathi, 2007). Likewise, the increased effort employees contribute to philanthropic employers is also unlikely to disappear when the owner philanthropy happens at the shareholder, rather than manager level (Balakrishnan, Sprinkle, & Williamson, 2011). Additionally, the ability to use CSR to build a goodwill reserve as a form of insurance in the case of poor financial performance (Peloza, 2006) or disaster of an industrial or environmental nature (Porter & Kramer, 2006) should extend, to a degree, to the owners as well as managers. The reputational insurance discussed in Peloza (2006) may be especially important to wealthy owners, as their reputation affects their ability to influence managers, institutions, and potential target firms. This set of literature, generally based in psychology theory (e.g. altruism, gift-exchange), supposes that social responsibility on behalf of the principal is generally associated with benefits to the firm. Balakrishnan et al. (2011) confirm this in a recent experimental study and find that, with fixed wages, agents contribute more effort to philanthropic principals than greedy principals. I move this theory into the multi-owner setting common among publically traded firms in the US and use it to form an expectation as to the market reaction to the charitable pledges of blockholders. I identify this hypothesis, speculating a positive market reaction to the philanthropic pledge of a blockholder, as H1P, my first hypothesis with a basis in psychology theory.

4

H1P: A firm will experience a positive market reaction to the news that a blockholder has made a major charitable pledge. H1P is strongly supported by experimental research that makes use of a simple principal agent model (Balakrishnan et al., 2011). Moving from the principal-agent model into a world where owners and managers are separate, however, introduces agency costs (Jensen & Meckling, 1976). Applying our existing understanding of CSR to the realm of blockholder charity necessitates examining whether or not agency problems involving blockholders exist that would undermine the validity of prior results that rely on simpler ownership models. The results of Shleifer & Vishny (1986) and Edmans (2009), among others, suggest that such an agency problem may exist. This blockholder literature on agency theory suggests that blockholders play an important role in monitoring management, and that small shareholders free ride on this monitoring. Blockholders act as reliable monitors of the firm (and managers in particular) for their own self-interest, as expending resources to monitor managers is wealthmaximizing to a point. The costs of monitoring managers are generally fixed, and blockholders hold the residual rights to enough of the firm to recoup the fixed costs of monitoring while dispersed shareholders tend not to. If a blockholder were to signal that his self-interest was weakening, or was lower than expected, small shareholders may take this as bad news. Small shareholders rely on the selfinterest of blockholders to monitor the firm and its managers, and weaker preferences for selfinterest on the part of the blockholder could mean weaker incentives for the blockholder to monitor managers. The results of Edmans (2009) suggest that self-interested monitoring by blockholders encourages managers to focus on long term firm profits and discourages managerial myopia. These results hold for small blocks of stock (greater than 5% but less than the threshold

5

for control), as even these small blockholders are generally well informed, and threaten to sell the stock if the (closely monitored) managers disappoint. Edmans (2009) labels this increased investment efficiency of managers monitored by self-interested blockholders a social benefit, as it increases the value of smaller, dispersed shareholdings as well. I use this stream of literature to motivate my first hypothesis based in agency theory (H1A), that the selfless charitable pledge of a blockholder reduces firm value. Note that this agency theory hypothesis, H1A, offers the opposite prediction of my first hypothesis based in psychology theory, H1P. H1A: A firm will experience a negative market reaction to the news that a blockholder has made a major charitable pledge. As discussed in motivating my first set of hypotheses on blockholder philanthropy, behavior observed (Balakrishnan et al., 2011) in simple principal agent environments may conflict with the reality of incentives in more realistic situations with multiple owners who are generally separate from management (Edmans, 2009). However, while altruism motivates agents to contribute more to philanthropic principals, experimental research suggests that agent contributions to the principal increase when the principal has more authority to be generous not only to philanthropic causes, but also to the agent (Cox, 2004). Balakrishnan et al., (2011) confirm this result, and find that the ability of the charitable principal to reciprocate agent contributions increases these agent contributions. This line of research generally concludes that agents believe principals more generous with charity will in turn be more generous to them (Godfrey, 2005). While the possibility of reciprocity may not serve to influence employee effort in the case of outside blockholder charity, it could serve to motivate employees to contribute more to the

6

firm when the philanthropic blockholder has some input on employee compensation. The data on blockholder charitable pledges I use includes the pledges of some directors. The philanthropy of these director blockholders may induce higher contributions of effort among employees, as prior research suggests these employees expect such generosity to flow to them as well. Director charity may have the greatest impact in increasing the effort level of upper level managers (like CEOs), as directors play an explicit role in setting the compensation contracts for these employees. This stream of literature goes further than that motivating H1P, which states that philanthropic principals elicit higher levels of effort from their agents. Specifically, the results of Cox (2004) and Balakrishnan et al. ( 2011) suggest that effort is increased to a greater degree when the principal has the ability to reciprocate the agents’ contributions, as would be the case when the principal serves on the board of directors. I test this hypothesis, the second stemming from psychology theory, as H2P. H2P: The charitable pledges of director blockholders will elicit a higher market reaction than the charitable pledges of blockholders who do not serve on the board. This research based on laboratory experiments suggests that the ability of philanthropic principals to reciprocate agent effort will lead to higher levels of effort and increased firm value. Again, it is important to identify agency problems that could complicate this relation when moving from the lab to the ownership structures observed in publically traded US firms. While directors do have the authority to set compensation contracts for senior managers, they also bear the fiduciary responsibility to monitor managers on behalf of the shareholders.

7

In general, research has shown that blockholding directors effectively respond to their economic and fiduciary duties when it comes to monitoring management on shareholders’ behalf. Beasley (1996) finds that blockholding directors lower firm fraud risk. Gul & Tsui (2001) find that the audit risk of firms with blockholding directors is lower, and that blockholding directors are effective at preventing managers from taking advantage of firms with high levels of free cash flow. Bhagat, Carey, & Elson (1998) examine firm performance and observe that not only do firms with blockholding directors see higher returns, but that they are more likely to replace failing CEOs. This literature consistently shows that shareholders rely on blockholding directors, even more so than outside blockholders, to serve as strong monitors. When these blockholding directors signal a weakening of their self-interested incentives to maximize their wealth, those who rely on these incentives to protect their own wealth (dispersed shareholders) may interpret this signal negatively. I use this agency problem to motivate my second hypothesis grounded in agency theory, H2A. Note that his hypothesis predicts the opposite of my second hypothesis based on psychology theories, H2P. H2A: The charitable pledges of director blockholders will elicit a lower market reaction than the charitable pledges of blockholders who do not serve on the board. Another class of charitable investor that appears frequently in my data is hedge fund 3

managers. While these managers do not personally own large blocks of stock, they exercise control over blocks through their position as a fund manager, and they face economic incentives similar to owners. Philanthropy by hedge fund managers likely endows their hedge fund with a 3

I use the terms hedge fund, institutional investor, and institution interchangeably throughout this manuscript (also, hedge fund manager and institutional investor principal). Most of the institutional investors in my sample are hedge funds or private investment partnerships (that operate similarly to hedge funds).

8

socially responsible reputation. Research into socially responsible institutional investors suggests that such a reputation may be valuable. Graves & Waddock (1994), for example, suggest that investors are attracted to socially responsible funds and are willing to pay a premium for them (Shank, Manullang, & Hill, 2005 find support for this conjecture). The results of Mackey, Mackey, & Barney (2007) and Pava & Krausz (1996) suggest that this is driven by a subset of investors who are willing to pay a very high premium for socially responsible funds. Beyond those investors willing to provide capital to socially responsible institutional investors at below market rates, a budding stream of literature has begun to investigate the role of social responsibility in opening up investment opportunities for institutions, as institutional CSR benefits the image of large investors (Bénabou & Tirole, 2010). Petersen & Vredenburg (2009) survey institutional investor principals to determine the benefits of this image enhancement. Beyond increasing their access to resources, the survey results illustrate that an improved image of the investing institution reduces problems with the government and surrounding communities, as well as softens public perception and reduces hostile media attention. This stream of research suggests a reputation for being socially responsible allows an institutional investor preferential access to resources, less government oversight, and the ability to attract investors willing to provide capital at discounted rates. When hedge fund managers join The Giving Pledge, their fund likely reaps these benefits from being recognized as socially responsible. I test this supposition in H3P, my third hypothesis based on psychology theory (altruism, reputational capital).

9

H3P: The charitable pledges of the principals of institutional blockholders (hedge fund managers) will elicit a higher market reaction than the charitable pledges of blockholders who are unaffiliated with an institution. While psychology based research suggests that the philanthropic activity of hedge fund managers may accrue benefits to the firm (generates altruism which leads to less oversight, better access to resources, etc.), agency problems may complicate the relation. Similar to directors, fund managers have not only economic incentives, but a fiduciary duty (to the shareholders of their fund) to monitor their fund’s holdings. I focus on the hedge fund monitoring literature to motivate my final hypothesis. This research generally finds that hedge funds that actively intervene (a form of monitoring) with managers create value for their shareholders. Clifford (2008) and Kahan & Rock (2007) are recent additions to this stream of literature that note the particularly strong monitoring role that hedge funds play. Hedge fund managers are heavily incentivized to actively monitor firms the fund owns, and research in hedge fund manager compensation suggests that fund performance is sensitive to fund manager incentives (see Edwards & Caglayan, 2001 and Agarwal, Daniel, & Naik, 2009). If the self-interested, wealth maximizing preferences of hedge fund managers weaken, as signaled by personal charitable pledges, then these managers may be less incentivized (by the prospect of profit sharing contracts) to pursue effort intensive monitoring activities as aggressively as shareholders previously expected. If this is the case, and charitable pledges by hedge fund managers do signal that their incentives to monitor the managers of their fund’s holdings for a share of fund profits is weaker than expected, then such philanthropy creates an agency problem.

10

I investigate this agency problem in H3A, my third hypothesis motivated by agency theory. H3A: The charitable pledges of the principals of institutional blockholders (hedge fund managers) will elicit a lower market reaction than the charitable pledges of blockholders who are unaffiliated with an institution. My hypotheses predict the anticipated market reactions to the philanthropic pledges of blockholders. My three hypotheses based in psychology theory anticipate positive market reactions to the announcement of blockholder charity. Behavioral preferences for philanthropy on the part of investors, customers, and employees should serve to increase the market value of firms whose major owners are charitable (H1P). This effect should be larger when these owners have the ability to reciprocate increased agent contributions to the firm (in the role of directors, H2P) and when the charity lowers the cost of capital and government oversight of investors (as in the case of hedge fund managers, H3P). I motivate an opposing set of hypotheses using agency theory. These hypotheses are based on the theory that small shareholders rely on the self-interest of blockholders to monitor managers, and that blockholder philanthropy signals a weakening of the wealth-maximizing preferences of blockholders on which smaller shareholders rely (H1A). Given the stronger monitoring roles that directors (H2A) and hedge fund managers (H3A) play, agency theory suggests that philanthropy on their part is particularly harmful for small shareholders (and as a result, destroys firm value). Data and Empirics To determine the effects of shareholder philanthropy on firm value, I examine the abnormal returns to a firm upon the announcement of a major charitable pledge by a large shareholder. I use the signatories of The Giving Pledge as my sample of charitable major

11

shareholders. The Giving Pledge is a charity drive, initiated by Warren Buffett and Bill Gates, that aims to secure pledges from wealthy individuals (median net worth in my sample is $2.2 billion) to donate to charity at least half of their wealth during their lifetime or upon their death. By examining the market reaction to the pledge announcement, this method abstracts from issues involving the specific charity these shareholders choose to donate to, the amount donated, and the source of the funds to be donated (as none of this information is released at the announcement of the pledge, or even decided upon at that time). Besides Gates and Buffett, notable signatories (not all appear in my sample) include Larry Ellison (Oracle co-founder), Paul Allen (Microsoft co-founder), Arthur Blank (Home Depot co-founder), Michael Bloomberg (Bloomberg L.P. founder and mayor of New York City), Barron Hilton (hotel magnate), Carl Icahn (activist investor), George Lucas (film producer), Elon Musk (co-founder of Paypal and Tesla Motors, founder of SpaceX), Ted Turner (media mogul), and Mark Zuckerberg (co-founder and CEO of Facebook). See Table 1 for a list of the signatories in my sample. My original sample includes all the philanthropists that signed The Giving Pledge from its inception in 2010 through the end of 2011. I identify 32 of these signatories as major owners of U.S. corporations through SEC filings (firms are required to disclose owners of blocks greater than 5%). These SEC filings are updated at least once a year, and I code signatories as major owners if they are included in a company’s filing in the 365 days prior to their signing The Giving Pledge. The 32 signatories I identify hold major stakes in 122 firms for a total of 124 observations (some firms have multiple major owners sign the pledge). Table 1 includes a column reporting the number of holdings for each of the signatories in my sample.

12

One of the 32 signatories in my sample is James Simons, recently retired founder, CEO, and manager of Renaissance Technologies, a $15 billion hedge fund. 44 of my 124 observations are Renaissance Technologies holdings, making James Simons a very influential observation in my sample. As such, I report all of my results in two panels, one including the 44 observations attributable to James Simons (for a total of 124 observations), and one excluding the 44 observations attributable to James Simons (for a total of 80 observations). My results are generally robust to his exclusion. To test H1P against H1A, I run event studies on the announcement dates on which the major shareholders announce their signing of The Giving Pledge. I use the value weighted index as a benchmark, and I rely on the market model to discern abnormal returns. I determine statistical significance of these mean (median) abnormal returns using the standardized crosssectional z score of Boehmer, Masumeci, & Poulsen (1991) (Cowan, 1992). While I tabulate abnormal returns for a variety of windows, I focus on the (-30, -1) runup period in my analysis. The super wealthy shareholders making the charitable pledges in my sample likely do so only after consulting with their lawyers, estate planners, investment managers, and accountants. Similar to merger and acquisition announcements, I expect such consultations to result in information leakage, such that (at least part of) the news is already incorporated into market prices when it is publically announced (both Cumming & Li (2011) and Gao & Oler (2012) discuss this run-up period return). I test H2P against H2A and H3P against H3A using regressions on the abnormal returns generated by the event studies. I pit H2P against H2A to determine whether director shareholder charity leads to a higher or lower market reaction. I am able to determine through the SEC filings if the charitable shareholder is a director. About 25% of the firm-events in my sample

13

come from firms in which the major charitable shareholder is also a director (about 39% of the remainder if I remove the 44 firm-events attributable to James Simons). I test H3P against H3A to learn whether the abnormal returns around the charitable pledges of hedge fund managers differ from those of individual shareholders. I determine whether the charitable owner owns his shares individually or controls them through a hedge fund while gathering individual holdings data from SEC filings. About half of the firm-events in my sample come from firms in which the major charitable shareholder controls his holding through an institution (about 30% of the remainder if I remove the 44 firm-events attributable to James Simons). Beyond the variables of interest for my hypotheses, I include a number of controls in the cross sectional regressions. Owner age may affect the market reaction if investors are concerned with the owner, his estate, or charity recipient selling off a large block of shares in the near future (soon before or upon their death, see Slovin & Sushka, 1993). Controlling for signatory age alleviates the concerns that the abnormal returns I observe are driven by liquidity or blockholder exit concerns. I also control for prior charitable shareholder giving, as market expectations may differ in cases in which the charitable shareholder has a history of being socially responsible. I control for these instances by including an indicator variable for whether or not the shareholder is 4

affiliated with a private foundation. This data comes directly from The Giving Pledge, as signatories report their prior association with charitable foundations. Additionally, I control for both charitable shareholder total net worth (natural log) and the percentage of net worth the charitable shareholder has tied up in the firm. The market reaction to 4

The Gates Foundation is a popular example and has received considerable gifts from both Bill Gates and Warren Buffett in the past.

14

the charitable pledge of higher net worth blockholders may be less severe, as they have less expected marginal utility of wealth to begin with. Likewise, the market reaction to an overinvested, undiversified major shareholder signing The Giving Pledge may reflect the strategic contributions expected of the undiversified shareholder in an attempt to protect their wealth. I draw net worth figures from The Giving Pledge, as participants report it at the signing 5

announcement. I compare participants’ net worth to the value of their major holding(s) using stock price data from the CRSP database. Furthermore, I control for whether the charitable shareholder controls the firm, defined as 6

owning more than 20% of the shares (as in Villalonga & Amit, 2006).

In these cases,

incentive alignment problems are already apparent (see Zingales, 1995; Nenova, 2003; and Shleifer & Vishny, 1997), and minority shareholders may actually welcome socially responsible actions by the controlling shareholder, as weaker preferences for self-interest may lead the controlling shareholder to lessen their expropriation of private benefits of control. Finally, I include a small set of firm level controls to proxy for the value of outside monitoring. External monitoring generally benefits diversified firms more than specialized firms, as diversification can create agency costs as the firm attempts to allocate resources between divisions (Denis, Denis, & Sarin, 1997 and Rajan, Servaes, & Zingales, 2000). Similar agency problems affect firms with foreign operations (Burgman, 1996). Given these issues, I include indicator variables in my regressions to proxy for firms that operate in multiple industries and countries. I also control for firm leverage, as debt has been shown to create both agency costs (e.g. Jensen & Meckling, 1976; W. S. Kim & Sorensen, 1986; and Mello & Parsons, 1992)

5

Edward Rose III did not report his net worth at the time of his signing. I estimate this figure by compounding his latest reported net worth by the market rate of return to his signing date. 6 Using their percentage of ownership as a continuous measure yields similar results.

15

and benefits (e.g. Jensen, 1986 and Korteweg, 2010) that could correlate with the value of outside monitoring. I also control for firm size, as evidence suggests outside monitoring is more beneficial in reducing agency costs to large firms (e.g. Moeller, Schlingemann, & Stulz, 2004 and Singh & Davidson III, 2003). I run cross sectional OLS regressions, including the above control variables, to test H2P against H2A and H3P against H3A. These regressions use the (-30, -1) day run-up announcement abnormal return as the dependent variable, but using other run-up windows (including and excluding the actual announcement date) yields similar results. Regression standard errors are 7

robust and clustered in two dimensions by year and signatory. Given my small sample size, I also include in my regression tables an F-statistic for each of my models. F-statistics corresponding to regressions significant at the p=0.05 level or better are in bold. Results Table 1 lists the names and details of the charitable shareholders included in my study. Many of these names are recognizable, and the typical signatory can be thought of as a very successful entrepreneur (usually in the technology or energy sectors) in his late sixties or early seventies with a net worth of around $2.2 billion. Table 2 reports the abnormal returns to firms that these signatories hold large stakes in at (and prior to) the announcement that they have signed The Giving Pledge. While the actual (-1, +1) announcement period abnormal returns are not statistically different from zero, the strongly negative run-up period abnormal returns suggests that news of the impending signing leaks and 8

is viewed poorly by investors. Panel A of Table 2 reports the mean (median) abnormal return in

7

Clustering standard errors by announcement date leads to similar results. In untabulated robustness tests I find that the run-up period abnormal returns do not reverse in the weeks after the announcement. 8

16

the (-30, -1) day run-up window to be -4.6% (-4.39%), significant at the p< 0.01 level (Z = -2.9). This suggests that the median firm in my sample (market value = $430 million) sees $20 million in shareholder value destroyed in the run-up to the announcement that a major shareholder has signed The Giving Pledge. For the mean firm in my sample (market value = $8.8 billion), the corresponding destruction in shareholder value is around $405 million. Panel B of Table 2 excludes the 44 observations attributable to hedge fund principal James Simons. These results are directionally consistent with the results of the full sample, but have slightly weaker statistical support. In general, I take this as evidence that the third of my observations attributable to James Simons are not driving the directional results of my study, and that beyond James Simons, major shareholder philanthropy destroys firm value. These event study findings indicate that in widely held, publically traded companies, 9

shareholder philanthropy destroys firm value. This settles the question between H1A (supported) and H1P (rejected), and suggests that costs of some sort overwhelm the benefits of principal philanthropy. Agency problems are a potential source of such costs, as the experimental results underlying the majority of my psychology based hypotheses abstract from such considerations. I test H2A and H3A (my agency theory hypotheses) against H2P and H3P (my hypotheses based in behavioral psychology), respectively, to investigate this possibility. Table 3 describes the variables in the cross-sectional regressions used to test H2P against H2A and H3P against H3A. In the broad sample (Table 3; Panel A) of 124 firms, it appears that the charitable shareholder is a director in 25% of the cases, a hedge fund manager in about half of the cases (only 30% in Panel B, where James Simons is excluded), the controlling shareholder in 23% of the cases, and generally has 9% of his net worth invested in the firm. 9

This finding is contrary to Balakrishnan et al (2011), but in a setting that incorporates the agency problems of dispersed ownership and the separation of ownership and management. I consider my results a boundary condition to Balakrishnan et al (2011).

17

Table 4 reports the cross-sectional regressions, incorporating the controls from Table 3, to test H2P against H2A and H3P against H3A. The dependent variable in all of these models is the (-30, -1) day announcement run-up abnormal return. Models 1 and 3 regress this abnormal return on the variables of interest for H2P/H2A (Director) and H3P/H3A (Institutional Shareholder Principal), respectively. Models 2 (Director) and 4 (Institutional Shareholder Principal) add control variables to Models 1(Director) and 3 (Institutional Shareholder Principal). Model 5 includes both indicator variables of interest, Director and Institutional Shareholder Principal, as well as the control variables. Panel A includes the entire sample, and Panel B excludes the 44 observations attributable to James Simons. I include the predictions of my various hypotheses on the Director and Institutional Shareholder Principal indicator variables in an abbreviated column on the far right. I first examine the Director indicator variable to test between H2P and H2A. I find strong evidence to suggest that agency problems are driving the negative abnormal returns to the announcement of shareholder philanthropy. While the regressions including the director indicator alone (Table 4; Panels A and B; Model 1) do not offer any predictive power (insignificant regressions based on the F-Statistic), it registers as negative and significant in both models that include control variables (Models 2 and 5). In Model 5, with the other control variables set at the median (mean), the run-up abnormal return to the announcement of a major outside shareholder’s pledge is -8.6% (-2.6%). For a shareholder serving on the board of directors, the analogous announcement run-up return is -17.2% (-11.2%). For the median firm, these percentages (median values) correspond to the philanthropic pledge of a major outside shareholder destroying $37 million in shareholder value and that of a director destroying $74 million in shareholder value. The mean firm in my sample is larger than the median by a factor

18

of about 20. As such, the economic significance of these results for the mean firm are larger by an order of magnitude. I next examine the Institutional Shareholder Principal indicator variable to test between H3P and H3A. In Models 3 and 5 of Table 4, Panel A, the Institutional Shareholder Principal indicator variable is significantly negative. This would suggest that, including the effects of other control variables, philanthropic pledges by institutional shareholder principals are particularly damaging to firms in which their institution holds a stake greater than 5%. If we exclude the effects of James Simons (Panel B of Table 4), the institutional investor principal responsible for one third of my sample, these results hold in direction, if not always significance. Model 5 is the only model in Panel B of Table 4 that is both statistically significant and includes the Institutional Shareholder Principal indicator variable. This variable appears with a strong negative coefficient which offers further support for H3A. Specifically, this alleviates any concern that James Simons is driving my findings on institutional investor principals. I evaluate Model 5 of Panel A (Table 4) for economic significance. With the other control variables set at the median (mean), firms see negative abnormal returns of -5.2% (-2.9%) in the run-up to the announcement that a noninstitutional blockholder has joined The Giving Pledge. For firms whose charitable blockholders are institutional investor principals (hedge fund managers), the corresponding negative abnormal run-up is -8.6% (-6.3%). These differences are statistically significant. For the median firm, this amounts to $22 million in shareholder value destroyed when a noninstitutional blockholder joins The Giving Pledge and $37 million in shareholder value destroyed when an Institutional Shareholder Principal joins The Giving Pledge. Again, the analogous losses for the mean firm in my sample are larger by at least an order of magnitude.

19

Conclusion Experimental research into principal philanthropy finds that such charity is generally an NPV positive project for the firm, as principal gifts are recouped through the agent’s increased effort (Balakrishnan et al., 2011). Other work in CSR supports this view, as charity affiliated with a specific company has also been linked to employee altruism (Aguilera et al., 2007), insurance against disaster (Peloza, 2006 and Porter & Kramer, 2006), and good outcomes in general (Beurden & Gössling, 2008). I test this hypothesis using a natural experiment, a collection of 32 major shareholders of 122 publically traded firms making comparable charitable pledges, and find that agency costs crowd out any benefits of principal philanthropy in my sample. While I examine only participation in The Giving Pledge, these results likely extend to more commonplace displays of owner, director, and fund manager philanthropy, at least in settings where they are expected to play a monitoring role. Prior literature on agency theory suggests that small shareholders rely on the self-interest of large shareholders to monitor the firm and its managers (Holderness & Sheehan, 1988 and Shleifer & Vishny, 1986). When these large shareholders make a charitable pledge, it sends a signal that they are not as self-interested as previously thought. For investors, this weakening of monitoring incentives appears to outweigh the benefits of CSR, and such announcements destroy about $20 million in shareholder value for the median firm (4.6% of firm value). This negative reaction is more pronounced when the charitable blockholder has been tasked with fiduciary monitoring responsibilities, specifically in the role of director or manager of a hedge fund’s stake. In particular, these results imply that directors and fund managers who control major stakes are failing in their fiduciary duties to their electing shareholders (the owners of the firm or

20

fund that elected/hired them) when they pledge their own funds to charity. Unless these fiduciary agents are able to bind themselves to remaining vigilant monitors, the only clearly evident solution to this issue is for directors and fund managers controlling major blocks to step down from their positions before embarking on major personal campaigns of philanthropy. To remain in their positions calls into question their incentives to monitor the firm, which creates an agency problem that destroys shareholder value of the very firm they bear a fiduciary responsibility to protect. This agency cost of shareholder philanthropy also complicates the suggestion that managers should reduce firm-driven CSR to allow more wealth to flow to the shareholders for their own charitable (and other) ends. The results of this paper clearly suggest that shareholder charity has its own costs, which should be considered in future analyses of the alternatives to corporate giving. Beyond the contributions of this paper to the literatures in corporate social responsibility and agency theory, I encourage other archival researchers to look to experimental work for inspiration. Experimental research in business and economics offers a multitude of predictions based on human behavior. Investigating how behavior changes when it emerges from the lab and encounters the frictions of reality, like agency costs, can further our understanding of real world incentives, both economic and psychological.

21

References 1. Agarwal, V., Daniel, N. D., & Naik, N. Y. (2009). Role of Managerial Incentives and Discretion in Hedge Fund Performance. The Journal of Finance, 64(5), 2221–2256. doi:10.1111/j.1540-6261.2009.01499.x 2. Aguilera, R. V., Rupp, D. E., Williams, C. A., & Ganapathi, J. (2007). Putting the S Back in Corporate Social Responsibility: A Multilevel Theory of Social Change in Organizations. Academy of Management Review, 32(3), 836–863. 3. Balakrishnan, R., Sprinkle, G. B., & Williamson, M. G. (2011). Contracting Benefits of Corporate Giving: An Experimental Investigation. The Accounting Review, 86(6), 1887– 1907. doi:10.2308/accr-10127 4. Banjo, S., & Guth, R. A. (2010, August 3). U.S. Super Rich to Share Wealth. Wall Street Journal. Retrieved from http://online.wsj.com/article/SB10001424052748704017904575409193790337162.html 5. Beasley, M. S. (1996). An Empirical Analysis of the Relation between the Board of Director Composition and Financial Statement Fraud. The Accounting Review, 71(4), 443–465. doi:10.2307/248566 6. Bénabou, R., & Tirole, J. (2010). Individual and Corporate Social Responsibility. Economica, 77(305), 1–19. doi:10.1111/j.1468-0335.2009.00843.x 7. Beurden, P. van, & Gössling, T. (2008). The Worth of Values – A Literature Review on the Relation Between Corporate Social and Financial Performance. Journal of Business Ethics, 82(2), 407–424. doi:10.1007/s10551-008-9894-x 8. Bhagat, S., Carey, D. C., & Elson, C. M. (1998). Director Ownership, Corporate Performance, and Management Turnover. Business Lawyer (ABA), (3), 885. 9. Boehmer, E., Masumeci, J., & Poulsen, A. B. (1991). Event-study methodology under conditions of event-induced variance. Journal of Financial Economics, 30(2), 253–272. doi:10.1016/0304-405X(91)90032-F 10. Burgman, T. A. (1996). An Empirical Examination of Multinational Corporate Capital Structure. Journal of International Business Studies, 27(3), 553–570. doi:10.2307/155438 11. Clifford, C. P. (2008). Value creation or destruction? Hedge funds as shareholder activists. Journal of Corporate Finance, 14(4), 323–336. doi:10.1016/j.jcorpfin.2008.04.007 12. Cowan, A. R. (1992). Nonparametric event study tests. Review of Quantitative Finance and Accounting, 2(4), 343–358. doi:10.1007/BF00939016 13. Cox, J. C. (2004). How to identify trust and reciprocity. Games and Economic Behavior, 46(2), 260–281. doi:10.1016/S0899-8256(03)00119-2 14. Cumming, D., & Li, D. (2011). Run-up of Acquirer’s Stock in Public and Private Acquisitions. Corporate Governance: An International Review, 19(3), 210–239. doi:10.1111/j.1467-8683.2010.00838.x 15. Denis, D. J., Denis, D. K., & Sarin, A. (1997). Agency Problems, Equity Ownership, and Corporate Diversification. The Journal of Finance, 52(1), 135–160. doi:10.1111/j.15406261.1997.tb03811.x 16. Edmans, A. (2009). Blockholder Trading, Market Efficiency, and Managerial Myopia. The Journal of Finance, 64(6), 2481–2513. doi:10.1111/j.1540-6261.2009.01508.x 17. Edwards, F. R., & Caglayan, M. O. (2001). Hedge Fund Performance and Manager Skill. Journal of Futures Markets, 21(11), 1003–1028. doi:10.1002/fut.2102 22

18. Gao, Y., & Oler, D. (2012). Rumors and pre-announcement trading: why sell target stocks before acquisition announcements? Review of Quantitative Finance and Accounting, 39(4), 485–508. doi:10.1007/s11156-011-0262-z 19. Godfrey, P. C. (2005). The Relationship Between Corporate Philanthropy and Shareholder Wealth: A Risk Management Perspective. Academy of Management Review, 30(4), 777–798. doi:10.5465/AMR.2005.18378878 20. Graves, S. B., & Waddock, S. A. (1994). Institutional Owners and Corporate Social Performance. Academy of Management Journal, 37(4), 1034–1046. doi:10.2307/256611 21. Gul, F. A., & Tsui, J. S. L. (2001). Free Cash Flow, Debt Monitoring, and Audit Pricing: Further Evidence on the Role of Director Equity Ownership. AUDITING: A Journal of Practice & Theory, 20(2), 71–84. doi:10.2308/aud.2001.20.2.71 22. Holderness, C. G., & Sheehan, D. P. (1988). The role of majority shareholders in publicly held corporations: An exploratory analysis. Journal of Financial Economics, 20, 317– 346. doi:10.1016/0304-405X(88)90049-9 23. Jensen, M. C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76(2), 323. 24. Jensen, M. C. (2001). Value Maximization, Stakeholder Theory, and the Corporate Objective Function. Journal of Applied Corporate Finance, 14(3), 8–21. doi:10.1111/j.1745-6622.2001.tb00434.x 25. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. doi:10.1016/0304-405X(76)90026-X 26. Jones, T. M. (1995). Instrumental Stakeholder Theory: A Synthesis of Ethics and Economics. Academy of Management Review, 20(2), 404–437. doi:10.5465/AMR.1995.9507312924 27. Kahan, M., & Rock, E. B. (2007). Hedge Funds in Corporate Governance and Corporate Control. University of Pennsylvania Law Review, (5), 1021. 28. Kim, W. S., & Sorensen, E. H. (1986). Evidence on the Impact of the Agency Costs of Debt on Corporate Debt Policy. Journal of Financial and Quantitative Analysis, 21(02), 131–144. doi:10.2307/2330733 29. Kim, Y., Park, M. S., & Wier, B. (2012). Is Earnings Quality Associated with Corporate Social Responsibility? The Accounting Review, 87(3), 761–796. doi:10.2308/accr-10209 30. Korteweg, A. (2010). The Net Benefits to Leverage. The Journal of Finance, 65(6), 2137–2170. doi:10.1111/j.1540-6261.2010.01612.x 31. Lim, A., & Tsutsui, K. (2012). Globalization and Commitment in Corporate Social Responsibility Cross-National Analyses of Institutional and Political-Economy Effects. American Sociological Review, 77(1), 69–98. doi:10.1177/0003122411432701 32. Mackey, A., Mackey, T. B., & Barney, J. B. (2007). Corporate Social Responsibility and Firm Performance: Investor Preferences and Corporate Strategies. Academy of Management Review, 32(3), 817–835. doi:10.5465/AMR.2007.25275676 33. Mello, A. S., & Parsons, J. E. (1992). Measuring the Agency Cost of Debt. The Journal of Finance, 47(5), 1887–1904. doi:10.1111/j.1540-6261.1992.tb04687.x 34. Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2004). Firm size and the gains from acquisitions. Journal of Financial Economics, 73(2), 201–228. doi:10.1016/j.jfineco.2003.07.002

23

35. Moser, D. V., & Martin, P. R. (2012). A Broader Perspective on Corporate Social Responsibility Research in Accounting. The Accounting Review, 87(3), 797–806. doi:10.2308/accr-10257 36. Nenova, T. (2003). The value of corporate voting rights and control: A cross-country analysis. Journal of Financial Economics, 68(3), 325–351. doi:10.1016/S0304405X(03)00069-2 37. Nichols, M. (2012, June 19). U.S. charitable giving approaches $300 billion in 2011. Reuters. New York. Retrieved from http://www.reuters.com/article/2012/06/19/us-usacharity-idUSBRE85I05T20120619 38. Orlitzky, M., Schmidt, F. L., & Rynes, S. L. (2003). Corporate Social and Financial Performance: A Meta-Analysis. Organization Studies, 24(3), 403–441. doi:10.1177/0170840603024003910 39. Pava, M. L., & Krausz, J. (1996). The association between corporate social-responsibility and financial performance: The paradox of social cost. Journal of Business Ethics, 15(3), 321–357. doi:10.1007/BF00382958 40. Peloza, J. (2006). Using Corporate Social Responsibility as Insurance for Financial Performance. California Management Review, 48(2), 52–72. 41. Petersen, H., & Vredenburg, H. (2009). Morals or Economics? Institutional Investor Preferences for Corporate Social Responsibility. Journal of Business Ethics, 90(1), 1–14. 42. Porter, M. E., & Kramer, M. R. (2002). The Competitive Advantage of Corporate Philanthropy. Harvard Business Review, 80(12), 56–69. 43. Porter, M. E., & Kramer, M. R. (2006). Strategy & Society: The Link Between Competitive Advantage and Corporate Social Responsibility. Harvard Business Review, 84(12), 78–92. 44. Rajan, R., Servaes, H., & Zingales, L. (2000). The Cost of Diversity: The Diversification Discount and Inefficient Investment. The Journal of Finance, 55(1), 35–80. doi:10.1111/0022-1082.00200 45. Schadewitz, H., & Niskala, M. (2010). Communication via responsibility reporting and its effect on firm value in Finland. Corporate Social Responsibility and Environmental Management, 17(2), 96–106. doi:10.1002/csr.234 46. Shank, T., Manullang, D., & Hill, R. (2005). “Doing Well While Doing Good” Revisited: A Study of Socially Responsible Firms’ Short-Term versus Long-term Performance. Managerial Finance, 31(8), 33–46. doi:10.1108/03074350510769794 47. Shleifer, A., & Vishny, R. W. (1986). Large Shareholders and Corporate Control. Journal of Political Economy, 94(3), 461–488. doi:10.2307/1833044 48. Shleifer, A., & Vishny, R. W. (1997). A Survey of Corporate Governance. The Journal of Finance, 52(2), 737–783. doi:10.1111/j.1540-6261.1997.tb04820.x 49. Singh, M., & Davidson III, W. N. (2003). Agency costs, ownership structure and corporate governance mechanisms. Journal of Banking & Finance, 27(5), 793–816. doi:10.1016/S0378-4266(01)00260-6 50. Slovin, M. B., & Sushka, M. E. (1993). Ownership Concentration, Corporate Control Activity, and Firm Value: Evidence from the Death of Inside Blockholders. The Journal of Finance, 48(4), 1293–1321. doi:10.1111/j.1540-6261.1993.tb04755.x 51. Taneja, S. S., Taneja, P. K., & Gupta, R. K. (2011). Researches in Corporate Social Responsibility: A Review of Shifting Focus, Paradigms, and Methodologies. Journal of Business Ethics, 101(3), 343–364. doi:10.1007/s10551-010-0732-6 24

52. Villalonga, B., & Amit, R. (2006). How do family ownership, control and management affect firm value? Journal of Financial Economics, 80(2), 385–417. doi:10.1016/j.jfineco.2004.12.005 53. Willett, M. (2012, September 25). 20 Generous Billionaires Who Could Be The Next Bill Gates. Business Insider. Retrieved February 25, 2013, from http://www.businessinsider.com/the-top-20-emerging-philanthropists-2012-9 54. Zingales, L. (1995). What Determines the Value of Corporate Votes? The Quarterly Journal of Economics, 110(4), 1047–1073. doi:10.2307/2946648

25

Table 1 Table 1 reports the names, number of major holdings, date of pledge, age at time of pledge, and net worth (in billions $USD) of the 32 major shareholders in my sample.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Major Holdings (Greater than 5%) at Name time of signing The Giving Pledge Nicolas Berggruen 1 Warren Buffett 10 Steve Case 1 Lee Cooperman 4 John Doerr 2 Larry Ellison 3 Phillip Frost 6 Bill Gates 12 Harold Hamm 1 Jon Huntsman 1 Carl Icahn 3 Irwin Jacobs 1 George B. Kaiser 4 Vinod Khosla 1 Sidney Kimmel 1 Ken Langone 1 Alfred E. Mann 1 Joe Mansueto 1 Michael Milken 2 George P. Mitchell 1 Pierre Omidyar 1 Ronald O. Perelman 4 T. Boone Pickens 1 Julian H. Robertson, Jr. 1 Edward W. Rose 1 David M. Rubenstein 3 Henry Samueli 1 Walter Scott, Jr. 1 Harold Simmons 5 James Simons 44 Tom Steyer 4 Shelby White 1 124

Date signed The Giving Pledge 8/25/2010 8/4/2010 12/9/2010 9/27/2010 8/4/2010 8/3/2010 4/28/2011 6/6/2010 4/19/2011 6/18/2010 9/21/2010 8/1/2010 7/26/2010 4/28/2011 11/18/2010 7/14/2010 8/4/2010 12/8/2010 12/8/2010 12/8/2010 8/4/2010 8/4/2010 8/4/2010 8/4/2010 4/28/2011 8/4/2010 12/1/2011 6/24/2010 3/10/2011 7/11/2010 6/29/2010 8/1/2010

Age at Pledge 49 80 52 67 59 66 76 55 66 73 74 77 67 56 81 75 84 54 64 91 42 67 82 78 70 61 57 79 80 72 53 71

Net Worth ($ Billion) $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $

2.3 46.0 1.7 2.2 2.5 41.0 2.4 66.0 9.7 0.9 14.8 1.5 10.0 1.4 1.2 1.6 1.0 1.7 2.3 2.0 8.2 12.0 1.2 2.6 0.3 1.9 1.8 2.1 7.1 11.0 1.3 1.0

26

Table 2: Abnormal Returns Table 2 reports the results of the event studies examining the run-up and event returns of the announcement that a major shareholder ( >5%) has signed The Giving Pledge. These results test between H1P against H1A. Table 2, Panel A Panel A reports the results of the event study using the value weighted index. This panel includes the 44 observations attributable to hedge fund manager James Simons. I measure the statistical significance of mean (median) abnormal returns with the Standardized Cross-Sectional Z Score (Generalized Sign Z Score). Window

Firm Events

(-45,-1) (-30,-1) (-15, -1) (-1,+1)

124 124 124 124

Mean Abnormal Return -6.29% -4.58% -3.27% 0.07%

Mean Abnormal Return Z Score -2.64*** -2.90*** -3.26*** 0.03

Median Abnormal Median Abnormal Return Return Z Score -5.05% -1.63 -4.39% -2.72*** -2.96% -2.72*** -0.19% 0.01

Predicted Sign H1A(-) H1P(+) H1A(-) H1P(+) H1A(-) H1P(+) H1A(-) H1P(+)

Table 2, Panel B Panel B reports the results of the event study using the value weighted index. This panel does not include the 44 observations attributable to hedge fund manager James Simons. I measure the statistical significance of mean (median) abnormal returns with the Standardized Cross-Sectional Z Score (Generalized Sign Z Score). Window

Firm Events

(-45,-1) (-30,-1) (-15, -1) (-1,+1)

80 80 80 80

Mean Abnormal Return -2.73% -2.56% -3.05% -0.51%

Mean Abnormal Return Z Score -0.86 -1.46 -3.11** -0.88

Median Abnormal Median Abnormal Return Return Z Score -1.45% -0.05 -1.70% -1.13 -2.55% -2.22** -0.35% -0.48

Predicted Sign H1A(-) H1P(+) H1A(-) H1P(+) H1A(-) H1P(+) H1A(-) H1P(+)

27

Table 3: Summary Statistics Table 3 reports the summary statistics of the variables that appear in the cross sectional regressions. These include the variables of interest for H2P/H2A (Director) and H3P/H3A (Institutional Shareholder Principal), as well as a host of control variables. Table 3; Panel A Panel A reports the summary statistics of the regression variables, including the 44 observations attributable to James Simons. Variable Director (Dummy) Institutional Shareholder Principal (Dummy) Net Worth (in billions, USD) Percent of Wealth Invested Controlling Shareholder (Dummy) Prior Charitable Foundation (Dummy) Charitable Shareholder Age Multiple Segments (Dummy) Foreign Income (Dummy) Leverage Market Capitalization (USD)

Firm Events 124 124 124 124 124 124 124 124 124 124 124

Mean

Std. Dev

0.25 0.54 17.04 9% 0.23 0.91 69.24 0.45 0.51 0.18 $8.8 Bil.

0.43 0.5 20.16 20% 0.43 0.29 8.91 0.5 0.5 0.16 $28 Bil.

1st Median Quartile 0 0 0 0 0 1 0.25 2.4 11 < 1% < 1% 1% 0 0 0 0 1 1 42 66 72 0 0 0 0 0 1 0 0.08 0.15 $3.2 Mil. $110 Mil. $430 Mil. Min.

3rd Quartile 0.5 1 12 4% 0 1 74 1 1 0.25 $2 Bil.

Max. 1 1 66 89% 1 1 91 1 1 0.75 $200 Bil.

Table 3; Panel B Panel B reports the summary statistics of the regression variables, not including the 44 observations attributable to James Simons.

28

Table 4: Cross Sectional Regressions Table 4 reports the results of the cross sectional regressions used to test H2P against H2A (Director) and H3P against H3A (Institutional Shareholder Principal). The dependent variable in these regressions is the (-30, -2) day run-up period prior to the announcement that a major shareholder has signed the giving pledge. Standard Errors are robust and clustered by year and individual director, of which there are 32 in my sample. T-statistics are reported in brackets beneath the coefficients. Coefficient statistical significance at the 10%, 5%, and 1% level is denoted with *, **, and ***, respectively. The number of observations, R2, and F-Statistic are reported for each model. Statistical significance of the models is denoted by F-Statistics in bold. Table 4; Panel A Panel A reports the results of the cross sectional regressions on the announcement run-up abnormal returns identified with the value weighted index. This panel includes the 44 observations attributable to James Simons.

Director Institutional Shareholder Principal Ln(Net Worth) Percent of Wealth Invested Controlling Shareholder (Dummy) Prior Charitable Foundation (Dummy) Ln(Charitable Shareholder Age) Multiple Segments (Dummy) Foreign Income (Dummy) Leverage Size Constant Observations R-squared F-Statistic

DV: Charitable Pledge Announcement Run-up CAR (-30, -1); Value Weighted Index Model 1 Model 2 Model 3 Model 4 Model 5 Predicted Sign 0.0055 -0.0720*** -0.0860*** H2A(-) H2P(+) [0.7609] [-4.4567] [-3.7297] -0.0514*** -0.0109 -0.0345*** H3A(-) H3P(+) [-7.7481] [-0.6930] [-11.7747] 0.0223*** 0.0278*** 0.0211*** [6.4884] [7.5316] [4.3843] 0.0757** 0.0603* 0.0649* [2.1154] [1.8141] [1.9531] 0.0657*** 0.0321*** 0.0545*** [8.6485] [4.5584] [4.0529] -0.1002*** -0.0931* -0.0969** [-3.1875] [-2.0121] [-2.5833] 0.0446 0.0685 0.0669 [0.4887] [0.6455] [0.8104] 0.0284*** 0.0176 0.0281*** [2.8099] [1.6564] [3.1581] -0.0236* -0.0235*** -0.019 [-1.8123] [-4.3037] [-1.4393] 0.0132* 0.0071 0.0253** [1.9556] [0.6788] [2.4381] 0.0124*** 0.0076*** 0.0120*** [5.5052] [4.0474] [3.8990] -0.0491*** -0.4506 -0.0199*** -0.4705 -0.5145 [-16.0397] [-1.1904] [-28.4057] [-1.0261] [-1.5426] 124 0.04% 0.04

124 21.17% 5.46

124 4.09% 4.64

124 18.05% 6.33

124 22.14% 9.29

29

Table 4: Cross Sectional Regressions (cont.) Table 4; Panel B Panel B reports the results of the cross sectional regressions on the announcement run-up abnormal returns identified with the value weighted index. This panel does not include the 44 observations attributable to James Simons.

Director Institutional Shareholder Principal Ln(Net Worth) Percent of Wealth Invested Controlling Shareholder (Dummy) Prior Charitable Foundation (Dummy) Ln(Charitable Shareholder Age) Multiple Segments (Dummy) Foreign Income (Dummy) Leverage Size Constant Observations R-squared F-Statistic

DV: Charitable Pledge Announcement Run-up CAR (-30, -1); Value Weighted Index Model 1 Model 2 Model 3 Model 4 Model 5 Predicted Sign -0.0259*** -0.0661*** -0.0769*** H2A(-) H2P(+) [-26.9171] [-3.3641] [-3.0758] -0.0271 0.0023 -0.0260*** H3A(-) H3P(+) [-1.1858] [0.2101] [-45.0020] 0.0283*** 0.0339*** 0.0258*** [5.5428] [7.3649] [4.3688] 0.1234*** 0.1211*** 0.1040*** [5.1527] [4.6922] [5.0993] 0.0434** 0.017 0.0429** [2.6757] [1.3704] [2.3343] -0.0977*** -0.0926* -0.0978** [-2.8267] [-2.0213] [-2.5514] 0.0966 0.1133 0.092 [0.9147] [0.9563] [0.9432] 0.0303* 0.0182 0.0299* [1.9279] [1.0592] [1.9311] -0.0175 -0.0201 -0.0132 [-0.7593] [-1.6879] [-0.5257] 0.0365*** 0.0284 0.0361*** [7.1457] [1.3173] [3.4389] 0.0031 -0.0018 0.005 [0.7696] [-0.5702] [1.1517] -0.0176*** -0.4915 -0.0199*** -0.4827 -0.4943 [-5.4900] [-1.2064] [-28.3407] [-0.9924] [-1.3238] 80 1.04% 0.86

80 25.21% 4.20

80 0.98% 0.42

80 20.97% 3.57

80 25.84% 4.46

30