Conflicts of Interest and Monitoring Costs of Institutional Investors: Evidence from Executive Compensation

Transcription

1 Conflicts of Interest and Monitoring Costs of Institutional Investors: Evidence from Executive Compensation by Andres Almazan * University of Texas Department of Finance Austin, TX (512) FAX: (512) Andres.Almazan@mccombs.utexas.edu Jay C. Hartzell University of Texas Department of Finance Austin, TX (512) FAX: (512) Jay.Hartzell@mccombs.utexas.edu and Laura T. Starks University of Texas Department of Finance Austin, TX (512) FAX: (512) lstarks@mail.utexas.edu May 2004 * The authors would like to thank Charlie Hadlock, Steve Kaplan, and seminar participants at the University of North Carolina, College of William and Mary, Purdue University and the University of Texas at Austin for their helpful comments.

2 Conflicts of Interest and Monitoring Costs for Institutional Investors: Evidence from Executive Compensation Abstract Although evidence suggests that institutional investors play a role in monitoring management, not all institutions are equally willing or able to serve this function. We present a stylized model of institutional monitoring and executive compensation to illustrate how potential business relations with the firm and the liquidity of the firm s stock can affect the intensity of institutional monitoring. The model predicts that institutions influence on managers' pay-for-performance sensitivity and level of compensation is reduced when institutions have greater potential for business relations with the firm, but that these effects are attenuated in firms with low liquidity. Our empirical results are consistent with these implications.

3 Conflicts of Interest and Monitoring Costs for Institutional Investors: Evidence from Executive Compensation 1. Introduction Monitoring by institutional investors is a potentially important governance mechanism for corporate management. Theory suggests and empirical evidence confirms that institutions can provide active monitoring capabilities that are difficult for smaller, more passive or less-informed investors. 1 For example, the forced ouster of the New York Stock Exchange CEO, Richard Grasso, over perceived excesses in his compensation package was fueled to a large extent by the vocal outrage of institutional investors (Wall Street Journal, September 17, 2003). However, the intensity of institutions monitoring can be limited by concerns about the liquidity of their portfolios (e.g., Bhide, 1994), fiduciary duties (e.g., Murphy and Van Nuys, 1994), potential business relations with the firm (e.g., Brickley, Lease and Smith, 1988), or the free-rider problem that appears due to the private cost of monitoring (e.g., Shleifer and Vishny, 1986). In this paper, we focus on two of these limiting factors: (1) the (existing or potential) business relations between corporations and their institutional shareholders and (2) the cost of institutional monitoring. Some institutions have a greater potential for business relations with the corporations in which they invest than do others and are consequently more likely to face conflicts between their business and fiduciary interests. Indeed, Brickley, Lease and Smith (1988) find that firms with greater holdings by institutions with potential business relations ( pressure-sensitive shareholders) have more proxy votes in line with management s recommendations, while firms with greater holdings by shareholders without such potential relations ( pressure-insensitive shareholders) experience more 1 Examples include Black (1992), Kaplan and Minton (1994), Kahn and Winton (1998), Del Guercio and Hawkins (1998), Gillan and Starks (2000), Noe (2002), and Almazan and Suarez (2003). 1

4 proxy votes against management s recommendations. 2 These findings lead to a natural question that we consider in this paper: Beyond institutions proxy-voting decisions, do business relations affect the institutions willingness to actively monitor the corporation? One area in which investors can actively monitor is in the assessment and rewarding of managerial performance, i.e., in the design of executive compensation. Examining the relation between institutional monitoring and executive compensation is interesting for several reasons. First, it considers the interrelation between two central governance mechanisms that appear to have become more important in the last decade. 3 Second, unlike many outcomes of the monitoring process, managerial compensation is observable, which allows for not only theoretical modeling, but also empirical testing. Third, an examination of the relation between institutional monitoring and executive compensation can help develop a better understanding of the nature of the agency problem between shareholders and managers. 4 Finally, whether the presence of different types of institutions leads to observable differences in compensation has important implications on the debate over the proper degree of institutional involvement in corporate governance. Formally, we develop a stylized model of a firm owned by two classes of shareholders with different monitoring abilities: Monitoring shareholders (institutions) who can assess managerial performance at a cost, and other shareholders who cannot monitor. In the model, differences in the composition of the shareholder base between the monitoring and nonmonitoring shareholders as well as differences in the incentives of monitoring shareholders lead to observable variation in managerial compensation. Specifically, the model implies that the pay-for-performance sensitivity of managerial 2 See also Pound (1988). Borokhovich, Brunarski, and Parrino (2000) provide additional evidence based on market announcements effects. 3 See, for example, Kaplan and Holmstrom (2001). 4 As we discuss below, the literature has debated whether pay arrangements can be seen as part of the solution to the agency problem or as part of the agency problem itself. See Hall and Murphy (2003) and Bebchuk and Fried (2003) for two opposing views in this matter. 2

5 compensation is increasing in the ownership of monitoring shareholders and decreasing in the costs of monitoring. It also implies that the level of executive compensation is decreasing in the ownership of the monitoring shareholders and increasing in their costs of monitoring. Our model predicts a complementary relation between monitoring by institutions and the degree of pay for performance in the compensation structure. This prediction stems from the fact that we model monitoring shareholders as monitors of compensation rather than of managerial effort or project selection. This modeling approach is in contrast with the simple principal-agent paradigm, where compensation and institutional monitoring substitute each other as a means to provide managerial incentives. 5 We test the empirical implications of our model by examining the relation between executive compensation structures and the concentration of equity ownership of different types of institutions. Following Brickley, Lease and Smith (1988), we consider as stronger monitors those institutions that have less natural potential for business relations with the corporations: investment advisers and investment companies. We refer to these institutions as (potentially) active institutions. Alternatively, we consider as weaker monitors those institutions that are more likely to have current or potential relations with the firms: bank trust departments and insurance companies. We refer to these institutions as (potentially) passive institutions. 6 Our empirical results are broadly consistent with the model s predictions. We find that, in general, the pay-for-performance sensitivity of managerial compensation is increasing in the concentration of active institutions ownership, but is not significantly 5 This role for institutions appears consistent with much of practice, where institutions have largely been focused on improving firms governance rather than dictating corporate strategy (e.g., see TIAA-CREF s policy statement on corporate governance, 6 Throughout the paper for simplicity we use the terms, active and passive to mean potentially active and potentially passive, respectively. In addition, the term active implies institutional investors who monitor through voice rather than large active investors who take over the firm as the term is used, for example, by Bethel, Liebeskind, and Opler (1998) or Denis and Serrano (1996). 3

6 related to the concentration of passive institutions ownership. We also find that the level of executive compensation is decreasing in the concentration of both types of institutional investors ownership, but is decreasing to a greater degree in the concentration of active institutions ownership. Finally, we find that the institutions monitoring is attenuated when the firm s stock price is less liquid, which we interpret as an indication of a high monitoring cost for institutions. These findings complement the Brickley, Lease and Smith (1988) evidence regarding differences in proxy voting across types of institutions, as well as evidence by Hartzell and Starks (2003) that documents systematic influences of institutional investors on managerial compensation. The latter paper, however, does not examine whether the intensity of monitoring differs across different types of institutions, nor does it examine what factors can explain differences in institutional monitoring. Our findings are also related to the results of Parrino, Sias and Starks (2003), who find a relation between CEO turnover and institutional selling (particularly by banks and independent investment advisers) and Pinkowitz (2003), who finds a relation between the success of hostile takeovers and mutual-fund selling. 7 The rest of the paper is organized as follows. The next section describes and analyzes the model. Section 3 presents the data and Section 4 our empirical findings. We conclude the paper in Section 5. 7 David, Kochar and Levitas (1998), Clay (2001), and Hartzell and Starks (2003) find clientele relations between executive compensation and institutional ownership, that is, evidence of greater total institutional ownership in companies with more pay-for-performance sensitivity and lower excess compensation, consistent with institutions preferring to invest in those firms. Further David, Kochar, and Levitas show that for the largest 200 corporations, institutions with less potential for a business relation with the corporations have stronger clientele effects. 4

7 2. Managerial compensation and monitoring shareholders In this section, we develop a model of managerial compensation in the presence of monitoring shareholders. We then discuss the empirical implications and limitations of our model. A. Model We consider an all-equity publicly-traded firm that operates in a risk-neutral economy in which the risk-free rate is normalized to zero. The firm can be owned by two classes of shareholders: institutional investors who can monitor the managers and other investors such as individuals who cannot monitor and as a result do not contribute to the firm s corporate governance. We denote the proportion of the firm owned by a monitoring institutional investor as α (0,1). For simplicity, we consider a setting with symmetric information between managers and shareholders. Events occur in three periods, t=0, 1, and 2. At t=0, the shareholders hire a manager to run the firm. 8 Managers have no wealth, are protected by limited liability, and have a zero reservation level of utility. When the manager is hired, neither the shareholders nor the manager knows the manager s skill level, but they agree that, with equal probability, it can be either H (high-skill manager) or L (lowskill manager). At t=1, the incumbent manager s skill is revealed, which in turn signals the future performance of the firm under his or her management. For simplicity, we denote the firm s cash flow under a high- or low-skill manager as H or L, respectively. After observing his or her revealed skill level, the incumbent manager makes a compensation proposal. At this point if the institutional investor incurs a monitoring cost c, then, with probability δ, a manager of medium skill M is found as a replacement, and with 8 We abstract from a board of directors and assume that the shareholders as a group directly hire a manager to run the firm. Similar results would be obtained by considering a board of directors that can be influenced by institutional investors. 5

8 probability (1-δ) no managerial replacement is found. 9 To decide whether to look for a replacement manager, the institutional investor weighs the incumbent manager s skill, the proposed compensation, the cost of monitoring, and the likelihood of finding a replacement. Finally, at t=2, the firm produces a liquidating cash flow, which depends on the ability of the manager in charge. An important goal of our analysis is to examine how shareholders monitoring incentives affect managerial compensation. That is, we focus on potential shareholder activism rather than on the shareholders ability to sell the shares. Consequently, we abstract from trading considerations and assume that all investors buy into the firm at time 0, and maintain their investment until the firm is liquidated at time 2. Finally, we consider two fundamental, interrelated conflicts between managers and shareholders: (i) the presence of substantial managerial control rents and (ii) the presence of managerial entrenchment. We model managerial control rents by assuming that monetary rewards play a secondary role for managers. Specifically, we assume that managers primary goal is control of the corporation. After retaining control, their secondary goal is to maximize their monetary rewards. We model managerial entrenchment by assuming that when managers are in charge of the corporation at t=1, they propose their own level of monetary compensation up to a limit K. 10 This managerial power can be limited, however, by institutional investors monitoring. We analyze the model under the following parametric assumption: H > M + K c/(αδ)> L, (1) which, as we show below, guarantees that the replacement decision depends in part on the tradeoff between cash flow and managerial compensation. In the analysis, we examine shareholders incentives to monitor the manager, the managerial reaction to 9 As we show, the ability to identify the talent of potential replacements can be essential to limit managerial power in the firm. Implicitly, hence, we are assuming that without such a technology, substituting an unqualified replacement for existing managers is impossible or would produce a great loss to the firm. 10 We discuss the factors that can affect K later in this section. 6

9 such monitoring, and the corresponding effect on the structure of managerial compensation. Because managers anticipate the threat of institutional-investor driven managerial replacements, there is an adjustment in the amount of compensation that the manager achieves in different states of the world. Proposition 1 formally describes the sensitivity of the salary to the signal when a monitoring institutional investor is part of the firm s shareholder base: Proposition 1: In the presence of a monitoring institutional investor with ownership α and monitoring cost c, a high-skill manager (who generates signal H) obtains a salary, w H = K, while a low-skill manager obtains a salary, w L = c/(αδ) (M L). The intuition for Proposition 1 comes from comparing the two constraints faced by a manager when proposing compensation at t=1: first, the salary limit, K, and second, the fact that if the salary demand is excessive, the institutional investor will replace the incumbent if a replacement is found (an event that occurs with probability δ), and will retain the manager if a replacement is not available (and hence will honor the managerial demand with probability (1-δ)). Formally, after a signal s={h,l}, the manager s salary demand w S solves: Max w S subject to: w S K (2) α (s w S ) δαm + (1 δ)α(s w S ) c. (3) Notice that by expression (1), constraint (2) is binding with a high signal (i.e., when s=h) and constraint (3) is binding with a low signal (s=l). Therefore, the following propositions regarding the structure of compensation follow: 7

10 Proposition 2. (Pay-for-performance sensitivity) Define the pay-for-performance sensitivity (PPS) of the manager s compensation to be the difference in managerial compensation as a function of the firm s cash flow (i.e., the shareholders wealth before managerial compensation). Pay-for-performance sensitivity is then given by PPS w H w L = K [c/(αδ) (M L)] = K + (M L) c/(αδ). (4) Accordingly, the pay-for-performance sensitivity of the manager s compensation is (i) increasing in the monitoring institutional investor s ownership in the firm (dpps/dα>0), and (ii) decreasing in the ratio of the cost of monitoring to its probability of success (dpps/d(c/δ)<0). Proposition 3. (Level of compensation) The (expected or average) level of compensation is given by W = ½ (w H + w L ) = ½ [K + c/(αδ) (M L)]. (5) Therefore, the level of compensation (i) decreases with the proportional ownership of the monitoring institutional investor (dw/dα<0), and (ii) increases with the ratio of the cost of monitoring to its probability of success (dw/d(c/δ)>0). B. Discussion In our model, the manager s ability to propose (and set) his or her own compensation is in the spirit of the managerial power hypothesis (e.g., Bertrand and Mullainathan, 1999, 2000; and Bebchuk, Fried, and Walker, 2002) which contends that entrenched managers can set their own compensation (i.e., extract rents) due to their ability to capture the board of directors. 11 If one views the CEO pay process as a 11 The argument that managers have the power to set their own compensation and consequently extract rents has been debated. Garvey and Milbourn (2003) argue that the Bertrand and Mullainathan results (i.e., managers who are paid for luck) can be due to executives fair compensation for bearing risk and that such results do not prove that managers have captured the compensation process. On the other hand, consistent with the rent extraction argument, Campbell and Wasley (1999) and Core Holthausen, and Larcker (1999) provide evidence that managers sometimes design compensation plans at the expense of the shareholders. 8

11 continuum, the managerial power hypothesis lies at one extreme with the managers having almost complete power to set pay. At the other extreme lies the agency (or contracting) model, in which the power to set pay is held by the shareholders who set pay to align the managers incentives with their own. Our model is not at the extreme end of the managerial power hypothesis, rather it is more consistent with Murphy s (2002) view that managers have bargaining power, but that this bargaining power is limited. In our model, there are two limits to managerial power. The first limit comes from the presence of monitoring institutional shareholders: If managers do not produce sufficient cash flows to justify their pay, they can be replaced (with some probability) or pressured to reduce their compensation. The second limit stems from the maximum compensation (i.e., K) that, even in the absence of shareholder pressure, a manager can obtain. Although we do not explicitly model the determinants of K, one can argue that other factors related to governance (e.g., takeover pressure), internal firm organization (e.g., availability of CEO successors) and outrage costs are likely to play a role. 12 While we have simply assumed that K is fixed, i.e., it does not depend on the firm value, similar results can be obtained if the compensation limit increases with firm value (e.g., K H > K L ). In this case, the presence of institutions would increase the sensitivity of compensation from what would be (K H K L ) in their absence to (K H (M+L c/(αδ)) as shown in Proposition 2. Hence, our results hold to the extent that institutional monitoring influences compensation more intensely when a low rather than a high value for the firm is predicted, i.e., when K L > M+L c/(αδ). Propositions 2 and 3 yield results consistent with the previous theoretical work of Shleifer and Vishny (1986), Huddart (1993) and Maug (1998a), who argue that large 12 For example, some compensation arrangements could cause embarrassment to the board of directors, could hurt managerial reputations, or could simply cause outsiders to develop perceptions that managers are expropriating rents. 9

12 shareholders can be important in the mitigation of agency problems. 13 In addition, these propositions yield testable hypotheses regarding the relation between managerial compensation and institutional monitoring. These hypotheses center around two model inputs: the amount of the firm owned by the monitoring institution, α, and the ratio of the cost of monitoring to the probability of monitoring successfully (i.e., finding a replacement manager), c/δ. In deriving a proxy for α for the empirical implementation of our model, we encounter two limitations. The first limitation derives from the assumption that monitoring comes from a single institution rather than from multiple institutions. Because of this assumption, the analysis results in the single institution s proportional ownership (i.e., α) as being the relevant independent variable for institutional monitoring. However, in the presence of multiple institutional owners, the proportional ownership does not reflect the incentives to monitor as each of the institutions may have a very small ownership interest in the firm, thus, leading to free-rider problems as pointed out by Shleifer and Vishny (1986). To address this limitation we employ the concentration of the institutional investors ownership as the relevant measure of α. More specifically, to test the extent to which potential conflicts of interest affect institutions monitoring, we separate institutions into two groups according to their potential business relations with the firm and examine whether ownership concentration in each group affects managerial pay patterns. Although the concentration of institutional ownership does not completely eliminate the free rider problem, it does capture the ownership by institutions with greater incentives to monitor. This construction is also consistent with Black s (1992) 13 Empirical evidence suggests that large blockholders have provided successful monitoring functions. See, for example, Agrawal and Mandelker (1990), Kaplan and Minton (1994), Kang and Shivdasani (1996) or Bethel, Liebeskind, and Opler (1998). Evidence on activist public pension funds has been more mixed. See, for example, Karpoff, Malatesta, and Walkling (1995), Carleton, Nelson and Weisbach (1998), Gillan and Starks (1998), or Del Guercio and Hawkins (1998). 10

13 contention that institutional investors have more influence when they have allies in the form of other institutional investors with large holdings. The second limitation to deriving a proxy for α arises because the total proportional ownership of the institutional investors may encounter an endogeneity problem in that these investors may have preferences for firms that have better executive compensation structures. Using the concentration of institutional ownership also helps ameliorate the endogeneity problem. Furthermore, as we discuss below, to ensure that endogeneity between executive compensation and the concentration of institutional ownership is not driving our results, we also perform tests that examine the relation between long-run changes in institutional ownership concentration and subsequent long-run changes in pay-for-performance sensitivity. We also need to derive a proxy for our second model input (c/δ). We need to allow the relative costs and benefits of monitoring (c/δ) to differ both across institutions and the firms in which they invest. In the context of the model, this separation corresponds to segmenting the cost of monitoring, c, into a shareholder-specific determinant of the costs of monitoring that arises due to differences in potential business relations with firms, plus a component that differs due to firm-specific characteristics. To test for variation in the cost of monitoring across institutions, we use the respective ownership of different classes of institutions. Although direct measures of firm-specific monitoring costs are hard to obtain, the difficulty in monitoring should be inversely related to the amount of information generated about the firm and reflected in its stock price, an aspect that can be captured by the stock s liquidity. Indeed, Holmstrom and Tirole (1993) and Garvey and Swan (2002) argue that more liquid firms have greater information flow (e.g., due to more informative prices and greater analyst following). 14 In our context, the greater degree of information reduces the difficulty of discerning how 14 Brennan and Subrahmanyam (1995) and Roulstone (2002) find that analyst following and liquidity are positively associated. 11

14 management is performing and identifying potential replacements. Consistent with these arguments, we use a measure of the stock s liquidity the inverse of its turnover as a proxy for the ratio of the cost of monitoring to the probability of successfully finding a replacement. We further differentiate the cost of monitoring across the two types of institutional investors by interacting this proxy with each type s concentration of ownership. 15 Although one could imagine alternative proxies for the cost and likelihood of successfully replacing a manager, the immediate alternatives present empirical problems. For example, one could use the homogeneity of a firm s industry as a proxy for the costs and benefits of CEO replacement (Parrino, 1997). But, by its nature, this is an industry-level variable, and one could not separate its effect from any other industryspecific effect (and industry has substantial explanatory power in managerial compensation). Variables that capture managerial entrenchment are additional logical candidates, but such variables are also prime candidates for endogeneity problems, making it difficult to interpret results and to obtain consistent parameter estimates. Thus, we use a liquidity measure, which captures across-firm variation, and is less likely to suffer from endogeneity problems. 3. Data Our initial sample consists of the 1,914 firms included on the Standard & Poor's ExecuComp database over the 1992 through 1997 time period. The database covers roughly 1,500 firms per year, including the 500 firms in the S&P 500 Index, the 400 firms in the S&P Midcap Index, and the 600 firms in the S&P Smallcap Index. For up to five 15 Garvey and Swan (2002) also argue that the more informative stock prices that result from increased liquidity are better benchmarks on which to base executive compensation. They find a direct relation between the use of incentive compensation and liquid stock prices. In contrast to their work examining this direct relation between compensation and stock liquidity, we focus on the interaction term between the concentration of institutional investor ownership and the cost of monitoring. 12

15 top executives from each firm, we retrieve details of their compensation package, including salary, bonus, long-term incentive plan payouts, stock and option grants and other compensation reported by the firms in their proxy statements. 16 In order to identify the relation between institutional investor monitoring and executive compensation, we restrict the sample to end before During the early and mid-1990s, the idea of tying executive compensation to firm performance through stock or option grants was generally viewed positively. (See, for example, Financial Times, 1995.) By contrast, during the last years of the 1990s, certain types of compensation (particularly option compensation) became increasingly controversial and even viewed negatively by some institutional investors. (See Lublin and Scism, 1999.) A. Measures of compensation We employ several different measures of the structure of managerial compensation. We use the level of pay, where pay is alternatively defined as salary or total direct compensation (i.e., the sum of salary, bonus, option and stock grants, longterm incentive plan payouts, and other compensation). In addition, we use three measures of the pay-for-performance sensitivity of managerial compensation. The first measure focuses solely on the options granted to managers: the sensitivity of option grants to changes in stock price (Yermack, 1995). The second measure focuses solely on cash compensation: the sensitivity of salary plus bonus to changes in shareholder wealth (Jensen and Murphy, 1990). Finally, the third measure combines these sources of compensation along with other types of compensation: the sensitivity of total direct compensation to changes in shareholder wealth (Jensen and Murphy, 1990) Some firms list less than five top executives in their proxy statement. 17 Substantial differences across measures of compensation suggest the use of alternative measures in the tests. For instance, the correlation between total direct compensation and salary plus bonus is

16 To calculate each executive s option-grant sensitivity, we use the methodology suggested by Yermack (1995). First, we calculate the delta of every option grant, C/ P (where C is the value of the call option and P is the price of the stock), by using the Black-Scholes model modified for dividends. (We derive dividend yields and volatilities from the Center for Research in Securities Prices (CRSP) data.) We then multiply the delta of the option grant by the number of options granted and divide by the number of shares outstanding at the beginning of the year. This number is the sensitivity of the option grant per dollar change in share value. Multiplying it by 1,000 gives the familiar dollar change in managerial wealth per $1,000 change in shareholder wealth. For years in which executives receive multiple option grants, the sensitivities are aggregated over each year for each manager. We do not include changes in the value of the managers previous stockholdings in our measure of compensation. Although managers may alter their portfolios and risk exposures in response to the composition of their pay package (Ofek and Yermack, 2000, or Bettis, Bizjak and Lemmon, 2001), we focus on the compensation components over which the board of directors has direct control and consequently the components that activist institutional shareholders could influence. This focus is consistent with the evidence in Core and Guay (1998) which shows that firms use the flow rather than the stock of equity incentives to reward past performance and to re-optimize incentives for future performance. B. Measures of institutional monitoring For every firm on the ExecuComp database, we obtain institutional equity holdings for each year between December 1991 and December 1996 from the CDA Spectrum database. 18 CDA Spectrum derives these holdings from institutional investors 18 Because we employ lagged institutional ownership variables in our tests, our institutional data precedes the compensation data by one year. 14

17 13-f filings. (Institutional investors with more than $100 million in equities must report their equity ownership to the SEC in quarterly 13-f filings.) Institutions file their holdings as the aggregate for their firm, regardless of how many individual fund portfolios they have. CDA Spectrum classifies institutional investors into five types: investment companies (mutual funds and closed-end funds), independent investment advisers (principally pension fund advisers), bank trust departments, insurance companies and others (miscellaneous institutions such as endowment funds or public pension funds). To identify the intensity of expected monitoring activity with an institutional type s expected ability to withstand pressure from the companies in which they invest, we follow Brickley, Lease and Smith (1988) and refine the CDA Spectrum institutional classification into two groups according to their likelihood of monitoring. We classify investment companies and independent investment advisers as (potentially) active and bank trust departments and insurance companies as (potentially) passive. The other category is a mix of endowment funds, self-managed corporate pension funds and a few public pension funds. 19 Because this group has a mix of potentially active and potentially passive institutions, we take the conservative approach and categorize them as passive. However, this category is a small proportion of the total institutional ownership (less than 5% for our sample period), thus, changing the group to active does not change our qualitative results. After dividing the institutions into the active and passive categories, we calculate the concentration of each respective group s ownership in the firm as the percentages of total institutional ownership held by any of the firm s five largest institutional owners that come from each group. That is, for a given firm, the concentration of potentially active institutions is the percentage of total institutional ownership held by active institutions that are among the five largest institutional investors of that firm. To capture the cross- 19 Although public pension funds are not required to make 13-f filings, some of the public funds choose to do so voluntarily. 15

18 sectional variation in the ratio of the cost of monitoring to the probability of monitoring successfully, we interact each of the concentration variables with the inverse of the firm s stock turnover. C. Control variables Studies of executive compensation as well as studies of institutional investment have found a number of systematic differences associated with certain firm characteristics. For example, executive compensation is related to firm size (Baker, Jensen and Murphy, 1988), firm performance (Smith and Watts, 1992) and firm growth opportunities (Smith and Watts, 1992, and Harvey and Shrieves, 2001). Institutional investment is related to firm size (Sias and Starks, 1997, and Gompers and Metrick, 2001) and firm performance (Nofsinger and Sias, 1999). Given these systematic relations, we include several control variables in our regressions. Our reported measure of firm size is market capitalization, but we also employed the alternative measures of net sales and total assets in our tests with no change in the qualitative results. We use Tobin s q ratio to control for firm growth opportunities. 20 To control for pay similarities within industries, we use industry dummy variables (at the two-digit SIC level). These variables also control for preferences institutional investors may have for particular industries. Year dummy variables allow both pay-for-performance sensitivity and changes in pay to vary year-by-year. Finally, we control for differences between the CEO and other top executives in two ways. First, we use data on all five executives in the regression and employ a CEO dummy variable, equal to one if the executive is the CEO and zero otherwise. In particular, this variable controls for differences in the effects 20 Using Compustat as our data source, we calculate Tobin s q as (the market value of equity less book value of equity plus book value of assets) divided by book value of assets. 16

19 of pay-for-performance sensitivity for CEOs versus the other top executives of the firm. Second, we run the regressions with the sample restricted to CEOs only. 21 We obtain the data on firm characteristics from CRSP and Compustat. To be included in the final sample, a firm must have data available from all four sources (Execucomp, CDA Spectrum, CRSP and Compustat) for a given year. This requirement results in a final sample of 36,352 firm-executive-year observations, spread over 1,836 firms (out of the 1,914 firms in our Execucomp sample). Table 1 provides the descriptive statistics for our sample with the managerial compensation variables in Panel A, the institutional investor ownership variables in Panel B, and the firm characteristic variables in Panel C. As shown in Panel A, the top five executives in the sample firms receive an average annual salary of about $301,000. The addition of option and stock grants, long-term incentive plan payouts, and other types of direct compensation brings their average annual total direct compensation to almost $1.25 million. The average annual change in compensation over our sample period is about $63,000 in salary plus bonus and about $201,000 in total direct compensation. Finally, the average option-grant sensitivity implies an expected change in value of almost $1 for each $1,000 change in shareholder wealth. Panel B provides summary statistics for the institutional ownership in the sample firms. Institutional investors hold an average of 52% of the outstanding equity in the sample firms. Statistics for our concentration measures show that large, potentially active institutional investors (independent investment advisers and mutual funds) hold about 31% of all institutionally-owned shares, and large, potentially passive institutional investors (banks, insurance companies and other institutions) hold about 13% of these shares. 21 In some cases, ExecuComp does not designate which of the executives is the CEO. In this case, we assume the executive with the highest base salary is the CEO. 17

20 Panel C of Table 1 provides summary statistics for three firm characteristics that are used in our empirical tests. Consistent with the high stock prices over our sample period, the average Tobin s Q ratio is The average market capitalization is $2.9 billion and the average share turnover for these firms is Empirical results In this section, we present the results from empirical testing the implications of the model developed in Section I. We first present the results from testing the hypotheses derived from Proposition 2 regarding the pay-for-performance sensitivity of managerial compensation. We then present the results from testing the hypotheses from Proposition 3 regarding the level of managerial compensation. A. Pay-for-performance sensitivity Proposition 2 implies that the pay-for-performance sensitivity of the manager s compensation is increasing in the institutional investor ownership especially for active institutions and decreasing in the ratio of the cost of monitoring to its likelihood of success. In Section A.1, we examine these implications by measuring the pay-forperformance sensitivity implicit in the stock option grants in isolation (Yermack, 1995). Then, in Section A.2, we examine the implications by measuring pay-for-performance sensitivity through a more comprehensive, ex post measure based on the effects on total compensation of changes in shareholder wealth (Jensen and Murphy, 1990). A.1. Option-grant sensitivity The interest in focusing on stock option grants in isolation is twofold. First, over our sample period, options became not only an increasingly important component of executive pay (see Murphy, 1998), but also an instrument of compensation that was 18

21 favorably viewed by institutional investors. Second, in contrast to other measures of pay-performance sensitivity, option-grant sensitivity is an ex ante sensitivity measure. Arguably, option sensitivity better captures the intended sensitivity of compensation to shareholder value rather than the realized sensitivity, which could be obscured by the occurrence of random shocks. Because many firms do not pay option grants or do not pay them every year, we use a Tobit regression to examine the relation between firms option-grant sensitivities and active versus passive institutional investor ownership concentration: (value of options granted per $1000 in shareholder wealth) it = β 1 (active institutional concentration it-1 ) + β 2 (passive institutional concentration it-1 ) + β 3 [(active institutional concentration it-1 )*(cost of monitoring)] + β 4 [(passive institutional concentration it-1 ) *(cost of monitoring)] + β 5 (shareholder wealth) it + β 6 (shareholder wealth) it-1 + Σβ k (other control variables it ) + Σβ y year dummy variables t, where the concentration of active (passive) institutional ownership is measured by the percentage of institutionally-owned shares held by the five largest institutional holders that are active (passive). The firm-specific cost of monitoring proxy (the inverse of the stock turnover) is interacted separately with the concentrations of the two types of investors, which allows us to examine how monitoring is reduced by these costs. The control variables are changes in shareholder wealth in the current and previous years, total institutional ownership, Tobin s q ratio, market capitalization, a dummy variable equal to one if the executive is a CEO, industry-level dummy variables equal to one for the two-digit SIC in which the firm operates, and year dummies equal to one if the observation is for the given year. We use the concentration of institutional ownership for the prior period because of our interest in the institutions influence. Tobin s q ratio and market capitalization are similarly lagged. For this regression as well as subsequent regressions in which the unit of observation is an executive-year, we calculate standard errors that are robust to heteroskedasticity and within-firm correlation (i.e., clustering of 19

22 errors by firm). This adjusts for a potential lack of independence across executives or time within each firm. The results of the Tobit regressions are provided in Table 2. The first two columns report the results of the regressions that include all five of the top executives in the firm. Columns three and four report the results of the regressions that restrict the sample to the CEOs. In all four specifications in Table 2, pay-for-performance sensitivity is significantly related to active institutional ownership concentration but not to passive institutional ownership concentration. This difference in sensitivity is statistically significant as shown by a Wald F-statistic for the equality of the regression statistics. In addition, the interaction terms between monitoring cost and institutional ownership concentration show that the effect of active ownership in compensation is substantially reduced in the cases with lower stock liquidity where monitoring is presumably more expensive and/or less likely to succeed. These results are consistent with Proposition 2, which indicates that pay-for-performance sensitivity is increasing in the ownership concentration of the active institutions, but decreasing in their relative monitoring costs. In addition, the differences between active and passive institutions is consistent with our initial assumption that an institution s intensity of monitoring activity is related to its current or potential business relations with the companies in which they invest. These results appear to have substantial economic significance. For instance, according to the regression in column (2) of Table 2, a one standard deviation increase in the concentration of active institutional investor ownership would be associated with a $0.39 increase in the top executives option grant sensitivity. Relative to the mean option-grant sensitivity of $0.99 and the median of $0.38, this effect is quite large. For the CEO in isolation, the economic significance is even greater: a one standard deviation increase in active investor concentration would be associated with an $0.81 increase in stock option sensitivity. This latter result is due to our finding that the option 20

23 grants of CEOs have more pay-for-performance sensitivity than those of other executives, which would be expected given the increased responsibility and discretionary ability they generally have. The coefficients on the control variables confirm previous findings in the literature. For instance, we find that the pay-for-performance sensitivity of option grants is increasing in total institutional ownership supporting the importance of clientele effects. In other words, institutional investors, as a group, are attracted to firms with greater pay-for-performance sensitivity of their option grants (see, for example, Clay, 2001, or Hartzell and Starks, 2003). 22 A.2. Long-run changes Table 2 provides evidence of differences between executive compensation structures and the type of institutional investor with concentrated holdings in the firm. While these differences are suggestive of the influence of institutions on compensation, examining lead-lag effects among ownership and compensation provide further confirmatory evidence of these effects. Specifically, we would expect that if the institutional ownership concentration of the active investors increases in a firm and these institutions provide monitoring of managerial compensation, there would be subsequent changes in the structure of the executive compensation. To examine this, we divide our sample into two subperiods, 1991 to 1994 and 1995 to For each firm, we calculate the change in active and passive ownership concentrations over the early subperiod, i.e., 1994 concentration less 1991 concentration. We next use the following procedure to estimate the change in optiongrant sensitivity across the two subperiods. First, we calculate the average sensitivity over the base period by summing the option deltas for the top executives of each firm in each year and taking an average (by firm). We then calculate the average option-grant 22 This is in contrast with our analysis, which it is performed using institutional ownership concentration variables that are unlikely to be affected by such clientele effects. 21

24 sensitivity over the later period in an analogous manner and measure the change in sensitivity by subtracting the earlier average from the later average. We regress the change in option-grant sensitivity for the versus the period on the following independent variables: the changes in institutional concentration over the period, interaction terms between the costs of monitoring and the changes in institutional concentration, and a set of control variables (Tobin s q, firm size, industry, and change in total institutional ownership). Table 3 provides the results of the regressions with the interaction terms included only in the regression shown in Column (2). Consistent with Table 2, for active institutions we find (i) a strong positive relation between the long-run change in option-grant sensitivity and the early change in ownership concentration and (ii) a significantly negative relation between the long-run change in option-grant sensitivity and the cost/benefit of monitoring-ownership concentration interaction term. For passive institutions, we find (i) a significantly positive relation between the long-run change in option-grant sensitivity and the early change in passive institutional concentration, and (ii) no significant relation between the long-run change and the relative cost of monitoring. These results suggest that while monitoring by all kinds of institutions has long-run effects on compensation, active institutions are more sensitive to the monitoring costs. This is consistent with institutions performing different types of monitoring. For instance, if monitoring by active institutions is affected more by the information about the firm generated in the market and incorporated into the stock price, then the effect of an increase in monitoring cost (i.e., a reduction in liquidity) will be more apparent in the monitoring intensity of active institutions A potential endogeneity concern on the positive relationship between institutional ownership concentration and pay-for-performance sensitivity can be raised (i.e., institutions choose to own firms which compensate managers more efficiently rather than influence firm s compensation practice). As Table 3 shows, changes in executive compensation are clearly preceded by changes in institutional ownership, a finding that renders this reverse causality interpretation implausible. In fact, if we run the reverse regression allowing changes in compensation to predict changes in institutional concentration of ownership, we find no significant relation. 22

25 A.3. Pay-for-performance sensitivity of cash and total direct compensation As an alternative measure of the executive pay-for-performance sensitivity, we consider the sensitivity reflected in the executive s cash or total direct compensation (Jensen and Murphy, 1990). Our tests on the relation between the categories of institutional investor concentration and this sensitivity are based on the following equation: (manager s compensation) it = β 1 (shareholder wealth) it-1 + (shareholder wealth) it *[ β 2 (active institutional concentration it-1 ) + β 3 (passive institutional concentration it-1 ) + β 4 [(active institutional concentration it- 1)*(cost of monitoring)] + β 5 [(passive institutional concentration it-1 ) *(cost of monitoring)] + Σβ k (control variables it )] + Σβ y year dummy variables t, where managerial compensation is measured by either salary plus bonus or by total direct compensation (the sum of salary, bonus, option and stock grants, long-term incentive plan payouts, and other compensation) and the control variables are total institutional ownership, Tobin s q ratio, market capitalization, a dummy variable equal to one for each firm's CEO, industry-level dummy variables equal to one for the two-digit SIC in which the firm operates, and year dummies equal to one if the observation was for the given year. 24 Table 4 shows the results from this regression for all executives in Panel A and for CEOs only in Panel B. The first two columns of each panel show the relation using salary plus bonus and the third and fourth columns show the relation using total direct compensation. Consistent with Table 2, we find substantial differences in the relation between the pay-for-performance sensitivity of the executives compensation and institutional concentration across the different types of institutions. The active institutional concentration has a significant positive relation, while the passive institutional concentration has no significant relation. Further, a Wald test shows that the 24 The year dummies enter the regression twice: once as intercept terms, and once interacted with the change in shareholder wealth. Thus, they control for time-specific variation in both changes in pay, and changes in pay-for-performance sensitivity. Hall and Liebman (1998) show that pay-for-performance sensitivity has increased since the 1980s. 23