Social Comparison and Reciprocity in Director Compensation

Abstract

In this article, we develop theory regarding one set of mechanisms through which increases in the compensation of directors are transmitted throughout the director labor market. In a longitudinal study using director compensation data from 1996 to 2005, we test hypotheses about how directors’ use of social comparison processes, and reciprocity between CEOs and the board, drive up the compensation level for boards of directors. Specifically, we argue and find that directors’ home firms and interlocked boards serve as salient comparison groups for board members.

Keywords

boards of directors compensation bonuses benefits top management teams upper echelon

Director compensation has risen dramatically in recent years. For example, a recent study demonstrates that during the period from 2001 to 2004, director pay rose by over 50%, compared with an increase of 24% in CEO compensation (Linck, Netter, & Yang, 2009). This rise in director pay has continued in recent years and has been attributed to a combination of factors in the director labor market including new legislation (e.g., Sarbanes–Oxley) and more stringent listing requirements, which have increased the risks and responsibilities associated with board service and reduced the supply of qualified directors (Linck et al., 2009). During this same period, an increasing amount of academic research has focused on the role that directors play in governing companies, including explanations of how boards influence executive pay (Barnea & Guedj, 2006; Brick, Palmon, & Wald, 2006; Conyon, Peck, & Sadler, 2009; Finkelstein, Hambrick, & Cannella, 2009). Yet despite the sharp increase in director pay, and the increase in research on the role of directors, there has been surprisingly little research concerning director compensation in general, and even less regarding the specific mechanisms through which director pay increases are transmitted throughout the director labor market.

Director compensation is an interesting and important context for a number of reasons. First, given the increased role of the board of directors in modern corporations, a better understanding of the factors that influence something as fundamental as their pay should be of interest to corporate governance and organizational scholars. Researchers have called the issue of director motivation one of the great unanswered questions of corporate governance research (Hambrick, Werder, & Zajac, 2008). By exploring the process by which director compensation increases are transmitted through the market, we hope to shed some light on what motivates directors to pursue compensation increases. Furthermore, given the substantial increase in director compensation in recent years (Linck et al., 2009), director pay now represents a nontrivial amount of money for firms. The average compensation level of corporate directors at large firms is enough for corporate directors to be classified in the top 3% of all wage earners in the country (Internal Revenue Service, 2011), from their board compensation alone. Although board compensation levels are still well below those of top executives, the level is high enough that employees, investors, and other stakeholders may start to scrutinize their pay. Finally, the compensation of directors is also important because it carries substantial symbolic meaning to interested parties regarding the board’s commitment to the firm as well as its commitment to the prudent use of the firm’s financial resources.

The limited research that has explored issues of director compensation tends to draw on economic or agency perspectives focusing on how director pay varies with performance (Yermack, 2004) or how it may influence the board’s incentive to monitor and promote shareholder friendly policies (Kosnik, 1987, 1990). This stream of research points out that the type of director compensation can be an important determinant of incentive alignment between directors and shareholders. However, one area that has been left largely unexplored is the process by which director compensation is determined and how social mechanisms play a role in that process. Because boards largely determine their own compensation, it is unclear how they decide what level of compensation is appropriate and how market forces that appear to be pushing director pay higher actually come to be translated into changes in director pay.

In this study, we seek to address this limitation of the corporate governance literature by introducing a social perspective on director compensation. We draw from the literature on social comparison processes and equity theory to develop our model of how director compensation is influenced by salient comparison groups such as directors’ home firms (i.e., the firms where outside directors work as full-time employees) and other interlocked boards (i.e., other boards on which outside directors sit). We argue that social comparison processes and reciprocity between the CEO and the board can help to explain recent increases in director compensation. Our major research questions are as follows: How do social comparison and reciprocity mechanisms affect director compensation? Do these social determinants of director compensation help to explain why director compensation has increased substantially over time?

By helping to better understand how the process of social comparison and the norm of reciprocity are likely to result in increased director compensation, our theory and results make a number of significant contributions to the corporate governance literature. First, this study develops theory regarding the mechanisms through which pay increases are transmitted through the director labor market. While recent research has explained how a combination of market forces have driven up the demand and reduced the supply of directors, our study begins to show how these forces are translated through an informal process of social comparison and reciprocity.

Second, while most prior work on director compensation uses economic perspectives such as agency theory, this study makes a notable contribution to the corporate governance literature by exploring how social forces may work to shape director compensation. While some studies of executive compensation have examined how social comparison processes and power dynamics affect executive pay (O’Reilly, Main, & Crystal, 1988; Porac, Wade, & Pollock, 1999), there has been little to no extant research on how social forces might affect director compensation. We extend the governance literature by applying a social theoretical lens based on the ideas of social comparison, perceived inequity, and reciprocity to the arena of director compensation. Specifically, we make a contribution by developing theory regarding the specific peer groups to which directors are likely to look for guidance when setting director compensation. Although extant research on executive compensation has suggested a number of referent groups that firms look to when setting executive compensation, this research has not considered which peer groups the board may look to when setting their own compensation. In addition, by developing theory regarding how reciprocity dynamics may help to explain the relationship between changes in CEO compensation and changes in board compensation, this study contributes to our understanding of how social dynamics between the CEO and the board may affect director compensation.

Furthermore, our study makes a contribution to the literatures on social comparison and compensation through our theorizing that directors may use themselves as important social referents. Almost all of the literature on social comparisons and equity theory attempts to determine which others an individual or group is likely to look to when making comparisons and determining whether their pay is equitable. Given that a selected individual or group becomes the basis for judging the equity of compensation, the choice of a specific referent individual or group is crucial. Scholars have struggled to predict the referent others used to judge equity of outcomes, as individuals sometimes select counterintuitive comparisons (Scholl, Cooper, & McKenna, 1987). We suggest that, for some contexts, perhaps the most valid choice of a comparison target for a given individual is that same individual in a different role. In our study, we suggest that directors may use themselves, acting in different role situations at different firms, as a valid social referent when making comparisons regarding pay and determinations of equity.

In addition to contributing to the governance literature, our exploration of how pay increases are transmitted across firms also makes a contribution to the literature on the diffusion of practices by showing how the social psychological processes related to the transfer of director compensation result in asymmetric information transfer through network ties. While prior work on the diffusion of information via network and interlock ties has generally assumed that information transfer is symmetric across all ties that are similar in nature, we argue and find that structurally similar ties may have dissimilar effects.

In sum, our study aims to bring a new perspective to a topic that has been relatively understudied in the corporate governance literature. In the following sections, we explain how social comparison processes and reciprocity are likely to drive increases in director compensation. We test our predictions with a sample of large firms from 1998 to 2005. We conclude by discussing the results and their implications for theory and practice.

Theory and Hypotheses

Despite the extensive literature on corporate governance in general and CEO compensation in particular, there have been relatively few studies that directly examine the compensation of directors. This is surprising given the fact that the board is the highest decision making body at the firm with power to hire and fire management, set executive compensation, and ultimately guide the strategic direction of the firm. Directors at large firms can often make large sums of money for what appears from the outside to be relatively little work. For example, reports claim that directors at Fortune 500 firms earned a median income of $234,000 in 2011 for as little as 4.3 hours of board-related work per week (Strauss, 2011). Not only is director pay more materially relevant than ever before, it also carries symbolic meaning. Firm policies and actions can often lead external constituents to make character attributions about the firm, which ultimately affect important outcomes such as firm reputation (Love & Kraatz, 2009).

Recently, there has been some attention given to director compensation in the finance literature, but almost none in the management literature (Brick et al., 2006; Yermack, 2004). This lack of research attention is curious given the central nature of the board of directors in resolving agency problems in current theories of corporate governance. Perhaps this oversight is a result of the fact that at first glance, director compensation may seem quite similar to the arena of executive compensation. If director compensation and executive compensation are overly similar, then despite the fact that director compensation has received little research attention, it may not be a useful area of research given that theory and findings related to director compensation will simply mirror those studied in executive compensation. However, despite the surface similarities, there are a number of significant differences between director and executive compensation that make the two settings sufficiently distinct to make director compensation an interesting area to examine.

Distinctive Features of Director Pay

There are a number of significant differences between director compensation and executive compensation. First, director compensation is unique in that directors are one of the few groups in public corporations that set their own pay. So, although agency perspectives view the board as the primary monitor of the agents (i.e., self-interested managers), the directors themselves can be viewed as primary actors who are balancing the interests of multiple constituents and therefore have the potential to act in ways that may not be in shareholder’s best interests (Lan & Heracleous, 2010). Also, as we argue in the following, the fact that boards set their own pay may cause boards to use and evaluate information from social referents in a way that is distinct from how they use and evaluate information during the executive compensation process.

Second, the uncertainty around the proper amount and structure of director compensation is likely even greater than that surrounding executive compensation. The issue of executive compensation has received considerable attention both by researchers and the popular press. In contrast, director compensation has received relatively little attention, and so there is little discussion of what is the appropriate amount of director pay and/or how that pay should be structured. In addition, the fact that there is little conclusive evidence about how the board may influence firm performance (Dalton, Daily, Ellstrand, & Johnson, 1998; Dalton, Daily, Johnson, & Ellstrand, 1999) makes this decision even more problematic.

Third, director compensation differs from executive compensation in the level and intensity of scrutiny that it receives from stakeholders. While justifications for CEO pay in firm proxy statements are extensive and draw on multiple lines of reasoning (Porac et al., 1999; Wade, Porac, & Pollock, 1997; Westphal & Zajac, 1994; Zajac & Westphal, 1995), explanations and justifications for director pay are either absent or so brief as to be useless to outside constituents. This lack of external scrutiny is puzzling given that research has demonstrated how members of the board are often suggested and appointed by the CEO and other board members. This director selection process is rife with actions of ingratiation and flattery (Westphal & Stern, 2006, 2007), and these dynamics create the opportunity for reciprocity in subsequent compensation decisions, especially given the fact that board compensation is generally somewhat inconspicuous.

Finally, director compensation is different from executive compensation in that the compensation package is designed for a team of individuals and is quite uniform across the board, unlike top executives, whose pay can differ dramatically (Siegel & Hambrick, 2005; Wade, O’Reilly, & Pollock, 2006). The board has characteristics that suggest it functions as a team. Specifically, it is a collection of individuals who are interdependent in terms of their tasks, who share responsibility for collective outcomes, and who see themselves and are seen by others as a social entity (Cohen & Bailey, 1997; Guzzo & Dickson, 1996; Hackman, 1987). Directors are interdependent in making governance decisions and are typically viewed as a collective entity. Thus, directors are generally compensated with a uniform annual retainer that represents the base compensation for serving on the board and can be thought of as analogous to CEO salary. Additional compensation comes in the form of stock (both grants and options), fees for attending meetings, and additional fees for serving on committees (such as the compensation or audit committee). Generally there are only small differences between the annual pay of directors, which tend to come from service on different committees, and these differences are relatively small. Consequently, director pay is generally treated as uniform across the board (Kosnik, 1990). Given the uniform nature of director pay and the fact that the board is a team, the board is likely to be especially sensitive to perceived inequity in compensation (Colquitt, Zapata-Phelan, & Roberson, 2005).

Economic Explanations of Director Pay

Given the differences between director and executive compensation, it is important to better understand the process of how director compensation is set. Early agency theoretic views of director compensation typically asserted that high levels of director compensation should be viewed negatively because it was thought that excess compensation would reduce objectivity toward executives and therefore restrict board control (Boyd, 1994; Fama & Jensen, 1983). However, director ownership of the firm’s stock was generally viewed as a positive contributor toward board control because it was assumed that directors who owned more stock would have more power or more of an incentive to control management (Boyd, 1994). Some of the earliest and most prominent work on director compensation in the management literature attempted to test this view that director stock ownership could motivate outside directors to more effectively monitor top management by reducing the incidence of specific practices, such as the payment of greenmail (i.e., the practice of forcing a potential target firm to buy back shares from an owner at a premium to avoid a takeover), that were widely considered to be antishareholder (Kosnik, 1987, 1990). Although empirical results were mixed, there was evidence that in some circumstances, director compensation can motivate directors to resist antishareholder practices (Kosnik, 1990).

Other studies of director compensation focused on characteristics of the firm or the activities of the board. For example, it was found that the number of board meetings and the size of the firm influence the level of director compensation, but pay was not related to performance (Hempel & Fay, 1994). Others found evidence that the proportion of outside directors on the board predicted the likelihood of adopting incentive plans for directors (Vafeas, 1999). Observers of boards and directors themselves have recently suggested that boards deserve higher levels of compensation because of increased workloads (Lublin, 2008). More recently, there have been a few studies in the finance literature that have examined director compensation. These studies have shown that the structure of director pay is similar to the structure of executive pay (Yermack, 2004), that the level is increasing (Farrell, Friesen, & Hersch, 2008), and that firms are increasing the level of stock owned by directors through required purchases, increased initial stock grants, or fixed-value equity grants whereby the overall value of the stock award is guaranteed (Becher, Campbell, & Frye, 2005; Farrell et al., 2008; Yermack, 2004). Others have looked at how legislation such as Sarbanes–Oxley has increased director pay by increasing demand for director services while making board service less desirable for many potential directors (Linck et al., 2009). What is consistent about these recent studies in the finance literature is their reliance on economic explanations for director compensation. Our focus is to extend this research by developing theory regarding how social forces, specifically the process of social comparison, influence director compensation levels and transmit increases in pay through the director labor market.

Social Influences of Director Compensation

According to social comparison theory, individuals are constantly driven to evaluate their own opinions and abilities (Festinger, 1954). In the absence of objective criteria, people look to similar others with whom they can compare and evaluate themselves (Festinger, 1954; Goodman, 1974). Social comparison is especially likely to occur in situations where there is uncertainty regarding appropriate behavior. For example, as discussed above, the appropriate level and mix of executive and director compensation can be viewed as an ambiguous situation because there are multiple factors that potentially affect firm performance and it is unclear how much variance in firm performance can be attributed to the board (Dalton et al., 1998, 1999). Because it is difficult to quantify the impact of the board, it becomes more difficult to accurately determine the appropriate level of compensation. This ambiguity is likely heightened because there are no objective standards by which to set director compensation (Main, O’Reilly, & Wade, 1995).

Due in part to this ambiguity, prior research suggests that the board is likely to engage in social comparison when determining CEO pay (see Finkelstein et al., 2009, for a review; O’Reilly et al., 1988). For example, research has shown that CEO pay is homogenized because of social comparisons and the use of compensation consultants (Bebchuk & Fried, 2004; Bizjak, Lemmon, & Naveen, 2008; Conyon et al., 2009; Wade et al., 1997). In addition, the social capital of executives influences their pay through a process in which executives with greater social status or connections than comparison groups receive more favorable compensation (Belliveau, O’Reilly, & Wade, 1996; Westphal & Zajac, 1995). For instance, when CEOs are certified by prominent others as being high status or a “star,” compensation rises accordingly (Graffin, Wade, Porac, & McNamee, 2008; Wade, Porac, Pollock, & Graffin, 2006). Social comparison processes may also affect executive pay because executive pay is published in corporate proxy statements, which removes secrecy and facilitates comparisons (Finkelstein et al., 2009). Because open information about pay typically promotes social comparison pressures (Pfeffer & Davis-Blake, 1990), executive pay may be viewed as a scorecard that indicates professional status relative to others (Finkelstein et al., 2009). Finally, research demonstrates that the board may look to the pay of similar others and their own pay to justify the amount of CEO pay at the focal firm (Porac et al., 1999). For example, O’Reilly et al. (1988) suggest that directors are likely to use their own level of compensation as a reference point when determining CEO pay.

Social comparison research on executive pay is based in large part on principles from equity theory, which is conceptually related to social comparison theory and is at the center of all compensation theory (Wallace & Fay, 1983). Equity theory offers predictions about how individuals react to overreward and underreward situations. According to equity theory, individuals make subjective assessments of the ratio of their inputs (e.g., effort) and outcomes (e.g., compensation) to those of referent others and experience dissonance when their perceived ratio is unequal to that of others. As a result, individuals often reduce effort or push to increase compensation to maintain a similar ratio to salient others.

Equity theory has been applied at both the individual and team levels of analysis (Rhodebeck, 1981). At the team level, Roberson and Colquitt’s (2005) network model views teams as a type of social network in which the patterns and intensity of ties between members influence inequity perceptions through two mechanisms, structural equivalence and cohesion. Structural equivalence refers to the extent to which people occupy similar positions in a network (Erickson, 1988), while cohesion refers to the frequency and intensity of interaction (Burt, 1987). With respect to boards, the structural equivalence perspective suggests that similar patterns of relationships between board members will allow individual directors access to the same information about potential comparison groups and facilitate a shared perspective about the fairness of their pay. At the same time cohesion should lead to the exchange of justice or equity information among board members by providing an opportunity for members to discuss and jointly interpret the team’s experiences (Roberson & Colquitt, 2005). Similarly, social cohesion can facilitate social contagion of equity perceptions where the views of individual directors spread throughout the board until converging to a unified view (Degoey, 2000). Thus, both the cohesion and structural equivalence perspectives suggest that board members’ evaluations of board relevant justice events converge into one view and allow for group-level social comparisons.

While social comparison processes have been shown to play a role in setting CEO compensation, extant research has yet to consider how social comparison processes may affect decisions regarding board compensation. The board sets director compensation, typically based on the recommendation of the compensation committee (which is composed of members of the board). In determining the appropriate level of compensation, directors must balance numerous considerations including the need to compensate sufficiently to attract top talent to the board while also aligning director incentives with those of shareholders, along with the need to demonstrate appropriate respect to powerful directors. Given the ambiguity surrounding the appropriate level of director compensation, directors are likely to look for social comparison targets on which to base decisions about director pay. The literature on anchoring and adjustment (Tversky & Kahneman, 1974) suggests that people make estimates based on some initial value and then adjust their initial estimates based on new information. However, the initial estimate usually has a dramatic effect on the final estimate. Boards are likely to seek out a salient basis of comparison for setting director compensation to serve as an anchor point for setting director compensation at the focal firm.

Research on social referents has shown that individuals seek out social referents based on the availability of information and the relevance of the referent (Goodman, 1974; Kulik & Ambrose, 1992). While much of the social comparison literature is focused on determining what individuals or groups are likely to serve as referents, in the case of boards, the most readily available and relevant information is likely to come from the directors themselves from their experience at other firms. In particular, the level of board compensation at directors’ home firms (e.g., where they work as full-time employees) or the compensation at their other board appointments should serve as salient reference points. These are not the only referents that can be used by directors (e.g., as we discussed earlier, evidence from the CEO compensation literature shows that there are a number of possible referents individuals can use when making comparisons). However, our goal here is to explore one particularly likely referent, namely, the directors themselves in other roles.

It is likely that this social comparison process of anchoring director pay based on readily available and relevant comparison groups will work to push director compensation higher. Work tying social comparison theory to pay finds that people will typically seek out similar referents with which to compare their level of pay (Goodman, 1974). To the extent that the focal individual makes more than the referent, they may increase outputs or rationalize their compensation, but in most cases such individuals will be less motivated to push for significant pay increases because doing so would increase feelings of inequity or tension (Adams, 1965). When individuals make less than a referent, they are typically less satisfied (Greenberg, 1990, 1993) and may be more likely to leave or push for increased compensation (Gomez-Mejia & Balkin, 1992). Attempts to increase compensation are especially likely in such a situation given research that suggests that feelings of deprivation or inequity can result when individuals compare themselves with similar others and find their own situation to be lacking (Kulik & Ambrose, 1992).

In all of these possible scenarios, individuals use referents to justify increases in pay or to maintain current levels. Individuals would rarely use social referents as a justification to decrease their pay. Because the board has the authority to set the level of director compensation, we would expect that when the gap between director pay at the focal firm and director pay of social referents increases, director compensation will increase at the focal firm. When director pay at the focal firm is higher than a referent, the board may see less need to increase pay, but in such cases the board will likely generate self-serving justifications for why their pay is higher. Consequently, it is unlikely that these comparisons with other boards would lead to voluntary decreases in director compensation such that the social comparison forces described above work to continually increase compensation over time. We are suggesting that compensation of directors, like other types of compensation, is likely to almost always rise and almost never fall. In other words, it will ratchet up, just as a mechanical ratchet is designed to turn in only one direction. While director compensation is likely to always be rising, our theory explains how social comparison forces should exacerbate this tendency for pay to increase.

Thus, when a board is deciding to set its compensation level, we suggest that the directors will use the compensation of their other boards (either for the firms where they also serve as a director or the boards of the firms where they are employed full-time) as a reference. As the evidence above shows, firms often use relevant comparison groups when setting appropriate compensation levels. The most relevant comparison groups for boards are other boards. Specifically, board members often sit on the boards of other public firms, and so they will be aware of the compensation from their other board appointments. In addition, board members are likely to be well aware of the compensation received by the board of directors at their home firm where they are employed full-time. Consequently, when determining appropriate pay, directors are most likely to use those groups with which they are most familiar, namely the other boards on which they sit, and the board of their home firm as appropriate social referents. Directors will have direct knowledge of the compensation at these firms and will also feel that these firms are appropriate comparison firms because of their service therein. They are likely to feel justified in awarding compensation at the focal firm at least as high as what is paid in these other board appointments. This leads to our first two hypotheses:

Hypothesis 1: The greater the difference between board compensation at directors’ other board appointments and compensation at the focal firm, the greater the increase in director compensation at the focal firm.

Hypothesis 2: The greater the difference between board compensation at directors’ home firms and board compensation at the focal firm, the greater the increase in director compensation at the focal firm.

In the previous discussion, we considered how the board might view the compensation of other boards as valid social referents when setting its own level of compensation. However, as discussed earlier, the compensation committee, which is responsible for the compensation of the board, is also responsible for setting the compensation of the CEO. In fact, setting the compensation of the CEO (and the other top executives) is considered one of its primary responsibilities. In this section we consider how the process of social comparison described earlier, along with the norm of reciprocity, may lead to a relationship between changes in the compensation of the CEO and changes in the compensation of the board.

First, we consider how the norm of reciprocity may help to explain how changes in CEO compensation can affect changes in director compensation. The norm of reciprocity suggests that actions will be returned in kind (Gouldner, 1960). A number of studies on board power suggest that powerful CEOs are able to exert influence over their level of compensation (Boyd, 1994; Brick et al., 2006; Westphal, 1998). CEO influence over the board often stems in part from feelings of indebtedness (i.e., reciprocity) by the board. Thus, one way for powerful CEOs to increase their own compensation could be to encourage the board to increase its own compensation. For example, F. Ross Johnson, former CEO of RJR Nabisco, expressed a desire for a high level of director compensation by saying, “If I’m there for them, they’ll be there for me” (Boyd, 1994: 339). So when the board votes to increase CEO compensation, we might expect that the board will also feel justified in increasing its own compensation (Brick et al., 2006). In addition, boards may increase CEO pay for various reasons including payback for a board appointment for which a director feels beholden to the CEO. When granting pay increases for reasons that are not directly tied to performance, directors may feel like they are also entitled to pay increases. In addition, even boards that are formally independent may not be socially independent in that they may be demographically similar to the CEO or have friendship or other social ties that may increase trust but also make directors less able to objectively monitor and control management (Westphal, 1999; Westphal & Zajac, 1995). Trust between the CEO and the board is important for effective board functioning (Westphal, 1999) but also creates opportunities for reciprocity that can be good or bad. When board members and executives are friends they may feel inclined toward helping each other by supporting mutual pay increases.

Social comparison processes between directors and the CEO will also likely result in reciprocal increases of director and executive pay. Although, as discussed previously, the board may have many potential reasons for increasing CEO pay, regardless of the true reasons for increasing a CEOs pay, boards typically justify increases in CEO pay as either (a) labor market related, (b) work level related, or (c) incentive alignment related (Zajac & Westphal, 1995). Labor market reasons would include the argument that increases in CEO pay are necessary to attract and retain superior managerial talent (Carpenter, Sanders, & Gregersen, 2001; Fama, 1980; Fama & Jensen, 1983; Harris & Helfat, 1997; Zajac & Westphal, 1995). Work-related justifications typically have to do with rewarding performance (Carpenter & Sanders, 2004; Westphal & Zajac, 1994) or focusing on the level of complexity and accompanying information processing demands required by the job (Henderson & Fredrickson, 1996). Incentive alignment justifications would be used to increase pay by giving stock or other incentives that should align managerial interests with those of shareholders. As the board and compensation committee prepares its formal justification for CEO pay increases, it is likely to internalize those reasons and come to believe that they also apply to the board as well as the CEO.

Pay is a measure of comparative success for executives (Finkelstein et al., 2009; Fredrickson, Davis-Blake, & Sanders, 2010), so when the board grants a pay increase to the CEO, it will likely see a highly comparable peer rising in status. In fact, research has concluded that the status derived from pay is more important than the actual financial reward itself (Main, O’Reilly, & Wade, 1993). Because the CEO is virtually always a member of the board, other members of the board are likely to view themselves as having similar status and will therefore feel justified in increasing their own pay. In addition, board members, including the CEO, tend to be a largely homogenous group, and the greater the similarity between individuals, the more likely that a person views them as a valid social referent (O’Reilly et al., 1988; Wade et al., 2006). Although the total level of CEO pay is much higher than the compensation of directors, the board is still likely to view the CEO as a valid social referent because they help set his or her pay and because of their joint board service. Furthermore, a large increase in CEO compensation makes pay more salient to all members of the board, likely triggering the norm of reciprocity regarding the board’s own pay.

In addition, individuals dislike high levels of pay dispersion (Bloom, 1999), so increases in CEO pay may make members of the board feel uncomfortable if they do not also receive pay increases. Recent research has shown that shared board membership decreases pay dispersion among non-CEO top management team members (Fredrickson et al., 2010), indicating that the board feels pressure to maintain equity among its members. Even though CEOs make much more than the average board member (at least at the focal firm), the fact that a CEO is receiving a large pay increase could cause feelings of inequity if directors do not also receive an increase (although a smaller one in absolute dollars). Consequently, pay increases to the CEO may lead to feelings of inequity among the board, which may prompt it to raise its own pay levels. Taken together, this leads to our final hypothesis:

Hypothesis 3: Increases in the compensation of the CEO will be associated with increases in director compensation.

Method

We collected data on all of the firms that maintained membership in the S&P 500 from 1996 to 2005 using a variety of archival databases including the RiskMetrics (formerly IRRC) Directors, which we used to identify home and interlocked firms and to collect board-level controls. To collect executive and director compensation, we used Execucomp’s director and executive compensation database. Finally, several of our firm-level control variables (e.g., performance, sales etc.) were collected from Compustat. This resulted in a sample of 281 firms. We chose firms from the S&P 500 because those firms are (a) the most prominent firms in the economy and (b) the firms with the most complete data on compensation of executives and directors. Using data from these archival sources, we created variables about the focal firms in our sample, and the firms to which these focal firms were tied through directors’ home firms and directors’ other board appointments. Missing data, creating change variables, and lagging of our independent variables reduced our final sample to 1,003 firm years for our total compensation models and 1,122 for our base compensation models covering the period from 1998 to 2005. All compensation variables were adjusted for inflation to 1996 levels using the consumer price index.

Dependent Variables

The primary dependent measure in our study is change in the level of director compensation. We decided to examine changes in compensation because our hypotheses not merely are concerned with overall levels of director compensation, but more specifically are arguing that tension caused by social comparisons will cause a reaction in the form of compensation changes. We are interested in the magnitude of this reaction. To calculate the change in director compensation, we obtained the level of director compensation from the Execucomp database and then calculated change in compensation by subtracting director compensation in Year t−1 from the current year. Our theory suggests that directors will be influenced by the tension between their pay at their other appointments (or their home firm) and the pay at the focal firm and that this will be reflected in changes in pay at the focal firm. However, it is possible that rather than focusing on the raw differences, directors are mentally calculating the percentage difference in the pay from their various sources and so they will push for similar percentage pay differences. Consequently, we also ran models where we calculated the DV as the percentage change in pay at the focal firm and our results were similar to those reported in the following.

Director compensation is made up of annual retainers, meeting fees, stock and options grants, and other director compensation and is generally fairly uniform across the board.¹ Thus, we measure director compensation at the board level. We created separate measures for change in the annual retainer, which represents the board’s base compensation, and total board compensation, which includes the annual retainer plus meeting fees and stock compensation. The annual retainer is somewhat analogous to CEO salary since it is fixed and not directly tied to performance, while total director compensation includes a performance-based component. Consistent with prior research (Boivie, Lange, McDonald, & Westphal, 2011; Sanders, 2001), stock options were valued using the Black–Scholes method (Black & Scholes, 1973).

Independent Variables

Our theory suggests that social comparison processes come into play when the board is deciding on the appropriate level of director compensation. Two relevant and salient sources of comparison are the level of director compensation at directors’ home firms (i.e., the firms where outside directors work as full-time employees) and the level of director compensation at other boards on which directors have appointments (i.e., interlocked firms). These two sources of compensation are likely to be used for comparison purposes because the level of compensation is known to the directors and because of their close association with those firms. To capture these comparisons, we measured the difference between the director compensation (both annual retainer and total) at the focal firm and the average level of director compensation at directors’ home companies and interlocked firms. Again, we will note that we used difference scores rather than levels because our theory suggests that it is differences that trigger feelings of inequity and discomfort and subsequent increases in compensation. We calculated the average director compensation from the boards of all outside directors’ home firms that had data available in Execucomp. In the same manner, we calculated the average annual retainer and total director compensation from all firms where outside directors held additional board seats. We then followed the procedure outlined by Bizjak and colleagues (2008) to create the difference between the comparison firms and the focal firm by subtracting the level of board compensation at the focal firm (for retainer and total compensation) from the average board compensation of the comparison group (home and interlock firms). Therefore, our resulting independent variables are home company retainer pay difference, home company total pay difference, interlock retainer pay difference, and interlock total pay difference. Positive values of our difference variables represent a situation where the focal firm pays its directors less than the comparison firms. Our theory suggests that as this difference becomes greater, firms are more likely to increase director compensation. As an additional measure, we created a set of dummy variables equal to one if the focal firm paid its board less than the home or interlocked firms respectively. We also created a similar set of measures to indicate that the firm paid its directors more than the comparison groups. In supplementary analysis using these indicator variables to predict changes in director compensation, the results were substantively unchanged from the models reported below.

Our final independent variable was change in CEO compensation, which we measured for both CEO salary and total CEO compensation. We obtained all CEO compensation data from the Execucomp database and captured change in CEO salary by subtracting CEO salary in Year t−1 from CEO salary in the current year. Change in total CEO compensation was calculated in a similar fashion, although total compensation included salary, bonus, and other awards (stock, restricted stock, options, other long-term incentive plan). For the models predicting changes in the annual director retainer, we used the change in CEO salary as the independent variable of interest since salary is analogous to director base compensation and may represent a more salient basis of comparison. For the models predicting change in total director compensation, we used change in CEO total compensation. Because total compensation can often be skewed, in supplementary analyses we ran models that used the natural log of total compensation to create the change in compensation variable, and the results were substantively unchanged from those reported in the following.

Control Variables

We include a number of variables as controls to rule out confounding factors that could be affecting our results. Because our construct of interest is the change in the level of director compensation, when including controls we tried to determine whether the proper functional form of each control variable would be to model the control as “stock” (e.g., the current level or amount of that particular variable) or as a “flow” (e.g., the difference in the variable from the prior year). We made our choice based on prior research as well as what made the most conceptual sense. However, in separate models not shown, we did run analyses where we included all of the controls as either stocks or flows, and the results of our hypothesized variables were unchanged.

There is some evidence suggesting that CEO status is related to executive compensation (Belliveau et al., 1996). Thus, we included several controls for director status since boards with higher status directors may demand larger increases in compensation. We created a variable total board appointments that consisted of the average number of directors’ other concurrent board appointments since board appointments are sometimes seen as an indicator of a high status director (D’Aveni, 1990). We also created a variable, S&P 500 CEOs that controlled for the number of directors that were also CEOs of an S&P 500 firm. These two variables were measured as stocks because there was not a lot of difference in the variables across years. In separate models, we also controlled for characteristics of the directors’ home firms including the average size (in revenue) and performance (return on assets) of directors’ home firms, and our results were substantively unchanged.

We also controlled for factors that may increase the information processing demands of the board (Henderson & Fredrickson, 1996). It could be that increases in compensation are made to account for the increased work that occurs as a firm grows in size (Tosi, Werner, Katz, & Gomez-Mejia, 2000). Thus, we controlled for change in firm size, measured as the difference in the firm’s yearly revenue compared with the prior year. In separate analyses we controlled for firm size using the number of employees, and the results were very similar to those reported here. Diversified firms may also be more complex and require increased information processing demands. We calculated firm diversification as a count of unique business segments (identified by NAICS/SIC codes) in which the firm has revenues. This variable was calculated as a stock because changes from year to year were minor. As an alternative, an entropy measure (Jacquemin & Berry, 1979) for diversification was calculated, and the results were substantively unchanged.

Because reciprocal ties between focal firm executives and outside directors may influence decisions regarding compensation levels, direct interlocks were included as a control and were coded as 1 if the CEO of the focal company sits on the board of a director’s home company. This variable was measured as a stock because it did not make sense conceptually to measure changes from the prior year. Because regulatory changes and other external forces, such as Sarbanes–Oxley, may influence board compensation, we included year dummy variables for each of the years included in our sample (excluding the first year, which is 1998, because we lose one year by lagging the variables and another year because we are calculating a change). We also ran separate models, not reported, where instead of yearly dummies we created a dummy variable that indicated whether the period was pre– or post–Sarbanes–Oxley, and the results were unchanged. Other governance characteristics may also influence compensation decisions, so we controlled for factors associated with governance effectiveness including the outsider ratio, the change in the size of the board, and board turnover. We measured the board size as a “flow” because an increase in the size of the board could reduce the workload on individual board members and thus affect the likelihood that the base levels of director pay will increase. We measured the outsider ratio as a stock because it does not change much from year to year. Board turnover is already conceptually modeling change from the prior year. Because changes in firm performance have been linked to changes in executive compensation (Tosi et al., 2000), we also included a measure of the change in firm performance measured as the difference in the firm’s return on assets from the prior year. In separate analyses, we controlled for the average board pay across the entire sample to help control for market forces that are influencing pay. The results were substantively unchanged from those reported here. Based on suggestions from one of our reviewers, we also ran models where we included the level of pay at the focal firm from two years prior to rule out the possibility that some firms simply pay at higher levels and that it is this high pay level in general that leads to increases in pay. Our results were unchanged from those reported here.

Analysis

Because we have yearly panel data, we used fixed effects linear regression analysis to test our hypotheses. Fixed effects models are the equivalent of including a dummy variable for each firm in the sample and thus control for unobserved firm characteristics that may influence the dependent variable (Greene, 2000). A Hausman specification test confirmed that fixed effects models were appropriate in our models. In all of our models, we lagged the independent variables by one year unless otherwise noted. We did not lag changes in CEO compensation since changes in director and executive compensation are likely to occur in the same period based on the arguments outlined previously. Because we have multiple years of data we wanted to ensure that our models were not biased due to the presence of serial correlation, consequently we also ran generalized least squares and generalized estimating equation models, and our results were unchanged.

In the management literature, there has been some concern over the use of change scores as dependent variables (Edwards, 1994, 1995). Change scores have been criticized because the difference between two component variables often has a lower reliability than the individual component measures (Bergh & Fairbank, 2002; Edwards, 1994, 1995). However, in our case, because we are measuring what Bergh and Fairbank (2002) refer to as “true change” in an objective variable, our reliability is necessarily 100% and simple difference scores are appropriate (Allison, 1990; Bergh & Fairbank, 2002). In so doing, we follow a long history of research that appropriately uses change scores to measure true change (Bizjak et al., 2008; Farrell et al., 2008; Henderson & Fredrickson, 2001; Jensen & Murphy, 1990).

Results

Table 1 displays descriptive statistics and a correlation matrix for all of the variables.

Table 1

Descriptive Statistics and Correlation Table

	M	SD	1	2	3	4	5	6	7	8
1. Change annual retainer	1785.87	8450.05	1
2. Change total compensation	3705.50	50032.63	.17***	1
3. Difference from home firm (retainer)	−3174.71	16109.90	.21***	.00	1
4. Difference from home firm (total)	−17267.11	76072.43	.08*	.35***	.15***	1
5. Difference from interlocked firm (retainer)	−4705.91	13840.26	.23***	.04	.74***	.14***	1
6. Difference from interlocked firm (total)	−13550.67	71101.67	.07*	.37***	.06^†	.83***	.14***	1
7. Change CEO salary (thousands)	25.35	175.65	.02	.00	−.03	.02	.00	.01	1
8. Change total CEO compensation (thousands)	1100.27	22500.00	−.04	.18***	−.03	.05	−.01	.03	.02	1
9. Avg. number of other appointments	1.38	0.61	.00	−.01	−.06*	−.11***	−.07*	−.13***	−.02	.01
10. Number of S&P 500 CEOs	1.06	1.10	.02	.00	.01	.02	−.05	−.03	−.02	.02
11. Board turnover	0.10	0.10	.02	−.01	.01	.04	.00	.00	.03	.00
12. Interlocks	0.04	0.20	.00	.02	.07*	.03	.03	.06^†	−.02	−.01
13. Change performance (ROA)	0.00	0.08	.03	.17***	−.02	.07*	.02	.03	.04	.03
14. Change board size	−0.11	1.29	−.05	−.07*	−.02	−.09**	−.01	−.09**	.02	−.01
15. Change firm size (sales)	1027.49	3328.36	.14***	.04	−.04	−.14***	−.03	−.17***	.08*	.08**
16. Segment count	2.22	1.25	.07*	.02	−.07*	−.03	−.05^†	−.03	.06*	−.02
17. Outside ratio	0.84	0.08	.02	−.01	.05^†	.04	.09**	.02	.03	−.05
	9	10	11	12	13	14	15	16	17
9. Avg. number of other appointments	1
10. Number of S&P 500 CEOs	.33***	1
11. Board turnover	.05	.06^†	1
12. Interlocks	−.08*	−.05	−.02	1
13. Performance (ROA)	−.02	.00	.03	.02	1
14. Board size	.02	−.05	.12***	.01	.01	1
15. Firm size (sales)	.12***	.08**	.03	−.03	.03	−.02	1
16. Segment count	.13***	.07*	.05	.00	.00	−.03	.08**	1
17. Outside ratio	.02	.16***	.06^†	−.08**	.01	.01	−.02	.05^†	1

Note: N = 1,003. ROA = return on assets.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table 2 shows an analysis of the average change in director compensation for firms with director compensation that is above and below their respective comparison groups. This table helps to test our assumption that pay will increase and not decrease. For example, we can see that the average increase in annual retainer for firms that paid directors less than the board members at directors’ home firms was $3,088, while firms increased the retainer by $1,029 when focal pay was higher than directors’ home firms. This pattern holds when the comparison group is interlocked firms and also for total compensation. It is interesting to note that in all cases, director pay increases, even after adjusting for inflation and even when the firm is already paying more than the comparison firms. So we see that even when the comparison groups pay at a lower level than the focal firm, pay is still likely to rise. However, as predicted, the increases are substantially higher when pay is below the comparison group average. Simple t tests confirm that all differences shown in the table are highly significant. While the results from Table 2 provide strong evidence that director pay has the tendency to increase and not decrease, our further analysis was designed to more explicitly test our hypotheses.

Table 2

Average Change in Director Compensation

	Annual Retainer		Total Compensation
	Comparison Group		Comparison Group
	Home Firm	Interlocked Firms	Home Firm	Interlocked Firms
Lower than comparison group	3088.21	3946.75	9345.34	12650.20
Higher than comparison group	1038.67	870.39	2039.70	1125.70
n	1,453	2,185	1,372	2,030

Table 3 shows the results of fixed effects linear regression analysis for changes in base director compensation, while Table 4 shows similar results for changes in total director compensation. Because of the high correlation between the interlocked firm and home firm variables, we ran models where we entered each independent variable separately (i.e., Models 2, 3, and 4 in each table) to reduce concerns about multicollinearity. We then include the results of a fully specified model (i.e., Model 5). Hypothesis 1 predicted that the difference between director compensation at directors’ other board appointments and director compensation at the focal firm would be associated with increases in director compensation. Model 1 is the baseline model for change in annual retainer and includes only the controls. Model 2 in Table 3 shows that interlocked firm difference is positive and significantly associated with increases in annual retainers (p < .001), and Model 2 in Table 4 shows a similar relationship for total compensation (p < .001). These results provide robust support for Hypothesis 1.

Table 3

Change in Base Compensation

	Model 1	Model 2	Model 3	Model 4	Model 5
Constant	10721.46^†	19284.80***	12853.58*	10992.33^†	19323.64***
	(6304.99)	(5802.96)	(6041.33)	(6300.33)	(5800.85)
Total board appointments	172.24	−31.36	171.52	215.65	28.90
	(1128.83)	(1032.49)	(1080.79)	(1127.93)	(1030.99)
S&P 500 CEOs	−119.90	282.69	13.43	−84.44	313.88
	(436.65)	(400.51)	(418.33)	(436.69)	(400.36)
Board turnover	1461.50	757.51	59.52	1207.02	326.43
	(3308.71)	(3026.45)	(3171.74)	(3308.61)	(3027.86)
Interlocks	−80.05	−607.42	−538.35	16.43	−534.27
	(2130.35)	(1948.73)	(2040.33)	(2128.85)	(1946.17)
Change in ROA	3915.30	3783.97	6719.949^†	3373.92	3609.52
	(4072.29)	(3724.32)	(3911.46)	(4080.52)	(3751.35)
Change in sales	0.19	0.16	0.17	0.17	0.15
	(0.12)	(0.11)	(0.11)	(0.12)	(0.11)
Change in board size	−589.203*	−364.31	−580.502*	−582.129*	−370.13
	(246.68)	(226.25)	(236.18)	(246.45)	(226.30)
Diversification	314.67	27.02	508.39	306.36	64.72
	(579.92)	(530.81)	(555.65)	(579.32)	(531.91)
Outside ratio	−14255.26 *	−22136.15***	−16648.49*	−14820.96*	−22576.57***
	(7169.31)	(6584.08)	(6869.37)	(7169.50)	(6582.49)
1999	476.40	218.30	147.39	523.11	238.48
	(1151.57)	(1053.35)	(1103.17)	(1150.68)	(1052.19)
2000	−420.34	125.73	−540.00	−376.28	120.46
	(1184.91)	(1084.45)	(1134.56)	(1183.94)	(1084.10)
2001	478.73	508.89	183.75	490.94	477.72
	(1203.53)	(1100.69)	(1152.78)	(1202.27)	(1099.61)
2002	2667.04*	2809.731*	2469.08*	2819.26*	2936.78**
	(1223.67)	(1119.16)	(1171.80)	(1225.67)	(1120.85)
2003	4089.14**	5266.62***	4559.25***	4229.38***	5408.96***
	(1264.62)	(1160.03)	(1211.93)	(1265.99)	(1160.58)
2004	5593.99***	7824.92***	6641.68***	5752.70***	8001.28***
	(1287.30)	(1189.49)	(1238.02)	(1289.35)	(1190.61)
2005	2378.86^†	5179.99***	4015.86**	2562.297^†	5430.51***
	(1343.52)	(1247.09)	(1299.19)	(1346.47)	(1249.79)
Interlock firm difference		0.45***			0.42***
		(0.03)			(0.05)
Home firm difference			0.26***		0.04
			(0.03)		(0.04)
Change CEO compensation				3.12*	3.38*
				(1.84)	(1.69)
N	1,122	1,122	1,122	1,122	1,122
R ²	.05	.21	.13	.06	.21
Δ R²		.16***	.08***	.01^†	.16***

Note: ROA = return on assets.

†

p < .10. *p < .05. **p < .01. ***p < .001 (one-tailed for hypothesized effects, two-tailed for controls). Standard errors in parentheses.

Table 4

Change in Total Compensation

	Model 1	Model 2	Model 3	Model 4	Model 5
Constant	5024.98	32947.38	11304.59	1424.08	25095.08
	(34048.86)	(31183.34)	(31835.44)	(33396.19)	(30220.41)
Total board appointments	−3185.65	−3760.40	−1473.76	−4183.57	−4040.68
	(6235.63)	(5696.04)	(5831.50)	(6117.55)	(5521.41)
S&P 500 CEOs	426.88	−213.75	755.68	−102.20	−512.80
	(2275.85)	(2079.49)	(2127.76)	(2233.81)	(2016.54)
Board turnover	−545.72	−4813.60	−13100.00	−7283.33	−16100.00
	(17876.22)	(16332.44)	(16753.75)	(17571.40)	(15899.42)
Interlocks	2878.12	−1187.55	28.48	5288.84	1246.13
	(11967.04)	(10936.07)	(11190.40)	(11743.34)	(10590.94)
Change in ROA	82141.21***	87374.87***	77118.19***	79190.38***	80991.58***
	(21367.17)	(19522.14)	(19980.33)	(20960.32)	(18931.09)
Change in sales	0.70	1.157*	0.91	0.27	0.67
	(0.64)	(0.58)	(0.60)	(0.63)	(0.57)
Change in board size	−3417.64*	−2170.25^†	−2174.38^†	−3224.40*	−1757.62
	(1348.65)	(1236.03)	(1266.25)	(1323.00)	(1198.28)
Diversification	1792.04	1240.86	1362.60	1563.73	950.28
	(3154.95)	(2882.19)	(2949.62)	(3094.17)	(2789.74)
Outside ratio	−1562.47	−21100.00	190.10	6100.78	−7625.80
	(38787.52)	(35465.27)	(36260.10)	(38061.50)	(34394.75)
1999	2850.40	1359.95	1673.81	2403.64	753.64
	(5927.74)	(5415.96)	(5542.55)	(5813.59)	(5242.36)
2000	3879.94	5273.18	3466.93	1774.70	2502.28
	(6218.11)	(5680.96)	(5813.00)	(6109.32)	(5511.23)
2001	−20114.33**	−19505.66***	−19239.25**	−19167.02**	−18271.36**
	(6322.66)	(5775.56)	(5911.19)	(6202.62)	(5592.28)
2002	−12667.32^†	−23316.08***	−21083.86***	−11544.85^†	−22961.01***
	(6490.22)	(5990.64)	(6119.44)	(6367.78)	(5803.93)
2003	5813.16	−2465.44	−713.08	6553.19	−2326.91
	(6949.94)	(6383.54)	(6526.37)	(6816.74)	(6181.17)
2004	7931.38	5804.45	5454.53	7685.14	5051.10
	(6982.45)	(6380.34)	(6531.62)	(6847.48)	(6177.01)
2005	2449.99	479.30	1635.61	2340.62	492.35
	(7277.95)	(6649.85)	(6804.08)	(7137.15)	(6436.17)
Interlock firm difference		0.44***			0.34***
		(0.04)			(0.05)
Home firm difference			0.34***		0.14**
			(0.03)		(0.05)
Change CEO compensation				0.63***	0.68***
				(0.11)	(0.10)
N	1,003	1,003	1,003	1,003	1,003
R ²	.08	.23	.19	.11	.28
ΔR²		.15***	.11***	.03***	.20***

Note: ROA = return on assets.

†

p < .10. *p < .05. **p < .01. ***p < .001 (one-tailed for hypothesized effects, two-tailed for controls).

Hypothesis 2 predicted a similar relationship with directors’ home firms as the basis for comparison. As shown in Model 3 of Table 3, the difference between the annual retainer at directors’ home firms and the annual retainer at the focal firm (home firm difference) predicted changes in base compensation (p < .001). Model 3 in Table 4 shows that the effect was similar for total director compensation (p < .001). These results provide strong support for Hypothesis 2.

Hypothesis 3 predicted that changes in CEO compensation would accompany changes in director compensation. Model 4 of Table 3 shows that changes in CEO salary were associated with changes in base director compensation, while Model 4 of Table 4 shows that changes in total CEO compensation were associated with changes in total director compensation. Although the coefficient is positive and significant as predicted in Model 4 of Table 3 (the base pay models), the addition of the variable does not produce a significant increase in R², although it does produce a significant improvement in R² in Model 4 of Table 4 (the total compensation models). Thus, Hypothesis 3 is partially supported. In supplementary analyses, we used lagged changes in CEO compensation to predict changes in director compensation. The results were not statistically significant suggesting that CEO pay and director pay tend to rise simultaneously.

In summary, all three of our hypotheses were generally supported, and the results provide good support for our theoretical arguments. More specifically, the results suggest that the compensation of interlocked firms and directors home firms are salient bases of comparison for boards when deciding on the appropriate levels of director compensation and that increases in director compensation tend to accompany increases in CEO pay.

Discussion

This study was motivated by trying to better understand how social comparison and reciprocity mechanisms affect changes in director compensation, and to see if these social mechanisms can help to explain in part why director pay has risen substantially in recent years. We believe that the theoretical arguments and empirical results presented in this article provide consistent evidence supporting our social perspective of director compensation and that the theory and evidence presented in this study provide a number of valuable contributions to the literature.

First, our study contributes to our knowledge of mechanisms through which pay increases disperse throughout the director labor market. Recent research about director compensation has shown that a combination of factors have driven up demand in the market and reduced the supply of quality directors (Linck et al., 2009), but what has been left unexplored is how these market forces are translated by directors into actual pay increases. We theorized and found evidence that director compensation is affected by the compensation that directors receive at their other board appointments, at their home firms, and by the compensation of the CEO at the focal firm. Thus, this study represents an important first step in understanding how decisions about director compensation are made and how labor market forces are translated into increased compensation. Increasing our understanding of this process is especially important given the fact that directors are one of the most unique groups in modern public corporations because they set their own compensation.

Second, we argued and found support for the claim that the greater the difference between the board compensation at directors other board appointments, or their home firms, the greater the increase in compensation at the focal firm. Thus, unlike prior work on director compensation, which typically uses economic perspectives such as agency theory, this article offers a social comparison perspective emphasizing team-level reaction to perceived inequity that complements prior work in this area. While prior research on social comparison processes and compensation has focused exclusively on the arena of executive compensation our study introduces the idea of social comparison in an important new context. Our theory and findings make a contribution to our understanding of how social comparison processes affect boards, especially with regard to their compensation. We draw on equity theory as well as the social network and reciprocity literatures to provide greater insight into how social comparison translates into hirer levels of director pay. We also extend prior work by identifying the unique social referents used by directors when determining their compensation. We argued and found support for the proposition that directors may use themselves, in other role contexts, as referent others when making social comparisons. We found strong support for our arguments that directors view their other board appointments, the boards of their home firms, and even the executives of the focal firm as valid social referents and are likely to use them as points of comparison.

Third, we found evidence that director pay and CEO pay are connected by showing that rises in executive compensation are positively associated with increases in board pay. Our theory suggests that reciprocity mechanisms may play a substantial role in this process whereby directors feel more justified in increasing their own pay when they have given the CEO a substantial increase. Because our results here are based on contemporaneous models we cannot make any claims of causality. What we find is that there appears to be a conditional correlation between executive and director pay increases. Although other factors may also contribute to the joint determination of executive and director compensation, this finding is an important first step to better understand how social mechanisms can play a role in determining how compensation decisions are made at the highest levels of the firm.

Finally, the social comparison mechanism that we describe suggests a nonlinear diffusion effect that is quite different from prior research. Prior work on the diffusion of information or practices across board interlock ties suggests a simple replication of policies through ties (Davis, 1991; Haunschild, 1993; Haunschild & Beckman, 1998). Our study suggests that because boards set their own compensation, when they undertake social comparisons across network ties, they are likely to use that information selectively and in a self-serving manner to justify pay increases. Thus, directors are especially attuned to network information that suggests they are underpaid, while information that indicates that directors are adequately compensated appears to have less of an effect. So in this manner, information will flow across these network ties in an asymmetric fashion.

Our findings have a number of practical implications. First, our study helps to shed light on how the recent trend of increasing director compensation in the market becomes increased pay at a focal firm. While much has been made of increased demands on directors in the wake of Sarbanes–Oxley and other corporate scandals (Linck et al., 2009), our results suggest that social comparison processes, particularly reactions to perceived inequity, play a substantial role in driving up director pay. The increased demands on directors may actually exacerbate these social comparison processes as directors feel even more justified in increasing pay when a disparity exists with a salient referent group. It appears that one response of boards to increased expectations from the financial community has been to increase the compensation that they provide themselves (Linck et al., 2009). This raises the important question of whether the increased pay for directors has produced an increased commitment by directors themselves or has increased overall board effectiveness. Prior research has reported fairly weak results when trying to connect the board with improved firm outcomes. Perhaps increased pay will also result in increased motivation for directors and ultimately increased board effectiveness.

Our results also suggest that because social comparison processes generally tend to lead to increases in compensation for directors, if corporate governance activists are worried about rising compensation levels, then they should work to reduce the rate at which compensation increases. It seems unlikely that directors will voluntarily reduce their compensation, but they may be willing to freeze it or to increase it at a slower rate. One possible implication of our findings is that if governance activists seek to influence director compensation, then they may need to work to increase the salience of social referents that receive lower levels of compensation. Although we believe that directors are more inclined to use higher-paid firms as their referents, perhaps if they are more vocally reminded of how lower-paid firms are also valid referents they will increase pay less frequently. In addition, increasing exposure of social comparison processes in director compensation may encourage external constituents such as institutional investors or the media to increase pressure on firms to pay directors for performance or according to the unique demands of a particular directorship rather than according to what high-paid comparison groups earn.

Future Research Directions

This research also suggests a number of promising interesting directions for future exploration. As demonstrated in our models, almost none of the control variables significantly predicted changes in director compensation (either in the annual retainer or in total compensation). In other analyses not shown, some of these variables such as organizational size did significantly predict the level of compensation. These results suggest a need to develop a more general model of overall director compensation. It appears that the level of director compensation is at least somewhat predicted by salient characteristics of the firm like size, performance, and complexity. However, increases in compensation seem to be affected primarily by characteristics from outside the firm. Future research could help to further explain when internal and external factors are most likely to affect director pay.

In addition, this research considers only a few particular categories of social forces that may affect changes in director compensation. Future research could explore other factors that affect changes in director compensation. Are directors aware of public opinion? Do corporate governance monitoring agencies’ recommendations result in changes to director compensation? Does the increased scrutiny of public boards from institutional investors, the media, and other watchdog groups have effects on director compensation? All of these questions would be potentially fruitful to explore. In addition, it would be useful to potentially examine other possible social referents that might be used as comparison groups including for example, other peer firms.

Another promising area of future research would be to more fully explore the relationship between executive and director compensation increases. Because we measure these pay increases contemporaneously we cannot rule out the possibility that these pay increases are caused by some unmeasured third cause.

In addition, this research used secondary data to show how social comparison processes affected compensation decisions. Future research could explore similar questions using primary data such as surveys or interviews to more fully explore the processes by which directors bring these social references into their discussions of what compensation changes are appropriate.

Finally, our study examines very large U.S. firms. These firms are the most likely to have directors who are also directors of other large public companies and to have high levels of executive compensation. Future research could explore whether the social comparison processes demonstrated here operate in the same way in smaller entrepreneurial firms, or in firms outside of the United States. Such research would help us understand in greater detail how directors in different contexts set their compensation and help to further advance a social perspective on director pay.

Footnotes

Acknowledgements

This article was accepted under the editorship of Deborah E. Rupp.

Notes

References

Adams

J. S.

1965. Inequity in social exchange. In Berkowtiz

(Ed.), Advances in experimental social psychology. New York: Academic Press.

Allison

P. D.

1990. Change scores as dependent variables in regression analyses. In Clogg

C. C.

(Ed.), Sociological methodology: 93-114. Oxford, UK: Basil Blackwell.

Barnea

Guedj

2006. “But, mom, all the other kids have one!”—CEO compensation and director networks. Working paper.

Bebchuk

L. A.

Fried

J. M.

2004. Pay without performance: The unfulfilled promise of executive compensation. Cambridge, MA: Harvard University Press.

Becher

D. A.

Campbell

T. L.

II Frye

M. B.

2005. Incentive compensation for bank directors: The impact of deregulation. Journal of Business, 78: 1753-1777.

Belliveau

M. A.

O’Reilly

C. A.

III Wade

J. B.

1996. Social capital at the top: Effects of social similarity and status on CEO compensation. Academy of Management Journal, 39: 1568-1593.

Bergh

D. D.

Fairbank

J. F.

2002. Measuring and testing change in strategic management research. Strategic Management Journal, 23: 359-366.

Bizjak

J. M.

Lemmon

M. L.

Naveen

2008. Does the use of peer groups contribute to higher pay and less efficient compensation? Journal of Financial Economics, 90: 152-168.

Black

Scholes

1973. The pricing of options and corporate liabilities. Journal of Political Economy, 81: 637-654.

10.

Bloom

1999. The performance effects of pay dispersion on individuals and organizations. Academy of Management Journal, 42: 25-40.

11.

Boivie

Lange

McDonald

M. L.

Westphal

J. D.

2011. Me or we: The effects of CEO organizational identification on agency costs. Academy of Management Journal, 54: 551-576.

12.

Boyd

B. K.

1994. Board control and CEO compensation. Strategic Management Journal, 15: 335-344.

13.

Brick

I. E.

Palmon

Wald

J. K.

2006. CEO compensation, director compensation, and firm performance: Evidence of cronyism? Journal of Corporate Finance, 12: 403-423.

14.

Burt

R. S.

1987. Social contagion and innovation: Cohesion versus structural equivalence. American Journal of Sociology, 92: 1287-1335.

15.

Carpenter

M. A.

Sanders

W. G.

2004. The effects of top management team pay and firm internationalization on MNC performance. Journal of Management, 30: 509-528.

16.

Carpenter

M. A.

Sanders

W. G.

Gregersen

H. B.

2001. Bundling human capital with organizational context: The impact of international assignment experience on multinational firm performance and CEO pay. Academy of Management Journal, 44: 493-511.

17.

Cohen

S. G.

Bailey

D. E.

1997. What makes teams work: Group effectiveness research from the shop floor to the executive suite. Journal of Management, 23: 239-290.

18.

Colquitt

J. A.

Zapata-Phelan

C. P.

Roberson

Q. M.

2005. Justice in teams: A review of fairness effects in collective contexts. Research in Personnel and Human Resources Management, 24: 53-94.

19.

Conyon

M. J.

Peck

S. I.

Sadler

G. V.

2009. Compensation consultants and executive pay: Evidence from the United States and the United Kingdom. Academy of Management Perspectives, 23: 43-55.

20.

Dalton

D. R.

Daily

C. M.

Ellstrand

A. E.

Johnson

J. L.

1998. Meta-analytic reviews of board composition, leadership structure, and financial performance. Strategic Management Journal, 19: 269-290.

21.

Dalton

D. R.

Daily

C. M.

Johnson

J. L.

Ellstrand

A. E.

1999. Number of directors and financial performance: A meta-analysis. Academy of Management Journal, 42: 674-686.

22.

D’Aveni

R. A.

1990. Top managerial prestige and organizational bankruptcy. Organization Science, 1: 121-142.

23.

Davis

G. F.

1991. Agents without principles? The spread of the poison pill through the intercorporate network. Administrative Science Quarterly, 36: 583-613.

24.

Degoey

2000. Contagious justice: Exploring the social construction of justice in organizations. Research in Organizational Behavior, 22: 51-102.

25.

Edwards

J. R.

1994. Regression analysis as an alternative to difference scores. Journal of Management, 20: 683-689.

26.

Edwards

J. R.

1995. Alternatives to difference scores as dependent variables in the study of congruence in organizational research. Organizational Behavior and Human Decision Processes, 64: 307-324.

27.

Erickson

B. H.

1988. The relational basis of attitudes. In Berkowitz

S. D.

Wellman

(Eds.), Social structures: A network approach: 99-121. Bingley, UK: Emerald.

28.

Fama

E. F.

1980. Agency problems and the theory of the firm. Journal of Political Economy, 88: 288-307.

29.

Fama

E. F.

Jensen

M. C.

1983. Separation of ownership and control. Journal of Law and Economics, 26: 301-325.

30.

Farrell

K. A.

Friesen

G. C.

Hersch

P. L.

2008. How do firms adjust director compensation? Journal of Corporate Finance, 14: 153-162.

31.

Festinger

1954. A theory of social comparison processes. Human Relations, 7: 117-140.

32.

Finkelstein

Hambrick

D. C.

Cannella

A. A.

Jr.

2009. Strategic leadership: Theory and research on executives, top management teams and boards. Oxford, UK: Oxford University Press.

33.

Fredrickson

J. W.

Davis-Blake

Sanders

W. G.

2010. Sharing the wealth: Social comparisons and pay dispersion in the CEO’s top team. Strategic Management Journal, 31: 1031-1053.

34.

Gomez-Mejia

L. R.

Balkin

D. B.

1992. Determinants of faculty pay: An agency theory perspective. Academy of Management Journal, 35: 921-955.

35.

Goodman

P. S.

1974. An examination of referents used in the evaluation of pay. Organizational Behavior and Human Performance, 12: 170-195.

36.

Gouldner

A. W.

1960. The norm of reciprocity: A preliminary statement. American Sociological Review, 25: 161-178.

37.

Graffin

S. D.

Wade

J. B.

Porac

J. F.

McNamee

R. C.

2008. The impact of CEO status diffusion on the economic outcomes of other senior managers. Organization Science, 19: 457-474.

38.

Greenberg

1990. Employee theft as a reaction to underpayment inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75: 561-568.

39.

Greenberg

1993. Stealing in the name of justice: Informational and interpersonal moderators of theft reactions to underpayment inequity. Organizational Behavior and Human Decision Processes, 54: 81-103.

40.

Greene

W. H.

2000. Econometric analysis: (4th ed.). Upper Saddle River, NJ: Prentice Hall.

41.

Guzzo

R. A.

Dickson

M. W.

1996. Teams in organizations: Recent research on performance and effectiveness. Annual Review of Psychology, 47: 307-338.

42.

Hackman

J. R.

1987. The design of work teams. In Lorsch

(Ed.), Handbook of organizational behavior: 315-342. Englewood Cliffs, NJ: Prentice Hall.

43.

Hambrick

D. C.

Werder

A. V.

Zajac

E. J.

2008. New directions in corporate governance research. Organization Science, 19: 381-385.

44.

Harris

Helfat

1997. Specificity of CEO human capital and compensation. Strategic Management Journal, 18: 895-920.

45.

Haunschild

P. R.

1993. Interorganizational initiation: The impact of interlocks on corporate acquisition activity. Administrative Science Quarterly, 38: 564-592.

46.

Haunschild

P. R.

Beckman

C. M.

1998. When do interlocks matter? Alternate sources of information and interlock influence. Administrative Science Quarterly, 43: 815-844.

47.

Hempel

Fay

1994. Outside director compensation and firm performance. Human Resource Management, 33: 111-133.

48.

Henderson

A. D.

Fredrickson

J. W.

1996. Information-processing demands as a determinant of CEO compensation. Academy of Management Journal, 39: 575-606.

49.

Henderson

A. D.

Fredrickson

J. W.

2001. Top management team coordination needs and the CEO pay gap: A competitive tests of economic and behavioral views. Academy of Management Journal, 44: 96-117.

50.

Internal Revenue Service. 2011. SOI tax stats—Individual statistical tables by size of adjusted gross income. Retrieved from http://www.irs.gov/taxstats/indtaxstats/article/0,,id=96981,00.html

51.

Jacquemin

A. P.

Berry

C. H.

1979. Entropy measure of diversification and corporate growth. The Journal of Industrial Economics, 27: 359-369.

52.

Jensen

M. C.

Murphy

K. J.

1990. Performance pay and top-management incentives. Journal of Political Economy, 98: 225-264.

53.

Kosnik

R. D.

1987. Greenmail: A study of board performance in corporate governance. Administrative Science Quarterly, 32: 163-185.

54.

Kosnik

R. D.

1990. Effects of board demography and directors’ incentives on corporate greenmail decisions. Academy of Management Journal, 33: 129-150.

55.

Kulik

C. T.

Ambrose

M. L.

1992. Personal and situational determinants of referent choice. Academy of Management Review, 17: 212-237.

56.

Lan

L. L.

Heracleous

2010. Rethinking agency theory: The view from law. Academy of Management Review, 35: 294-314.

57.

Linck

J. S.

Netter

J. M.

Yang

2009. The effects and unintended consequences of the Sarbanes–Oxley act on the supply and demand for directors. Review of Financial Studies, 22: 3287-3328.

58.

Love

Kraatz

2009. Character, conformity, or the bottom line? How and why downsizing affected corporate reputation. Academy of Management Journal, 52: 314-335.

59.

Lublin

J. S.

2008. As firms flounder, directors quit. Wall Street Journal, November 21.

60.

Main

B. G. M.

O’Reilly

C. A.

III Wade

1993. Top executive pay: Tournament or teamwork? Journal of Labor Economics, 11: 606628.

61.

Main

B. G.

O’Reilly

C. A.

Wade

1995. The CEO, the board of directors and executive compensation: Economic and psychological perspectives. Industrial and Corporate Change, 4: 293-332.

62.

O’Reilly

C. A.

III Main

B. G.

Crystal

G. S.

1988. CEO compensation as tournament and social comparison: A tale of two theories. Administrative Science Quarterly, 33: 257-274.

63.

Pfeffer

Davis-Blake

1990. Determinants of salary dispersion in organizations. Industrial Relations, 29: 38-57.

64.

Porac

J. F.

Wade

J. B.

Pollock

T. G.

1999. Industry categories and the politics of the comparable firm in CEO compensation. Administrative Science Quarterly, 44: 112-144.

65.

Rhodebeck

1981. Group deprivation: An alternative model for explaining collective political action. Micropolitics, 1: 239-267.

66.

Roberson

Q. M.

Colquitt

J. A.

2005. Shared and configural justice: A social network model of justice in teams. Academy of Management Review, 30: 595-607.

67.

Sanders

W. G.

2001. Behavioral responses of CEOs to stock ownership and stock option pay. Academy of Management Journal, 44: 477-492.

68.

Scholl

R. W.

Cooper

E. A.

McKenna

J. F.

1987. Referent selection in determining equity perceptions: Differential effects on behavioral and attitudinal outcomes. Personnel Psychology, 40: 113-124.

69.

Siegel

P. A.

Hambrick

D. C.

2005. Pay disparities within top management groups: Evidence of harmful effects on performance of high-technology firms. Organization Science, 16: 259-274.

70.

Strauss

2011. Company directors see pay cataapult; retainers, stock gains lift compensation. USA Today, October 26.

71.

Tosi

H. L.

Werner

Katz

J. P.

Gomez-Mejia

L. R.

2000. How much does performance matter? A meta-analysis of CEO pay studies. Journal of Management, 26: 301-339.

72.

Tversky

Kahneman

1974. Judgment under uncertainty: Heuristics and biases. Science, 185: 1124-1131.

73.

Vafeas

1999. Determinants of the adoption of director incentive plans. Journal of Accounting, Auditing & Finance, 14: 453-474.

74.

Wade

J. B.

O’Reilly

C. A.

III Pollock

T. G.

2006. Overpaid CEOs and underpaid managers: Fairness and executive compensation. Organization Science, 17: 527-544.

75.

Wade

J. B.

Porac

J. F.

Pollock

T. G.

1997. Worth, words, and the justification of executive pay. Journal of Organizational Behavior, 18: 641-664.

76.

Wade

J. B.

Porac

J. F.

Pollock

T. G.

Graffin

S. D.

2006. The burden of celebrity: The impact of CEO certification contests on CEO pay and performance. Academy of Management Journal, 49: 643-660.

77.

Wallace

Fay

1983. Compensation theory and practice: Kent human resource management series. Boston, MA: Kent.

78.

Westphal

J. D.

1998. Board games: How CEOs adapt to increases in structural board independence from management. Administrative Science Quarterly, 43: 511-538.

79.

Westphal

J. D.

1999. Collaboration in the boardroom: Behavioral and performance consequences of CEO-board social ties. Academy of Management Journal, 42: 7-24.

80.

Westphal

J. D.

Stern

2006. The other pathway to the boardroom: Interpersonal influence behavior as a substitute for elite credentials and majority status in obtaining board appointments. Administrative Science Quarterly, 51: 169-204.

81.

Westphal

J. D.

Stern

2007. Flattery will get you everywhere (especially if you are a male Caucasian): How ingratiation, boardroom behavior, and demographic minority status affect additional board appointments at U.S. companies. Academy of Management Journal, 50: 267-288.

82.

Westphal

J. D.

Zajac

E. J.

1994. Substance and symbolism in CEOs’ long-term incentive plans. Administrative Science Quarterly, 39: 367-390.

83.

Westphal

J. D.

Zajac

E. J.

1995. Who shall govern? CEO/board power, demographic similarity, and new director selection. Administrative Science Quarterly, 40: 60-84.

84.

Yermack

2004. Remuneration, retention, and reputation incentives for outside directors. Journal of Finance, 59: 2281-2308.

85.

Zajac

E. J.

Westphal

J. D.

1995. Accounting for the explanations of CEO compensation: Substance and symbolism. Administrative Science Quarterly, 40: 283-308.