Sense-Making in Compensation Committees: A Cultural Theory Perspective

Preferences

Where do preferences come from? As Thompson, Ellis, and Wildavsky (1990, p. 55) observe:

This is the great unanswered question in the social sciences. Indeed for the most part, it is the great unasked question. Economists take preferences as given, as external to the system being considered. They can, of course, handle the consequences of changes in preferences, but they cannot say anything about where these preferences come from. Political scientists are interested in how people get what they want through political activity. But in both cases, the prior question of why people want what they want is not normally posed. Similarly, decision-making theorists focus on the means used to achieve a given end but rarely ask how people come to prefer that end over some other end.

Research on CCs is not immune to this characterization and has, until now, largely ignored the question of the origin of preferences with regard to compensation policies. Under the agency theory umbrella, most of the research carried out takes for granted that the committee’s preferences are (or should be) essentially motivated by a will to balance the asymmetry between the wants of executives and the interests of shareholders (Perkins & Hendry, 2005). On this basis, studies attempt to measure under which conditions—composition, independence, experience, etc.—committee members are most able to achieve this preferential objective (see for instance Daily et al., 1998; Laux & Laux, 2009). Articulated around institutional theory, another stream of research investigates the strategies mobilized by CCs to legitimize their recommendations and decisions vis-à-vis stakeholders (Berrone & Gomez-Mejia, 2009). According to this perspective, Main et al. (2008, p. 225) find for instance “that concerns with legitimacy push remuneration committees towards an institutional isomorphism in processes and practice”, while Ogden and Watson (2008, p. 711) show how “remuneration committees, concerned about potential stakeholder criticism, devoted considerable time and resources to ensure that their choice of comparators and performance metrics would be regarded as legitimate”.

In spite of the insights they have engendered, these two theoretical perspectives tend to underestimate the question of CC members’ preference formation. Either preferences are taken for granted, or research is confined to the understanding of how they are legitimized. However, in neglecting the construction of preferences, current research on CCs is not very helpful in addressing the issue of soaring compensation, which is commonly viewed as resulting from: (a) unbounded arbitrariness and irresponsibility on the part of CC members; or (b) technical isomorphism, in which members mechanistically apply remuneration schemes viewed as legitimate in director compensation networks but whose outcomes are not necessarily perceived favorably in the wider community. In short, there is a vacuum about why CC members do what they do.

Before addressing this research problem, it should be recognized that the adoption of executive compensation policies now constitutes a central stake for organizations. Not only because it is often claimed that their performance depends on their abilities to recruit and motivate skilled executives (McDonald, Khanna, & Westphal, 2008; McDonald, Westphal, & Graebner, 2008), but also, and perhaps mainly, because various scandals have brought to the fore reputational risks (Power, 2007) tied to the levels of compensation awarded. The work of CCs is therefore more and more understood as being about managing risks—in a context where economic and moral considerations closely intertwine (Harris, 2009).

Under this angle, Mary Douglas’s cultural theory appears particularly appropriate to study CC members’ ways of thinking. One of the main intuitions underlying her theory is the conception of risk management as a socially constructed process involving the distribution of responsibilities and the allocation of blame (Mikes, 2011; Power, 2007). Never fully objective or knowable outside of belief systems and moral positions, risks for Mary Douglas are always embedded in cultures which give them meaning (Lupton, 1999) and legitimize organizational forms and actions (Gephart et al., 2009). As pointed out by Douglas and Wildavsky (1982, p. 5), the construction of dangers and risks fundamentally relates to the notion of preferences: “Risk should be seen as joint product of knowledge about the future and consent about the most desired prospects.” In sum, Douglas’s perspective is well adapted to the examination of the choices and preferences of individuals whose work, in the current context, is increasingly located at the crossroad of risk and morality.

Drawing preferences from social relations

According to Douglas’s cultural theory, preferences are formed from the most basic desire of human beings—how we wish to live with other people and others to live with us. Her theory identifies four primary ways of perceiving, organizing and justifying social relations: individualism, hierarchy, egalitarianism and fatalism. These four ways of life, also called cultural biases, operate in every conceivable domain of social life (Ellis & Thompson, 1997). Theoretically, the fourfold typology is derived from two kinds of societal control: grid and group. ‘Grid’ concerns the amount of classification and rules to which individuals are subject to in the course of their interactions. In other words, grid relates to the extent of autonomy a person has over the selection of roles in their daily life (Linsley & Shrives, 2009). Grid therefore addresses agency and attempts to evaluate the extent to which individuals are free to negotiate their own behavior. ‘Group’ represents the individual’s commitment to their community (Linsley & Shrives, 2009). More specifically, it relates to the amount of moral pressure that a community puts on its members in terms of the claims it makes over them, and the boundaries it draws around them. The notion of group therefore addresses communitarian identity and attempts to evaluate the extent to which the individual belongs to an existing social unit, or conversely, has to forge his or her own social network.

For instance, when admission to the social unit is hard to achieve, the unit is more exclusive and conscious of its boundaries and group strength tends to be high. An extreme case might be the monastic community whose members renounce their private property upon entering and depend on the corporate body for all of their material and social needs. By contrast, group strength is low when people negotiate their way through life on their own—that is to say, when individuals pick and choose with whom they associate, according to their preoccupations and perceived interests. Grid is high whenever roles are distributed on the basis of explicit public classifications, such as gender, age or position in a hierarchy. On the other hand, a low-grid environment is one in which role distribution is not heavily circumscribed, being instead focused on personal abilities to compete and negotiate for positions, and on formal conditions that ensure equal access and opportunity to compete.

Assigning two values (high and low) to these two dimensions, cultural theory identifies four types of cultural biases (see Figure 1): individualism (low grid, low group), hierarchy (high grid, high group), egalitarianism (low grid, high group) and fatalism (high grid, low group). In the paragraphs below, we detail how each cultural bias reflects distinct values and conveys a certain view of the world, particularly regarding attitudes towards risk and blame (Douglas, 1992a).

OPEN IN VIEWER

Figure 1.

The Grid/Group Diagram

Locating and discussing Douglas’s theory

Mary Douglas’s followers have predominately come from the field of public policy studies, seeking to better understand how individuals and groups interpret and address contemporary sociopolitical issues (Ellis & Thompson, 1997; Thompson et al., 1990; Verweij & Thompson, 2006), such as climate change (Verweij et al., 2006), legislation on safety belts (Adams, 2006), or the politics of social housing in Denmark (Jensen, 1999). This dense and varied production contrasts with the rather modest use of Douglas’s work in the organizational analysis literature (Gephart et al., 2009). One of the first notable contributions is that of Altman and Baruch (1998), who rely on two case studies in Israeli organizations to introduce, at a very general level, the basis of grid/group analysis by stressing its explanatory power and originality in comparison to other analytical models, such as those of Miles and Snow (1978) and Hofstede (1991). This work is followed by Lane and Quack (1999) as well as Hendry (1999). Lane and Quack (1999) employ Douglas’s cultural theory to investigate differences in terms of risk management in German versus British banks. They argue that hierarchy and individualism, each shaping attitudes towards risk in distinct ways, are respectively emphasized in German and British risk management templates. Hendry (1999) relies on cultural theory to analyze changes in social structures in the 1980s–90s, in the US/UK business and management arenas. Fundamental to Hendry’s study is the relationship between hierarchy and individualism. He discusses how, particularly in the US, business organizations were then increasingly permeated with an individualistic culture, although the hierarchical culture was exerting a degree of influence.

Arguably, Douglas’s major theoretical contribution is to reorganize the potentially infinite diversity of ways of thinking and doing into a few specific cultural types. In the words of political scientist Harry Eckstein (1997, p. 31), the four cultural biases encapsulate “a great many meanings into a limited set of supermeanings”. However, at the organizational level, the deployment of such broad categories requires some clarification in terms of two key issues, namely, organizational change and cultural pluralism:

Before Douglas’ theory can be applied to the problems of contemporary societies and organizations, two questions arise: Given the self-reinforcing quality of the cultural types, how does the theory account for and accommodate structural change? What happens when different collectivities characterized by different cultural types come into contact and overlap in their memberships? (Hendry, 1999, p. 563)

In this context, the following point developed by Douglas (1997, p. 128) takes on special significance:

A cultural bias is a point of view, with its own framing assumptions and readily available solutions for standardized problems. Scattered persons not in any group at all may share a similar cultural bias.

In other words, cultural groups or factions rarely exist in pure form. In practice, especially in contemporary society, most of people’s lives cannot be neatly partitioned. Membership to multiple collectivities is the norm and it is very common for these to be of different cultural types. The same individual may well live within a predominantly hierarchical society, work within a competitive marketplace and worship as a member of an egalitarian protestant church (Thompson et al., 1990). Thus, Douglas’s cultural theory assumes that, in the field, the four biases of perceiving and organizing interact at each conceivable level of social organization and can account for an endlessly changing and complex social environment. Interaction occurs especially through conversation between individuals, and also through the mind of single actors internally debating how to conduct themselves and make decisions in their daily undertakings.

Since individuals tend to argue from different premises anchored in distinct forms of biases, cultural theory implies logics of contention. However, contention can be “all to the good” as “each way of organizing and perceiving distils certain elements of experience and wisdom that are missed by the others” (Verweij et al., 2006, p. 6). For instance, an individualistic environment can benefit from egalitarian concerns against inequalities, and from the regulatory capacities of hierarchy to enforce property rights and contracts.

Douglas and Wildavsky (1982) also conceptualize the relationship between the different cultural biases through the duality of center and border. They note that Western social thought and organization theory are usually focused on the rivalry opposing hierarchy to individualism (Adler, 2001). Yet in spite of pursuing different ends, hierarchy and individualism share a number of social appreciations— especially an aversion to threats of radical change to the fabric of society. The future is seen as an extension of the present; as a result the two cultures are not significantly sensitive to long-term, low-probability risks. In the words of Douglas and Wildavsky (1982, p. 181), “Market and hierarchy make a formidably stable combination.” While individualism brings innovativeness and entrepreneurship to the center, hierarchy contributes stability. In contrast, the egalitarian ethos aims to criticize the injustices engendered by the dominance of the center, having a vested interest in bad news of impurity and pollution. A different future is expected if the center continues to reign, and this future is expected to be bad. As a result, the egalitarian ethos expects discontinuity between present and future, and its social role can be viewed as that of the “border”:

The center sees the future as continuous with the present. It has numberless ways of discounting information to the contrary. It expects disturbances and setbacks in the normal course, but it also expects to weather them. In the long run it is optimistic. The border foretells imminent disaster. It does not believe in the long run, and in the short run it is pessimistic. (Douglas & Wildavsky, 1982, pp. 122–3)

Expecting discontinuity, the border nonetheless is paradoxically optimistic about the perfectibility of humanity, professing faith in human goodness as the basis for a better society (Douglas & Wildavsky, 1982). The border’s favorite mechanism to promote its critical agenda is the small group, seen as being conducive to equal participation in governance. The small group structure is also iconic, given the philosophy’s opposition to the encompassing institutions of the center.

Moreover, Douglas and Wildavsky (1982) maintain that while the border constitutes a form of permanent opposition to the center, it has no intention of governing, and is not really interested in exercising power. The border is not stable enough to maintain a center. Ironically, the border’s viability depends on the center’s capability; the border needs a stable opposing center against which it can define itself.

In sum, beginning with the premise that cultural biases exert influence on people’s views regarding what to prefer and what to abhor, cultural theory upholds that individuals’ preferences do not arise mysteriously from within the individual (Wildavsky, 1987). Drawing on Douglas’s cultural theory, we propose that a pertinent way to develop a better understanding of CCs is to analyze how CC members perceive and construct executive compensation and their committee’s work, in the light of values and preferences derived from cultural biases. Our research also contributes to the importation of Douglas’s work into the organizational analysis literature—not only at the methodological level, through reliance on individual interviews to study the influence of cultural biases in organizations, but also at the theoretical level, through an analysis of the intertwining between cultural change and the promotion of diversity.

Importantly, cultural biases are integral to social organizations (Douglas & Wildavsky, 1982). When we say, therefore, that some CC members are biased toward stressing executive compensation policies from a specific cultural scheme, we are not saying that there is a totally pure and overarching unbiased line of reasoning, but rather that all stakeholders involved in the construction and sense-making around executive compensation policies—whether CC members, CEOs, consultants, policy-makers, shareholders, journalists, etc.—are inevitably culturally biased and that each bias, as a point of view, needs to be acknowledged (Douglas, 1996).

In terms of the knowledge development process, our research can be understood as belonging to the theory illustration category, in that we seek “to establish the plausibility of a specific theoretical perspective by demonstrating its capacity to illuminate some previously unappreciated aspect” (Keating, 1995, p. 69) of CC members’ ways of thinking and doing—particularly in providing insight into members’ production of moral and rational comfort. As emphasized in the discussion and conclusion, illumination includes the development of reflective views on the very important notions of change and regulation.

The individualistic bias: normalizing the atypical

We begin our empirical journey by examining how the individualistic bias affects CC members’ line of reasoning. Recall that the setting for this individualistic story is a world in which humans are inherently self-seeking and atomistic—but simultaneously creative and innovative. Innovativeness should not be constrained; markets will be able to identify innovations which benefit society, and how to fairly compensate its promoters. Consistent with this characterization, several interviewees maintain that market pressures translate into fair remuneration conditions. In the following quote, one CC member argues that market-based remuneration provides a sense of social justice:

Obviously executive compensation is high. Stockholders are the first to recognize this. However, compensation committee members want to ensure that compensation is fair towards the individual who performs well. To ensure fairness, we need to take into account how other companies in the industry pay their CEO. We want our CEO to have the same remuneration, perhaps even a little higher if we are very satisfied of his performance. So this is how I interpret fairness. Otherwise it would be quite easy to conclude: “It does not make any sense! Two million a year! It’s too high. This guy is not worth it!” (CC 4)

This sense of fairness is basic to low-grid competitive pressures because it extends the individual scope for negotiating. Further, a free contracting environment implies a form of egalitarian distributive justice (Douglas, 1978), concerned with distributing benefits in proportion to the individual’s contribution. Those who make a greater productive contribution to their group deserve to receive more benefits. Those who put in the most get the most out. Politically, the individualistic bias is also visible through the strong opposition of some interviewees to governmental interference in the regulation of compensation policies. For instance:

Governance in and of itself doesn’t make a company successful. Talented CEOs and boards of directors make companies successful. And so I think we’ve gone overboard, and I think we’re going to go overboard again with all this new regulatory talk. Today I was at a luncheon with a prominent democrat and he talked about the new buzz about more regulation. He said, everybody’s focused on regulation. And you’d think he’d be really in favor of it, being a democrat, but he’s not. Because he fears it will be overkill again. SOX [i.e., Sarbanes-Oxley Act of 2002] didn’t prevent the financial challenges that we have now, did it? (CC 24)

This opposition is not just symptomatic of a basic conflict of political preferences, but reflects a deeper sense of historicity in which tradition is continually questioned. In contrast to hierarchic cultural regimes which often rely on historical narratives to justify present actions, individualistic regimes submit all claims of historicity to a competitive screening (Douglas, 1978). Because market conditions are always changing, the past is not a valuable referent to evaluate the present, and comparisons in remuneration levels over time are inappropriate:

What we went through in the financial markets from 1990 until 2000 is just the magic of Disney World. From 1896, it took the Dow Jones 72 years to reach 1,000 points—right before the oil crisis of 1973. It then took 20 years for the Dow Jones to reach 2,000 points, in 1990. This was followed, again, by a crisis. And then, all of a sudden, in 1992, it was just taking several months for the Dow Jones to increase by 1,000 points. Before this frantic movement, stock prices were increasing by 8% or 10% a year. It was taking ten years for stocks’ value to double. This was the environment which historically characterized stock option remuneration. But when companies were giving stock options in the 1990s, the stocks’ value was sometimes tripling in the same year! Lay people then came to interpret the magnitude of option compensation very negatively. The reality is, however, that the magnitude of option compensation is just the result of market trends in stock prices. (CC 3)

Markets are therefore seen as ahistorical. In line with this conception of time, rule-based regulation is seen as paralyzing the vitality of market mechanisms and restraining the scope of individual negotiations. Accordingly, several interviewees maintain that the regulation of compensation can only be provisional because the markets’ future is constantly full of uncertainty. The following quote, where a form of self-regulation is advocated, illustrates this point:

The problem regarding government-based security regulation is lack of adaptability. … But fundamentally speaking, I’m much more positive on the role of organizations which measure companies’ commitment towards corporate governance. The fact is that we all want to perform well in these rankings. Improvement pressure is therefore much more likely to ensue from rankings than from rules dictated by government. Indeed it typically just takes three or four years for people to find ways to bypass government rules. The power to ensure that things are dealt with appropriately resides in the market, through the involvement of its managers and institutional investors. (CC 11)

Douglas’s cultural theory considers that the individualistic setting is characterized by a propensity towards change. However, this inclination does not make CC members susceptible to all “prophecies” announcing potential failures in remuneration policies and means to avoid them. In market regimes, prophets must rhetorically promote their claims like any other people or idea, competitively (Douglas, 1992a); and as also expressed in the former quote, claims which naturalize the incentives of profit maximization seem to constitute more persuasive prophecies than those which aim to bring order through government intervention.

The individualistic bias also provides a “villain” to blame in case of failure. In this respect, CC members were asked not only to make sense of the recent accumulation of highly visible scandals regarding executive compensation, but also to comment on the increasing discrepancy between executive compensation and the average worker’s pay. A first common pattern emerging from several interviews is that failures are attributed to individuals, with the fault primarily residing in atypical “bad apple” executives and CCs:

The main benefit of stock options is “pay as you go”. When stockholders make money they are able to give money to the executives; when they don’t make money they don’t have to give money to the executives. Some atypical cases have been blown out of proportion because people think that these atypical cases constitute the norm. However, I know many CEOs who do not make lots of money through stock options. As a matter of fact, the number of CEOs who do not benefit significantly from stock options is much higher than the number of CEOs who benefit significantly from options. (CC 1)

In 2002 a number of business scandals emerged simultaneously. People then got the impression that all executives were dishonest. I then decided to calculate some proportions to put the scandals in perspective. In North America, 28,000 companies were then listed companies. I added up the number of scandals; the total was 16. I don’t know if you remember the 2004 summer Olympics that took place in Athens—people got the impression from the media that all athletes were taking steroids. Was this true? In all, during these games they tested 3,500 athletes out of 6,000. Twenty-two cases were found to be positive. The same rationale applies to business. Most people in business are actually honest—but this is not what we see in the media, which enjoys making scandalous reports. (CC 3)

While in the first quote experiential claims are mobilized to contextualize the frequency of excessive compensation cases, in the second quote the interviewee’s argument is predicated on quantification and analogical reasoning. Specific figures are invoked, thereby conveying a sense of accuracy and factuality. Significant consequences ensue from the tendency to apportion blame in an individualistic fashion, with a few “bad apples” having infected an otherwise robust system. As collective sense-making is centered on atypicalities, the programmatic and normative aspects of CEO remuneration remain unquestioned. Further, while a number of CC members recognize that market mechanisms exert inflationist pressures on executives’ compensation packages, the blame is not allocated on the market itself. Instead, the interviewees deplore the excessive development of regulation, forcing companies to publicly disclose executive compensation, thereby creating an incontrollable spiral of social isomorphism:

The requirement to disclose public companies’ executive wages has had a pernicious effect. … Why should my compensation be lower than this other CEO given that I’m as good as him? Since our company is the biggest in the industry does our CEO deserve more compensation than the other CEOs? … You know, if you own a hockey team, you want to have good scorers. But to have good scorers you need to pay for good scorers, whose value is determined by the market. If you’re not willing to pay good scorers on the basis of their market value, then none of them will accept to play in your team! Thus, a kind of incontrollable pattern has resulted in significant pay increases in the last few years. (CC 1)

I said at the time that disclosure would drive up compensation. And that’s exactly what it did. And all the advisors and all the companies knew what X got and everybody knew what Y got and there was a huge push to equalize compensation and that meant meeting the higher ones. The very people who wanted all this disclosure are now complaining about what they regard as excessive compensation. … But as I said from the beginning, disclosing it is just going to ratchet it up. (CC 24)

Another common individualistic pattern implies the denial of failure. From this perspective, compensation issues are seen through the vocabulary of demand and supply, thereby reinforcing the imagery of the invisible hand of economics in determining wages and contracts. Through this imagery, claims of excessive remuneration are denied; remuneration is not and cannot be excessive when CC members obey market laws. Blaming is unwarranted since failure is refuted. Accordingly, a variety of justificatory schemes are mobilized by CC members to rationalize high-remuneration cases. For instance, interviewees express the view that inflationist pressures on executive compensation ensue from a rarity of supply—that is to say a scarcity of talents. Therefore, in a context of free market negotiations between executives and CCs, the balance of power tends to be perceived as structurally unfavorable to the latter—hence the inevitability of paying higher wages:

You know, the wage increase trend is a quite complex matter. Inflation plays a role in the process, as well as the shortage of candidates. In each industry and company, people on compensation committees would like to favor new CEOs being chosen from within the ranks of the company. However, in a number of cases good candidates cannot be found within the company. So when you look outside the company’s boundaries, you rapidly find out that there are not hundreds of good candidates available. (CC 4)

Is it difficult to find a CEO who can actually provide huge value added to my business? It’s very difficult. … In 1993 we fired the CEO [of Corporation Z] and we brought in [Name of new CEO]. [Corporation Z] was a second rate company in Canadian life insurance. Now it’s number two in the world. [First name of new CEO] has done this in a 15-year run. Finding a guy like that is next to impossible. To find a Michael Eisner [former CEO of Walt Disney Company] is next to impossible. (CC 25)

As implicitly recognized in these quotes, the perception of a supply shortage is reinforced by the adhesion of interviewees to a glorification of talented executives, virtually magnified as heroes upon which rests much of the companies’ success, and who must be paid in consequence. This representation of executives is also symptomatic of an individualistic bias, where substantive signs of collective realizations are fragmented to the point where the only recognized achievement is individual; collective referents in explaining organizational performance are overshadowed. Along this bias there are no moral limits or social constraints for rewards, as long as the individual at the top delivers on their promises:

There are some really talented CEOs, you know. If you had the main stockholders of [Corporation X] in front of you, they would tell you: “We don’t know any other individual who could have done better for the company than [Name of CEO of Corporation X]. True, he received $500 million in stocks—but the company’s performance over the last 15 years is astounding. Over this time period stockholders benefited every year from a 22% rate of return! No other CEO could have done better. This guy is definitively irreplaceable.” Our company is certainly smaller than [Corporation X], but in terms of performance our current CEO is excellent and it would definitively not be easy to replace him with someone as qualified. (CC 7)

When it’s time to recruit a new CEO no board member on earth is going to say: “Hey, we should select an average individual.” It’s just natural to search for the best individual because companies fundamentally need someone with top capabilities. So when we identify an individual with top capabilities, the remuneration package which is offered to the individual needs to reflect his top capabilities. … When [Corporation Z] buys [Corporation Y], they cannot pay [Name of CEO of Corporation Z] less than [Corporation Y]’s CEO. And [Corporation Z] is a global organization; in order to lead this organization we need the best, one of the best in the world. (CC 11)

Competencies in managing a public corporation are therefore seen as undeniably tied to organizational performance; in so doing, it makes sense to pay a high price in “purchasing” these valued competencies. The latter quotes also indicate the kind of tautological reasoning at work within CC members inclined to individualistic views. As they firmly believe in the market’s capacity to determine the remuneration of the best executives, they cannot allocate compensation that is under the market, unless they admit to recruiting executives who underperform. This line of thought, palpable in several of our interviewees, echoes Ezzamel and Watson’s (1997) “bidding-up” hypothesis, whereby executives who are underpaid relative to the market have their pay raised, but those who are overpaid receive no parallel downward adjustment.

As highlighted above, the dismantling of collective classifications in the name of individual competitiveness is a fundamental and constant justifying point of reference for the individualistic culture. Individuals’ abilities are glorified and markets should not be constrained in providing stockholders with much-sought-after talents, abilities and ideas. Yet paradoxically, this low-grid ethics of individual equality is very often confronted with high inequalities engendered by the competitive environment. The individualistic bias professes equality, but engenders inequality (Douglas, 1978). Based on her anthropological observations, Douglas maintains that low-grid dynamics, despite their claim of adherence to an ethic of equality, never recognize that their own processes of screening out and competitive selection deeply contradict their initial calls for egalitarian individualism. Instead, proponents of low-grid values develop theories to account for the contradiction without demolishing the central ideological edifice at stake:

Rather than dismantle the institutions which protect and promote the competition between individuals, a typical low grid solution to the moral dilemma is to subscribe to a theory of unequally distributed native ability: “Too bad we can’t all be clever…” “You can’t make a silk purse out of a sow’s ear.” And so the pressures to exclude are justified. (Douglas, 1978, pp. 35–36)

During the interviews, we explored CC members’ thoughts regarding the large and increasing disparities between compensation of top executives and that of the average workers. Several interviewees develop arguments consistent with the theory of unequally distributed native ability. In the following excerpt, the interviewee views compensation differences between executives and non-executives as two distinctive worlds which cannot be compared:

Interestingly, what may appear as inconsistency or injustice to other cultural regimes is rationalized by the individualistic culture as morally acceptable considering the jobs are as different as their pay. Markets ensure that people are paid in accordance to their worth and abilities, thereby providing a convenient means to dismiss “inappropriate” comparisons.

Overall, our analysis suggests that the hard core of the individualistic culture is protected from radical doubt with regard to executive compensation by several normalizing strategies which allow stratospheric compensation schemes to be maintained on what is perceived as a moral and rational basis. Inequalities in labor wages are attributed to the individuals’ varying degrees of competencies, while allegations of excessive compensation are rationalized through invocations of the laws of the market. At the end of the day, the individualistic claim that humans are equal citizens is not diminished by the factuality of social inequalities—given that humans are viewed as not being equally endowed with personal qualifications for success.

The hierarchic bias: protocolizing compensation

Our analysis identifies the hierarchic bias as another cultural type significantly influencing sense-making and preferences with respect to executive compensation. When an individual’s social environment is characterized by strong group boundaries and binding prescriptions, the resulting social relations are hierarchical. Individuals in this social context are subject to the control of other members in the group, and the demands of socially imposed roles, rules and protocols (Thompson et al., 1990). Along this bias, boards, and CCs in particular, do not “rubber stamp” the laws of the market or CEOs’ requests; instead CCs assiduously submit compensation claims and requests to the gaze of standard protocols of evaluation. Fairness is assumed to result from compensation being determined and sanctioned by standard operating and control procedures.

As the following quotes indicate, boards and CCs are perceived as powerful instruments of control in the hierarchical chain of corporate command:

I think the whole issue of governance has improved. I don’t think there’s any doubt about it. We all have our doubts about Enron and the legislation that followed, which was heavy-handed and bureaucratic, and to some extent, wasteful and so on. But it put a certain fear into directors and gave them a certain amount of courage. I don’t think there’s any doubt that the pendulum has swung. That’s not a universal truth—but I think on balance that boards are much more willing to recognize their job, which is to govern, to oversee, and to direct the management. (CC 22)

I am absolutely against say-on-pay because the question of management performance evaluation belongs to the board. If shareholders think the board is not doing its job, they should change the board. How can one, in the context of an annual general meeting, adequately assess the CEO’s performance or that of the management team in general? How can one evaluate the strategy, and how it was applied by the CEO? I think that it is the board’s responsibility and it should remain so. (Other 5)

The two excerpts signal from different angles a hierarchic conception of governance, predicated on trust and acceptance of authority ensuing from organizational and regulatory structures (Douglas, 1986). The first excerpt defends a positive view of post-Enron regulation which presumably translates into a reinforcement of the board’s authority. The second manifests hostility regarding shareholders’ interference in board decisions related to executive compensation: say-on-pay disrupts established evaluative mechanisms and authority structures. Altogether, these quotes convey the idea that compliance with rules and established organizational order translates into hierarchic authority being protected “from contamination by the levels below” (Thompson et al., 1990, p. 9). The second excerpt also reveals a sense of fairness based on: (a) the hierarchic delegation of power: “If shareholders think the board is not doing its job, they should change the board”; and (b) the power of expertise: “How can one, in the context of an annual general meeting, adequately assess the CEO’s performance?” This association between order and knowledge is far from unexpected. As observed by Coyle (1997, p. 63), such cohabitation is symptomatic of hierarchic views:

When value agreement and factual certainty are high, the corresponding organizational form (and culture) is hierarchy. In these cases, group identity and role differentiation are strong, and these variables identify the upper right corner of the cultural model. It is knowledge that provides order; when knowledge is complete, the organization is like a fully analyzable machine, held together by standard operating procedure.

Accordingly, being hopeful about the power of human reasoning, one of the fundamental features of hierarchy consists of mobilizing expertise grounded in mechanization. The mechanization of expertise is reflected in our interviews in several ways. For instance, expertise implies adhesion to compensation rules predicated on quantifiable objectives and measurable criteria:

At the beginning of the year, we set objectives. They relate to three components: profit; business growth; and general costs. And the weight of each component is also specified. So we set the objectives at the beginning of the year, and at year-end we say: “Here are the results.” It is mathematical. There is a remaining 15%. In [the case of our CEO] there remains a 15% of qualitative evaluation from the board. So what I do in January I send a little survey to each board member. I think there are eight questions or something like that and I say: “Give him a mark from 1 to 5.” After that, I add them up and we add this 15% to the mathematical calculation of the three other components. (CC 7)

In order to benefit from the long-term compensation plan, the individual needs to beat the industry average. The goal is to exceed by 4% the industry’s ROE [return on equity]. The minimum is 0% in order to get shares. To begin with, there is a three-year cycle, with a potential number of allocated shares. At the end of year three we measure the ROE and compare it to the industry according to a scale from 0 to 8%. If the individual gets 4, then it’s the number of shares determined at the beginning. If it’s 8, then we double. If it’s zero then no shares are allocated. I consider that such a scheme measures the organization’s relative performance. With a scheme like this, it’s clear that ROE will be the number one item upon which people concentrate their efforts. (CC 11)

Further, in the words of CC 2: “The easy part [of a compensation policy] is to apply the recipe. However, the recipe needs to be elaborated beforehand.” Importantly, most of the “recipes” detailed in our interviews rely on quantifiable benchmarking techniques:

This business [i.e., CEO compensation] is one of the most measured businesses on a quantitative basis around so you can set up benchmarks which are calculated as what the average of everybody is doing and you can compare yourself to the benchmarks. And you can set targets which are some percentage equally or exceeding by some percentage and run the thing through on that basis. (CC 18)

As the following excerpt indicates, adhesion to the ideal of measurability and mechanization of compensation policy is not universal and contrasts with the smell test of the individualistic bias: “pragmatic in style of reasoning and casting suspicion on grand theories” (Douglas, 1997, p. 130):

It’s a moving target, there is no basic formula. CEO compensation includes a high degree of value judgment. These are not rules defined on a piece of paper. For a small part, but ultimately not. You know the guy has been performing well for the past twenty years, or five years, and then suddenly he has a bad year: what do you do? Do you say: “I am going to destroy him”? Or do you tell yourself there is a good chance that he will perform exceptionally well in the next three to four years and therefore you forgive him? (CC 12)

The individualistic culture being receptive to compassion fed by anticipation of earnings is noteworthy and, as such, the excerpt illustrates the unpredictable paths and connections that cultures may develop in the field. Nonetheless, the key point we wish to stress is that, contrary to pragmatic and individualistic abilities in developing experiential and intuitive-based anticipations of opportunities, the hierarchic bias tends to promote dispassionate and depersonalized judgments shaped by reliance on standardized protocols and ways of doing. Along this line, the legitimacy granted to CC members does not depend on their stature or individual charisma, but rather on their control of the substantive knowledge required to appropriately execute the committee’s evaluative protocols. The latter are not formalized algorithms that can be operated through deskilled human involvement; instead their application necessitates nimbleness and experience:

Our methods, our systems have become more complicated, you know, with a mixture of salary, incentive bonus, stock options and restricted stock, looking at vesting periods that are longer. It’s become a more complicated task in the last couple of years than it would have been years before. (CC 17)

As a result of this complexity, CC members influenced by the hierarchic bias tend to promote a recruitment strategy within CCs in which strong individual leadership is toned down in favor of human resource expertise:

I think it’s important that we have the same ground rules for compensation as for the audit committee. That is, everybody should be literate and there must be a test of—I don’t mean a written test, but a significant measure of literacy, and one should be expert. Now we don’t have that. In very few compensation committees is there anyone expert. And you know, the reason is, the kind of people that join boards are typically CEOs, former CEOs, former chairs retired whatever, but usually mainstream executives who did not develop that expertise. So I’m saying … that probably we should have more vice-presidents of human resources on boards. (CC 21)

One ought to be looking for something called human resource expertise and human resource literacy on the compensation committee. And in my experience there are very few compensation committees that can adequately hold themselves out to have sufficient degree of human resource expertise and human resource literacy to do their job. (CC 18)

Furthermore, expertise in executing emotionally detached evaluative protocols is often perceived as benefiting from the specialized knowledge of consultants:

[The decision to rely or not to rely on outside expertise] depends on how comfortable the board is with its own expertise. How comfortable the board is with the status quo in terms of the process and the system, or the method of compensation. But if you want to change anything from the existing mix or the existing parameters, or you think you’re out of line in terms of competitive practice, then you might want to bring in [Name of consulting firm]. (CC 21)

Now you could argue that the human resources or the vice-president of human resources ought to have that, too—but I think a consultant sees—in any field of consulting, I guess—sees a very wide range of practices, and has a broader scope of knowledge. (CC 21)

However, the use of consultants is not exclusive to hierarchy. As argued by Abrahamson (1996), business people are generally receptive to techniques that are framed as being at the forefront of management processes. This is likely to be the case under an individualistic environment where the advantage goes to the innovating individuals who can spring a surprise on competitors (Douglas, 1978). As suggested by March and Olsen (1989), in such settings, actors through a “logic of appropriateness” may tend to adopt fashionable best practices in response to perceived uncertainty. Indeed, in some of our interviews, remuneration consultants appear to exert a significant influence over CCs in the exposure and appropriation of the latest compensation policy techniques. For example, in the following quote, the interviewee claims that the company’s CC benefits from consultants’ input in complying with the latest set of best practices:

Well, we’re saying [to the consultant] we need the benefit of your wisdom on this and we need to tap into your creativity because you see a whole lot more plans and schemes and things than we would see and we want you to be able to bring to us the best practices that you see around. Our consultant does most of the really major companies in Canada. I want them to say this is your vanilla formula but if you were to tweak this and tweak that and look at it from this point of view, it would better meet your objectives. (CC 24)

Nonetheless, consultants are often hierarchically mobilized—for example, because of their substantive expertise in helping apply complex evaluative protocols:

It’s become a complex field and there are a lot of tax implications and there are a lot of variables in terms of subsets of options or stock grants and deferred this and deferred that. And the expertise that these folks have developed is pretty significant. (CC 15)

Hierarchy is also characterized by peculiarities regarding the process of allocating failure and blame:

Hierarchists cannot blame the collective. “System blame”, blaming the relationship between the parts and the whole on which they pride themselves, would amount to self-destruction. Instead, hierarchies are famous for their blame-shedding techniques. … Blame is shifted to deviants who do not know their place. (Thompson et al., 1990, p. 59)

Taken into the context of executive compensation, blame in case of failure is allocated to deviant CC members who destabilize the foundations of the hierarchical vision, namely, the authority of the CC and the reliability of its expertise. Accordingly, in the following excerpts, blame is allocated to CCs that are deficient in observing their protocols and authoritative powers with regard to management:

We expect a lot and in a number of cases CEOs don’t work very hard. And I’m still shocked by that and how much money they make. It’s scary, it’s really scary but a lot of them don’t have a whole lot of accountability. They’ve got board members who sit back and go to a meeting and a dinner once a quarter, and are happy that you’re not losing money—so what incentive do they have to do better? (CC 8)

If poor results lead to excessive compensation, it’s because somewhere there is a formula that was deficient in its conception. It’s probably because the compensation committee didn’t read the future correctly, made bad forecasts or basically created a formula thinking everything would turn out right you know. (CC 2)

These two quotes also implicitly reflect a distinctive feature of the hierarchic culture, concerned with ensuring that exuberant behavior never goes too far, that the ball remains within the zone of equilibrium and is formalized “in analytic operational formulae” (Douglas, 1997, p. 130). Unbridled experimentation is therefore negatively perceived. Instead, hierarchic comfort is constructed through the quest for average. The inclination is towards the production of certainty and predictability as generated through formulas being comparatively applied; that which breaks from norms of reasonableness is negatively perceived:

In a world where there is excess on both sides, some people are literally being exploited, we don’t talk about them, but they exist. To find a formula which appears reasonable in all of this is the challenge we face, to find a happy medium. (CC 1)

As observed by Thompson et al. (1990, p. 61): “controlling envy is of paramount importance in hierarchy because its way of life institutionalizes and justifies inequality. Ostentation is reserved for the collectivity.” One of the key dangers for the hierarchic organization is the loss of internal cohesion, not least through claims of excessive inequalities in compensation across hierarchical ranks. The more important the remuneration gap between hierarchical echelons, the higher the risk of individual strategies to promote individual gains—thus threatening internal cohesion and respect of collective order. Inequalities within the hierarchic organization need to be reined in, given that to endure over time, the organization needs to constrain internal rivalries: “its success depends on not allowing one member’s personal glory to be distinguished from the collective honour” (Douglas & Wildavsky, 1982, p. 90). The following excerpt is consistent with this view:

I think fair has to do with treating the whole organization as decently as you can in terms of fair income, fair salaries, fair benefits and so on. I think the relativities within the compensation structure are important. I hate it when I see a CEO earning much more than his direct reports. I think the relativity as you go down the structure should have a certain kind of consistency. So that would be the foundation of fairness. So fairness is probably more a bottom up thing than it is a top down thing. (CC 22)

In short, in the context of CCs, the hierarchic culture emphasizes experts, systems of expertise and protocols as indispensable mechanisms of rationality, which are used to determine morally acceptable compensation boundaries across hierarchical ranks. However, the hierarchic bias can be a victim of its own faith in structured protocols and expertise. On the one hand, by naturalizing benchmark techniques as forms of objective evaluation, hierarchy is not conducive to sensitivity with respect to inflationist pressures. That is, a mechanical focus on the average of the industry is not likely to engender a rigorous questioning of the evolution of the average. On the other hand, driven by a sense of legitimacy derived from the formal duties and powers they possess, CC members may tend to isolate themselves from pressures and criticisms that come from the outside. As observed by Douglas (1992a, p. 66): “One unintended consequence of setting up a successful bureaucracy … is that its viewpoint tends to be insensitive to political outcomes.” Hierarchy can therefore expose the organization to the risk of misperceiving societal expectations regarding executive compensation.