Journal of Management Studies 29:2 March 1992 0022-2380 $3.

50 STAKEHOLDER-AGENCY
THEORY CHARLES W. L. HILL THOMAS M. JONFS school of Btcriness Administration, University of
U'ashington ABSTRACT Taking agency theory. and stakeholder theory as points of departure, this
article proposes a paradigm that helps explain the following: (1) certain aspects of a firm's
strategic behaviour; (2) the structure of managementstakeholder contracts; (3) the form taken by
the institutional structures that monitor and enforce contracts between managers and other
stakeholders; and (4) the evolutionary process that shapes both management-stakeholder
contracts and the institutional structures that police those contracts. INTRODUCTION Over the last
decade, agency theory has emerged as the dominant paradigm in the financial economics
literature (Jensen and Meckling, 1976; Ross, 1973). As developed in that literature, agency theory
has been primarily concerned with the relationship between managers and stockholders.
However, recently authors in the management field have begun to explore the implications that
agency theory might have for the disciplines of organizational behaviour, organizational theory,
and strategic management (e.g., Eisenhardt, 1985, 1988, 1989; Kosnik, 1987). One area that
remains relatively unexplored concerns the ability of agency theory to explain the nature of the
implicit and explicit contractual relationships that exist between a firm's stakeholders. In addition
to managers and stockholders, stakeholders include employees, customers, suppliers, creditors,
communities, and the general public. The agency theory view of the firm as a nexus of contracts
between resource holders (stakeholders) suggests that this may be a promising avenue for
investigation. Taking agency theory and stakeholder theory as points of departure, the purpose of
this article is to propose a paradigm that helps explain the following: (1) certain aspects of a firm's
strategic behaviour; (2) the structure of management-stakeholder contracts; (3) the form taken by
the institutional structures that monitor and enforce contracts between managers and other
stakeholders; and (4) the evolutionary process that shapes both management-stakeholder
contracts and the institutional structures that Address fw rcprinl: Charlcs W. L. Hill, Department of
Management and Organization, School of Business Administration, DJ-lo, University of
Washington, Seattle, Washington 98195, USA. 132 C.W.L. HILL AND T.M. JONES police those
contracts. Like agency theory, this paradigm suggests that the firm can be seen as a nexus of
contracts between resource holders. Unlike agency theory, the paradigm encompasses the implicit
and explicit contractual relationships between all stakeholders. Stated simply, the resultant model
is a generalized theory of agency: one of stakeholde-agency. Although similar to agency theory in
many respects, stakeholder-agency theory is based on assumptions concerning market processes
that are substantially different from those underlying the finance version of agency theory. The
result is a paradigm whose predictions are not always consistent with those of agency theory.
While agency theory operates on the assumption that markets are efficient and adjust quickly to
new circumstances, here the existence of short to medium-run market inefficiencies are admitted.
The result is the introduction of power differentials into the stakeholder-agent equation. Although
the idea of power differentials is at variance with the traditional agency approach, in our view the
approach developed here increases the explanatory power of the paradigm. AGENCY THEORY An
agency relationship is defined as one in which one or more persons (the principal(s)) engages
another person (the agent) to perform some service on their behalf which involves delegating
some decision-making authority to the agent Uensen and Meckling, 1976; Ross, 1973). The
cornerstone of agency theory is the assumption that the interests of principles and agents diverge.
According to agency theory, the principal can limit divergence from hislher interests by
establishing appropriate incentives for the agent, and by incurring monitoring costs designed to
limit opportunistic action by the agent. Further, it may pay the agent to spend resources (bonding
costs) to guarantee that he/she will not take certain actions that would harm the principal, or to
ensure that the principal will be appropriately compensated if helshe does take such action. That
is, the agent may incur ex-ante bonding costs in order to win the right to manage the resources of
the principal. Despite these devices, it is recognized that some divergence between the agent’s
actions and the principal’s interests may remain. Insofar as this divergence reduces the principals’s
welfare, it can be viewed as a residual loss. The sum of the principal’s monitoring expenditures,
the agent’s bonding expenditures, and any remaining residual loss are defined as agency costs.
Further, agency theory asserts that natural selection processes favour governance structures that
economize on agency costs (Fama and Jensen, 1983; Jensen, 1983). By governance structures,
agency theorists mean the mechanisms that police the explicit and implicit contracts between
principals and agents (Demsetz, 1983; Fama, 1980; Fama and Jensen, 1983). These include the
structure of law governing corporate behaviour and its attendant legal apparatus, monitoring
mechanisms (such as the board of directors), and enforcement mechanisms (such as the market
for corporate control and the managerial labour market). Although applied primarily to the
stockholder-manager relationship, Jen- STAKEHOLDER-AGENCY THEORY 133 sen and Meckling
(1976) argue that agency theory ‘will lead to a rich theory of organizations which is now lacking in
economics and the social sciences generally’ (p. 309). Jensen and Meckling view the implicit
contract between stockholders and managers as just one of the nexus of contracts that form the
legal fiction known as the modern corporation. Other contracts that could be considered within an
agency framework include those between managers and the various primary interest groups of
the firm or stakeholders. STAKEHOLDERS The term stakeholders refers to groups of constituents
who have a legitimate claim on the firm (Freeman, 1984; Pearce, 1982). This legitimacy is
established through the existence of an exchange relationship. Stakeholders include stockholders,
creditors, managers, employees, customers, suppliers, local communities, and the general public.
Following March and Simon (1958)) each of these groups can be seen as supplying the firm with
critical resources (contributions) and in exchange each expects its interests to be satisfied (by
inducements). Stockholders provide the firm with capital. In exchange, they expect the firm to
maximize the risk-adjusted return on their investment. Creditors provide the firm with finance and
in exchange expect their loans to be repaid on schedule. Managers and employees provide the
firm with time, skills, and human capital commitments. In exchange, they expect fair income and
adequate working conditions. Customers supply the firm with revenues and expect value for
money in exchange. Suppliers provide the firm with inputs and seek fair prices and dependable
buyers in exchange. Local communities provide the firm with locations, a local infrastructure, and
perhaps favourable tax treatment. In exchange, they expect corporate citizens who enhance
andlor do not damage the quality of life. The general public, as tax payers, provides the firm with a
national infrastructure. In exchange, they expect corporate citizens who enhance andlor do not
damage the quality of life and do not violate the rules of the game established by the public
through their legislative agents. Spec@ Asset Investments Stakeholders differ with respect to the
size of their stake in the firm. The magnitude of an individual actor’s stake is a function of the
extent to which that actor’s exchange relationship with the firm is supported by investments in
specific assets (Williamson, 1984). Following Williamson (1984, 1985), by specific assets we mean
assets that cannot be redeployed to alternative use without a loss of value. For example,
employees with general-purpose skills and knowledge can leave the firm and be replaced without
productive loss to either the worker or the firm (assuming efficient labour markets). In such cases,
their ‘stake’ in the firm is low. Alternatively, employees with skills that are uniquely tailored
(specialized) to the requirements of the firm cannot leave without bearing substantial exit costs in
the form of the lower rent stream that their skills can earn in the next best application. The ‘stake’
of such employees in the firm is high. This distinction is important: compared to 134 C.W.L. HILL
AND T.M. JONES actors with a low stake in the firm, actors with a high stake will demand more
comprehensive incentive mechanisms and governance structures in order to safeguard their asset-
specific investments in the firm. The Unique Role of Management Whatever the magnitude of
their stake, each stakeholder is a part of the nexus of implicit and explicit contracts that
constitutes the firm. However, as a group, managers are unique in this respect because of their
position at the centre of the nexus of contracts. Managers are the only group of stakeholders who
enter into a contractual relationship with all other stakeholders. Managers are also the only group
of stakeholders with direct control over the decision-making apparatus of the firm (although some
stakeholders, and particularly the suppliers of capital, have indirect control). Therefore, it is
incumbent upon managers to make strategic decisions and allocate resources in the manner most
consistent with the claims of the other stakeholder groups. The unique role of managers suggests
that they can be seen as the agents of other stakeholders; hence the term stakeholder-agency
theory. It would be incorrect, however, to suggest that all the other groups of stakeholders are
therefore principals in the sense implied by agency theory. In agency theory, principals hire agents
to perform some service on their behalf. Stockholders and some customers apart, few
stakeholders can be said to hire managers (in the case of employees the reverse is clearly true).
Nevertheless, there is a parallel between the general class of stakeholder-agent relationships and
the principal-agent relationships articulated by agency theory. Both stakeholder-agent and
principal-agent relationships involve an implicit or explicit contract, the purose of which is to try
and reconcile divergent interests. In addition, both relationships are policed by governance
structures. Moreover, many of the concepts and much of the language of agency theory can be
applied to stakeholder-agent relationships. All of this suggests that principal-agent relationships,
as defined by agency theory, can be seen as a subset of the more general class of stakeholder-
agent relationships. UNDERLYING ASSUMPTIONS Our main assumptions concern the efficiency of
the market mechanism. These assumptions have implications for the existence of power
differentials between the parties to a contract. By a power differential, we mean a condition of
unequal debendme between the parties to an exchange (Emerson, 1962; Pfeffer, 1981). That is,
for two entities, A and B, there is a power differential in A’s favour when B depends upon A more
than A depeids upon B. (While this definition is suitable for the purposes of this article, a fuller
discussion of the definitions of power can be found in the work of Lukes (1974) and Wrong ( 1988).
Agency theorists see the firm as surrounded by efficient markets that adjust quickly to new
circumstances (Barney and Ouchi, 1986). They make the rather heroic assumption that markets
are in or near an efficient equilibrium STAKEHOLDER-AGENCY THEORY 135 (Fama, 1980; Fama and
Jensen, 1983; Jensen, 1983). The efficient markets assumption implies that principals and agents
have freedom of entry into and exit from contractual relationships. According to mainstream
agency theorists, if an agent (principal) does not like the terms of a contract offered by a principal
(agent) and/or the governance structures that police that contract, he/she can always seek a
better alternative. If a shortage ot’ agents (principals) results, the principals (agents) will .be
compelled by market forces to adopt more acceptable incentive mechanisms and/or governance
structures. Of course, this argument ignores the obvious fact that a better alternative might not
always be available (for a discussion of the implications, see Burt, 1983). Putting this criticism aside
for a moment, however, the mainstream agency theory argument suggests that contracts
between principals and agents, along with the governance structures that police those contracts,
are determined by market forces. Thus, within the framework of mainstream agency theory, they
must be seen as having efficiency properties. This tidy logic breaks down if the efficient market
assumption is dropped (Perrow, 1986; Putterman, 1984). If the markets that surround the firm are
inefficient, as occurs when alternative contracting opportunities are limited, the existence of
power differentials between principals and agents must be admitted. If agents are unable to exit
from a contractual relationship without taking a substantial loss (because ‘better alternatives’ are
not available), or if the supply of agents exceeds the demand for agents by principals, power shifts
towards the principal. Similarly, if principals are unable to dismiss agents, or if there is a shortage
of agents, power shifts towards the agents. This is important, since power differentials can
materially affect both the content of principal-agent contracts and the structure of governance
mechanisms policing those contracts. Given this, it is important to state our assumptions with
regard to market efficiency and equilibrium. There are two points of disagreement between
stakeholder-agency theory and agency theory. The first concerns the speed with which markets
adjust to new circumstances, while the second concerns the assumption of equilibrium. In contrast
to agency theory, we view market adjustment processes as being characterized by friction
(Williamson ( 1985) makes a similar point and argues that this friction results in transaction costs).
Due to friction, once created disequilibrium conditions may persist for a prolonged period of time
before an efficient equilibrium is re-established. The resulting disequilibrium conditions imply the
existence of power differentials between the parties to an exchange. Sources of Friction Barriers to
entry and exit constitute one source of friction (Porter, 1980). Barriers to entry and exit impede
adjustment processes and may allow power differentials arising from an initial disequilibrium to
persist in a market for significant periods of time. Open systems theory suggests a further source
of friction. Managers and other stakeholders can to a degree shape or enact their environment
(Pfeffer and Salancik, 1978; Weick, 1979). If a disequilibrium situation is perceived as being to their
advantage, managers may be able to slow down the adjustment process by appropriate strategic
investments (c.g., 136 C.W.L. HILL AND T.M. JONES investments designed to increase entry
barriers, by collusion, by predatory pricing, etc.). Organizational inertia may be a further source of
friction (Hannan and Freeman, 1984). As a result ofdisequilibrium, one party to a contract may be
disadvantaged. Correcting the disadvantage may require the innovation of new incentive
structures and/or monitoring and enforcement mechanisms. However, the ability of the
disadvantaged party to innovate may be hindered by strong inertia forces. Due to inertia, it may
be difficult to alter established routines and procedures for monitoring and enforcing
managementstakeholder contracts so as to reflect new realities. Pressures such as sunk costs,
political coalitions, the tendency to consider precedents as normative standards, and a simple lack
of imagination all limit the degree to which incentive, monitoring, and enforcement structures
respond quickly to new circumstances. Having said this, in the long run we believe that market
processes work to select out the most inefficient organizational forms. Despite barriers to entry
and exit, attempts by managers to shape the environment to their advantage, and inertia, in the
long run there are grounds for believing that the market system does achieve a rough balance
between the efficient and inefficient (Alchain, 1950; Nelson and Winter, 1982). Although the
adjustment process can be slowed down, it cannot be halted altogether. While we certainly do not
agree that only the most efficient survive, we think it is highly probable that the most inefficient
organizational forms lose ground and are eventually selected out. Equilibrium and Market
Processes If change was a rare event, the above arguments would imply that equilibrium
situations in which the most inefficient organizational forms have been selected out are
commonplace. However, one of the central features of the real world is that the only constant is
change. Although market process work towards some kind of equilibrium, change constantly alters
the direction in which that equilibrium is to be found. As argued by Schumpeter (1942), such
change often occurs due to the process of creative destruction triggered by innovation. Moreover,
Schumpeter suggested that innovation is itself a product of the competitive process. Thus, ongoing
change may be a persistent endogenous feature of capitalism. Alternatively, change may be due to
exogenous macro-environmental trends (demographics, social-political Factors, macro-economic
change, etc.). No matter how it arises, ongoing change creates a situation of permanent
disequilibrium and hence, of persistent power differentials between stakeholders and managers.
However, because of the random nature of change, power differentials are themselves unlikely to
remain unidirectional. While change at one point in time may favour managers, change in a
subsequent period may shift the balance of power towards other stakeholder groups. Building on
this perspective, we follow modern ‘Austrian’ economists in arguing that the focus of theoretical
attention should be on market processes rather than equilibrium conditions in efficient markets,
since the latter is little more than a convenient fiction (Kirzner, 1979; Knight, 1921; Littlechild and
STAKEHOLDER-AGENCY THEORY 137 Owen, 1980; Nelson and Winter, 1982; Schumpeter, 1942).
While a drive towards efficiency may characterize the business system, in the sense that the most
inefficient producers ultimately get selected out, we view short- and medium-term inefficiencies
arising out of disequilibrium conditions as commonplace.['] This suggests that power differentials
arising out of disequilibrium conditions between managers and other stakeholders are an essential
determinant of the nature of many stakeholder-agent contracts and the structures that police
those contracts. In sum, our view of market dynamics is fundamentally different from that which
characterizes most of the agent-principal literature. We do not assume equilibrium, although we
do assume that market processes work in such a manner that, in the long run, inefficient incentive
structures and monitoring and enforcement mechanisms are selected out, while more efficient
structures and mechanisms evolve to replace them. However, due to barriers to entry and exit, the
ability of stakeholders and managers to enact their environment, and inertia, the adjustment
process is plagued by significant frictions. By forcing attention onto adjustment processes under
disequilibrium conditions, this perspective adds richness to the discussion of stakeholder-agency
theory. DIVERGENT CLAIMS, UTILITY LOSS, AND CONTRACTING COSTS The interests of principals
and agents diverge primarily because these different groups have different utility functions. In
turn, this can lead to direct conflict over the use to which resources are put (for example, see
Jensen, 1986). Agency theory focuses on the divergence of interests between managers and
stockholders. The argument can be traced back to the managerial discretion literature of the
1960s (Baumol, 1959; Marris, 1964; Williamson, 1964). This literature theorizes that stockholders
are wealth maximizers, while managers maximize a utility function that includes remuneration,
power, job security, and status as its central elements. The agency/ managerial literature
postulates that satisQing the claims of stockholders involves maximizing the efficiency of the firm
(Fama, 1980), while satisfying the claims of management requires increasing the size of the firm
(remuneration, power, job security and status are argued to be a function of firm size). In turn, it
has been argued that the desire to increase firm size results in a managerial preference for
maximizing the growth rate of the firm, principally through diversification (Amihud and Lev, 1981;
Aoki, 1984; Marris, 1964). Further, the discretion literature postulates a trade-off between growth
maximization and efficiency maximization. Beyond a certain point, the greater investments in
growth, the lower investments in maximizing efficiency. Thus, divergent claims give rise to an
agency conflict between managers and stockholders. Stakeholder-agency theory postulates that
other stakeholder groups also place claims on the firm that, if satisfied, reduce the amount of
resources that management can channel towards the pursuit of growth through diversification.
Satisfying employee claims for higher wages, consumer claims for greater quality and/or lower
prices, supplier claims for higher prices and 138 C.W.L. HILL AKD T.M. JONES more stable ordering
patterns, and the claims of local communities and the general public for lower pollution and an
enhanced quality of life, all involve the use of resources that might otherwise be invested by
managers in maximizing the growth rate of the firm. Thus, an agency conflict is inherent in the
relationship between management and all other stockholders (for a transaction cost interpretation
of this phenomenon see Williamson, 1985). This is not to deny that to a degree the claims of
stakeholders and managers also converge. For example, satisfjling employee claims for higher
wages and better working conditions may improve employee productivity and thus provide
management with greater resources. Similarly, devoting resources to controlling pollution may
result in local communities being more receptive to future proposals by management for
expanding its operations. However, our contention is that beyond a certain point, this
convergence of interest is replaced by divergence. If uncorrected, the divergence between
management and stakeholder preferences with regard to the way in which a firm allocates its
resources will result in a failure of stakeholders to maximize their utility. The difference between
the utility that stakeholders could achieve if management acted in stakeholders’ best interests,
and the utility that is achieved if management acts in its best interest, can be referred to as a
utility Loss. In the absence of incentive, monitoring, and enforcement structures that serve to align
the interests of managers and stakeholders, utility loss may be substantial. The function of
incentive, monitoring, and enforcement structures is to minimize utility loss by correcting for the
divergence of interests between management and stakeholders. The concept of utility loss leads
to a somewhat broader definition of agency costs than that typically given in the agency literature.
To distinguish this from the agency definition, from now on we shall talk in terms of contructing
costs. These are defined as the reduction in utility that stakeholders bear by channelling resources
to support incentive, monitoring, and enforcement structures, as opposed to using those
resources directly to satisfy their utility function, plus any remaining or residual utility loss. For
example, imagine that the maximum amount of utility that stakeholders can derive from a given
relationship is 100 units, but that management preferences result in stakeholders only getting 60
units, resulting in a total utility loss of 40 units. If stakeholders devote resources equal to 10 units
of utility to establishing incentive, monitoring, and enforcement mechanisms, they may increase
the utility they derive from the relationship to 90 units, resulting in a net gain of 20 units. The
remaining residual utility loss is 10 units. Thus, contracting costs are equal to the 10 units of utility
that are sacrificed to support incentive, monitoring, and enforcement mechanisms, plus the
residual utility loss of 10 units. INTEREST ALIGNMENT MECHANISMS One way of minimizing the
utility loss that arises from a divergence of interests involves introducing ex-ante interest
alignment mechanisms into the STAKEHOLDER-AGENCY THEORY 139 contracting scheme. In the
agency literature, management and employee stock option plans are the most widely discussed of
these mechanisms (Demsetz, 1983). Stock option plans serve to induce managers and employees
to pay more attention to maximizing stockholder wealth, since that will simultaneously mazimize
their own wealth. On a more general level, offering tax breaks for investments in pollution
containment equipment is an example of how local communities and the general public (through
their legislative agents) use incentives to try and align management interests with their own. In
addition, to gain access to their resources, stakeholders may demand that managers absorb ex-
ante bonding costs in order to demonstrate their commitment to satisfying stakeholder interests.
For example, consider the consumer contemplating entering an exchange relationship with a
manufacturer of consumer durables. The purchase of durables presents consumers with a difficult
agency problem. Consumer durables are purchased infrequently and involve large expenditure. In
such circumstances, the consumer is vulnerable to opportunistic action on the part of
management. Management may misrepresent the quality or durability of the product in an
attempt to close the sale. The agency problem is solved by the ex-ante introduction of a warranty
into the contracting scheme. This specifies management’s obligation to correct defects or provide
suitable compensation in the event of substandard quality. The warranty is a bonding mechanism
that communicates to consumers a commitment on the part of management to a certain standard
of quality. More generally, a bonding mechanism is an example of the use of credible
commitments (Williamson, 1985). Establishing a credible commitment requires that managers
post a ‘hostage’ or bond forfeitable upon malperformance (Alchain and Woodward, 1988). The
concept has been used to explain certain characteristics of a firm’s relationships with its suppliers
and consumers (although it is hardly limited to this context). For example, when a supplier has to
make substantial investments in specialized assets in order to enter into trade with the firm, it is
also exposing itself to the possibility of opportunistic abuse by management (Williamson, 1985).
Once the supplier has made the investment, it is effectively ‘locked in’ to the relationship and
cannot exit without reducing the value of those assets. Management may use this fact to go back
on any ex-ante agreement and drive down the prices charged by the supplier. As insurance against
this possibility, the supplier may demand a similar ex-ante investment in dedicated assets on the
part of the firm. This locks both parties into a mutually dependent relationship in which power is
symmetrically distributed. Examples include reciprocal trade agreements, most-favoured-buyer
clauses, inflexible prices, posted prices, exclusive territories, franchise-specific investments, patent
pools, and union shop agreements (Williamson, 1985). In all these cases, the underlying objective
is to establish mutual dependency between managers and other stakeholder groups so that
interests are more closely aligned. As a general rule, the use of credible commitments to bond
managers and stakeholders will be greater the greater the investments in specialized assets
required of either stakeholders or managers to support a given exchange relationship. 140 C.W.L.
HILL AND T.M. JONES MONITORING AND ENFORCEiMENT MECHANISMS AND STRUCTURES
Interest alignment mechanisms apart, the contracts between stakeholders and managers are
primarily implicit (Mitroff, 1983). Stakeholders supply the firm with resources on the implicit (tacit)
understanding that their claims on the organization will be recognized. To ensure that this occurs,
a number of institutional structures have evolved that serve the function of monitoring and
enforcing the terms of implicit contracts. Agency theory generally refers to such institutions as
governance structures. Our change of terminology reflects a broadening of emphasis. Specifically,
we are concerned with more than just quasi-independent or third party governance (such as the
board of directors, the market for corporate control, or the legal superstructure of society). We
are also concerned with institutions that have evolved to represent and further the interests of a
given set of stakeholders (such as labour unions and consumer unions) precisely because such
institutions have utility lossminimizing properties. Thus, the term institutional structures subsumes
the term governance structures. Monitoring Structures An information asymmetry exists between
managers and stakeholders. As insiders, managers are in a position to filter or distort the
information that they release to other stakeholders. Management control over critical information
complicates the agency problem. It makes it difficult for stakeholders to identify if management is
acting in their interests. The obvious response is for stakeholders to gather more information
about management activities. However, while individual stakeholders can and do undertake their
own monitoring of management performance, the costs of gathering and analysing additional
information may be prohibitive. This is particularly likely when stakeholders are diffused. Diffusion
refers to a situation where a stakeholder group contains many individuals or entities, no one of
which has command over a significant proportion of the group’s total resources. In such
circumstances, ceteris paribus, no one individual or entity may be able to finance the extensive
information-gathering and analysis necessary to reduce significantly the information asymmetry
between managers and stakeholders. In turn, this gives managers greater discretionary control
over the use to which the firm’s resources are put, increasing the residual loss that stakeholders
have to bear. The response to the monitoring problem has been the evolution of a wide range of
institutional structures that serve to economize upon the costs of information-gathering and
analysis. Some of these structures are enshrined in legislation (e.g., the requirement that public
companies publish consolidated annual accounts). Other institutions have evolved in an attempt
to exploit the profit opportunities of gathering, analysing, and then selling information to
stakeholders (e.g., stock analysts’ services, consumer reports, etc.). Still others have arisen as non-
profit organizations that exist in part to monitor the degree to which managers act in the best
interests of certain stakeholder groups (e.g., Consumer Watch, Infact, labour unions). The
common theme found in all of these structures is their ability to achieve economies of scale in
information- STAKEHOLDER-AGENCY THEORY 141 gathering and analysis, primarily through the
employment of specialists. The consequence of such devices is a reduction in utility loss.
Enforcement Mechanisms and Structures The function of enforcement is one of deterrence.
Enforcement mechanisms are articulated by stakeholders prior to any resource exchange in an
attempt to deter management from maximizing its utility at the expense of stakeholders. The
success of enforcement mechanisms depends upon their credibility (Schelling, 1960), and those
lacking credibility will be ignored by management. In such circumstances, any attempt to put
enforcement mechanisms into effect will involve costs that outweigh the benefits of reducing the
utility loss from management opportunism. In short, mechanisms that are not effective deterrents
will fail (as do laws that are commonly ignored by the general population). Law as a deterrent.
Establishing credible deterrents in the context of stakeholder-management relationships requires
enforcement mechanisms that are supported by a broad consensus of stakeholders, and, which
are effectively communicated to management ex-ante. Certain legal penalties have this character
(laws against insider trading, antitrust regulations, pollution regulations, etc.). Indeed, it can be
argued that much of the structure of law relating to business activity in society reflects critical
points of conflict in stakeholder-agent relationships. That is, legislators, as representatives of
certain stakeholder interests, have enacted into law enforcement mechanisms that, because they
are credible deterrents, serve to economize on utility loss. Exit as a deterrent. The legal approach
to resolving principal-agent conflicts constitutes only one way of establishing a credible threat. A
more general approach involves the establishment of a credible threat to withhold resources from
the firm if management fails to serve stakeholder interests. That is, to threaten exit from the
exchange relationship (Hirschman, 1970). Such threats may be more effective than legal penalties.
Only in rare situations are legal penalties likely to jeopardize the survival of the firm. Indeed, many
firms view such penalties as a ‘normal cost of doing business’. In contrast, by denying the firm
access to critical resources, stakeholders can threaten its very survival (Pfeffer and Salancik, 1978).
In a sense, the threat of exit is an underlying theme of many stakeholdermanagement
relationships. For example, if dissatisfied with product quality, consumers can always take their
business elsewhere. Similarly, if dissatisfied with a firm’s performance, stockholders can always
sell their stock. Thus, credible threats to exit can be enacted through the market mechanism.
However, market action suffers from a number of weaknesses. First, there is a co-ordination
problem among diffused stakeholders that in certain circumstances makes collective action
problematic. Exit may not be a very effective deterrent if members of a stakeholder group are
unable to act in unison to impose demands on management. For example, while employees may
be unhappy about working conditions, individual complaints or threats of exit may do little to
persuade management to improve conditions, particularly if 142 C.W.L. HILL AND T.M. JONES
there is a ready supply of replacement labour. Similarly, while consumers as individuals may
disapprove of the pollution implications of a given product (e.g. auto exhaust, plastic containers,
air conditioning fluid) the threat of individual exit may be futile, particularly if no cost-effective
alternative exists. Thus, consumers may continue to purchase the products, even though as
individuals they are unhappy about the implications of doing so, and would prefer the firm to
devote resources to developing less harmful alternatives. The institutional response to the
problem of achieving collective action among diffused stakeholders has been the evolution of a
number of structures that perform the co-ordination function. Examples include labour unions,
consumer unions, and special-interest groups. By providing centralized direction, these structures
economize upon the costs of co-ordination and establish the credibility of the exit mechanism.
Thus, labour unions may initiate a strike if management fails to meet their demands for better
working conditions. Similarly, special-interest groups may initiate a consumer boycott if the firm
continues to produce products that they consider to be harmful (e.g., Infact’s consumer-led
boycott of Nestle’s was designed to halt the company’s questionable infant formula marketing
practices in Third World countries). A more intractable weakness of market action is that it may
lack effectiveness in those situations where stakeholders are ‘locked in’ to an exchange
relationship by specific asset investments. Suppliers, customers, employees, or communities who
have invested in specialized assets in order to enter into an exchange relationship with the firm
may not be able to exit without incurring substantial exit costs. The exit costs consist of the
reduced rents from specialized assets that can be earned in their next best application. Other
things being equal, such barriers to exit reduce the credibility of any threat to exit as a contract
enforcement mechanism. This is serious given that actors who make specific asset investments in
the firm are by definition among the most important of its stakeholders (their future is most
closely aligned to that of the firm). However, certain bonding mechanisms have the additional
character of increasing the credibility of the threat to exit among stakeholders who have invested
in firm-specific assets. For example, a union shop agreement can be viewed as a bonding
mechanism by which management agrees to hire only union labour as a means of safeguarding
employees investments’ in firmspecific human capital. This bonding mechanism limits
management’s ability to abrogate any previously agreed labour contract. If they do, they face the
possibility of a strike (exit), the threat of which is made credible by the inability to hire non-union
labour. Notwithstanding such examples, however, the threat of exit may be limited in such
circumstances, in which case stakeholders may have to resort to voice as an enforcement
mechanism (Hirschman, 1970). Voice as a deterrent. In certain circumstances voice may be the
most effective enforcement mechanism. Voice is often the least costly mechanism to adopt.
Newsworthy publicity comes cheap, yet it can severely damage managerial reputations and the
intrinsic value of a manager’s human capital. To be effective, however, voice must be articulated
by interest groups that have a STAKEHOLDER-AGENCY THEORY 143 legitimate claim to represent
stakeholder interests. Again, certain institutional structures such as labour unions, consumer
unions, and special-interest groups arguably have this characteristic. This reinforces our earlier
conclusion that interest groups can be viewed as institutional structures that have evolved to
economize on contracting costs. STATIC EQUILIBRIUM In our view, due to the pervasive nature of
change, much of the business system is in a state of almost permanent disequilibrium. Despite
this, there is value in discussing what we would expect to find if the business system were ever to
achieve equilibrium. Although this is something of an abstract and teleological exercise, such a
discussion tells us something about the end towards which dynamic processes propel the system.
Here we discuss the factors determining the complexity of the institutional structures that we
would expect to find in an equilibrium situation; later we focus on disequilibrium. A Static Model If
equilibrium were ever reached, institutional structures would display efficiency properties.
Specifically, stakeholders would increase the complexity of institutional structures up to that point
where the marginal benefits of doing so (in terms of a reduction in utility loss) were equivalent to
the marginal costs of maintaining those structures (in terms of the utility that has to be sacrificed
to support them). Given this, it is probable that in equilibrium managers still retain some
(diminished) discretionary control over the use to which the firm’s resources are put. The
argument is explained with reference to figure 1. The horizontal axis of figure 1 measures the
complexity of existing institutional structures. The least complex structure is that of the market
mechanism. More complex structures involve increasingly extensive monitoring and enforcement
mechanisms. Thus, consumer watchdogs such as Ralph Nader’s Consumer Watch, or the
development of labour unions, can be seen as adding complexity to the institutional structures
that police the managementstakeholder interface. The vertical axis measures units of utility. A
positive relationship between the complexity of available institutional structures and the costs of
those structures (in terms of the utility that has to be sacrificed to support them) can be
postulated. If working efficiently, the market system, because it is a decentralized mechanism,
imposes the lowest costs on stakeholders. More complex structures impose additional costs. For
example, ultimately consumers underwrite Consumer Watch through donations. Employees
underwrite labour unions through subscriptions. Similarly, if stakeholder pressures result in certain
regulations being enacted into law, ultimately stakeholders, as taxpayers, underwrite the
commensurate increase in legal apparatus. However, due to the benefits of specialization it seems
likely that economies of scale in information-gathering and analysis exist. Thus, initially the costs
144 t Utility C.W.L. HILL AND T.M. JONES 0 CI Complexity of institutional SfTUCtUTCS Figure 1 (in
terms of utility) of maintaining institutional structures will increase at a decreasing rate with
increasing complexity. This is illustrated in figure 1 where the cost function increases at a
decreasing rate up to the point of inflection a. Past a diminishing returns to specialization are likely
to set in and costs will increase at an increasing rate. The benefits to stakeholders of maintaining
institutional structures can be measured in terms of the reduction in utility loss that such
structures achieve. The benefit function in figure 1 is shown to increase at a decreasing rate,
symbolizing decreasng returns to increasingly complex structures; that is, increasing management
resistance to reductions in their discretionary control over the use to which resources are put.
Eventually, the function will approach the line bb', where 0-6 symbolizes the total utility loss
arising from an ex-ante divergence of interests. The equilibrium condition in figure 1 involves the
stakeholders devoting resources to increasingly complex institutional structures up to that point
where the marginal benefits of such expenditures are equivalent to the marginal costs. It is
worthwhile for stakeholders to bear the costs of establishing and running institutional structures
of 0-C1 complexity.['] Note, this equilibrium point involves a reduction in total utility loss of 0-c.
The remaining utility loss is equivalent to 6-6. Thus, c-b represents the resources still under the
discretionary control of management once the claims of stakeholders have been satisfied. The
logic of our earlier arguments suggests that these resources will be devoted to investments in
maximizing the growth rate of the firm. Another way of viewing c-b is as a measure of the
incentive stakeholders have to develop more efficient institutional mechanisms. More precisely,
the 145 STAKEHOLDER-AGENCY THEORY gross returns to innovation are equivalent to the
discounted present value of &- b,c where the subscript t refers to successive time periods. For C1
to represent a true equilibrium, the perceived returns to innovation must be equivalent to the
perceived costs of innovation. The costs of innovation refer to the costs of overcoming resistance
to change (in terms of the utility that must be sacrificed) and imposing new institutional structures
upon the implicit or explicit contract. An example of these costs might be the costs in terms of
both money and emotion to employees of supporting a strike to get their labour contract with
management renegotiated. If the perceived returns to innovation are greater than the foreseeable
costs, it will pay stakeholders to devote resources to the development of more efficient
institutional structures. The implications of this point are developed later. Extensions A
shortcoming of this model is that it glosses over the problems created by the conflicting claims of
different stakeholder groups. Obviously, the claims of different groups may conflict (e.g.,
stockholder demands for greater dividends conflict with employee demands for higher wages).
However, on a more general level, each group can be seen as having a stake in the continued
existence of the firm.[31 Where opinions differ between stakeholder groups is on how the firm’s
resources should be allocated between investments, and the most desirable time pattern of
organizational rent streams. If the different stakeholder groups engage in open conflict over this
issue, the net effect may well be to damage the firm and all involved with it (as when employees
go on strike or consumers boycott its products). Thus, different stakeholder groups have an
incentive to co-operate, rather than incur the costs of open conflict (for a theoretical discussion of
this see Aoki, 1984). An equilibrium solution to this type of problem can be found in the literature
on co-operative game theory. Although beyond the scope of this article, it should be noted that it
is possible to model what has been referred to as an ‘organizational equilibrium’ (Aoki, 1984). This
is a state in which no one group of stakeholders can increase its utility without risking a higher
expected loss of utility owing to the possible withdrawal of co-operation by the other
stakeholders. A rational stakeholder would not disturb such a state by making a demand for
greater control over how the firm’s resources are invested (for a theoretical proof of this
argument see Aoki, 1984). With reference to institutional structures, the implication of such an
organizational equilibrium is that each stakeholder group will adopt increasingly complex
structures up to the point that is consistent with the cooperative solution. That is, no one group
will attempt to establish additional institutional structures if doing so would upset the
organizational equilibrium and precipitate open conflict between stakeholders. Management’s
role in this process is one of an interest mediator. Management is assigned the difficult task of
balancing conflicting demand so as to achieve a co-operative solution. Management is hardly a
passive player, however. Management can be viewed as trying to expand its bargaining position
with respect to different stakeholder groups. Under the restrictive conditions of neoclassical
equilibrium, such an exercise would be fruitless. 146 C.W.L. HILL AND T.M. JONES However, an
Austrian perspective of the market process leads to a very different conclusion. POWER
DIFFERENTIALS AND MARKET PROCESS It was argued earlier that due to the pervasiveness of
change, extensive disequilibrium is the norm. Moreover, although we view markets as being
ultimately efficient, we theorized that the adjustment process is characterized by considerable
friction due to inertia, the ability of managers and stakeholders to slow down adjustment by their
strategic investments, and entry and exit barriers. The implication of prolonged disequilibrium is
that in practice, power differentials arising from a condition of dependency between principals
and agents are commonplace and may persist for some time. The advantaged party may use such
differentials to further entrench its position and modify institutional structures to its advantage. Of
course, power differentials do not always work to management’s advantage. For example, labour
shortages arising from unanticipated macroenvironmental change will increase the bargaining
power of employees relative to managers, enabling them to impose tighter constraints on
managers (e.g., to demand higher wages, better working conditions, more extensive grievance
procedures, and employee directors). Often, however, power differentials will be in management’s
favour. Moreover, by virtue of their position at the nexus of the implicit and explicit contracts that
constitute the firm, and because of their control over the decision-making apparatus of the firm,
managers may be better positioned to exploit power differentials than individual groups of
stakeholders. Thus, in the remainder of this section we will focus on management strategies for
establishing and/or exploiting power differentials, the implications of power differentials for
institutional structures, and stakeholder responses to management actions. Establishing and
Exploiting Power Dijjferentials Starting with the convenient fiction that in the ‘beginning there was
an efficient equilibrium’, disequilibrium can be seen as either the product of a firm’s own
innovative efforts, or the result of an exogenous shock. However created, management may try to
take advantage of the resulting turbulence and uncertainty to engineer a situation in which the
firm’s stakeholders are more dependent on management than management is upon them. This
involves undertaking strategic actions that reduce the concentration of stakeholder power and/or
increase the concentration of management power. The concentration of stakeholder power can be
reduced by strategies designed to dffue the control over critical resources exercised by
stakeholder groups. For example, with reference to stockholders, targeted stock buybacks along
with new stock issues may be used to reduce ownership concentration and increase shareholder
dispersion. Dispersion makes it more difficult for stockholders to monitor and enforce their
implicit contract with management (Berle and Means, 1932). In a similar vein, management may
diffuse supplier power by developing alternative sources of supply (assuming that alternatives
STAKEHOLDER-AGENCY THEORY 147 are available). Management may reduce customer power by
building a more diverse customer base through product and market diversification. Management
may limit the power of local. communities and the general public by both national and
multinational diversification. And finally, it has been argued that the way in which management
has organized production in the workplace and has exercised control through bureaucratic
mechanisms has significantly reduced the power of employees to oppose management policies
(Braverman, 1974; Clawson, 1980; Edwards, 1979). Increasing the concentration of management
power requires strategies that increase the amount of resources under management control.
These include horizontal mergers and acquisitions to increase concentration within an industry,
vertical integration to gain power over suppliers and customers, and co-operative agreements
between the managers of different firms including joint ventures, interlocking directorates,
purchasing alliances, and price leadership agreements (Pfeffer and Salancik, 1978). The common
theme underlying these strategies is that they restrict the choice set of stakeholders, thereby
altering the configuration of resource dependencies. For example, horizontal acquisitions increase
the buying power of the firm by limiting the number of independent customers to whom suppliers
can sell. All of these strategies are undertaken to increase management power rather than
maximize efficiency. Their ultimate objective is to loosen the constraints imposed by stakeholders
and give management greater discretionary control over the firm’s resources. Without a
commensurate increase in productive efficiency, the additional bureaucratic costs of running an
expanded organization or of achieving intra-organizational co-ordination imply that declining
efficiency will be one result of such strategies. Thus, in an efficient market, firms that pursue such
strategies will be selected out by the competitive mechanism. However, the view of competitive
dynamics advocated here suggests that disequilibrium gives managers the opportunity to build
such power differentials. Of course, it is possible that the ability of managers to pursue strategies
that increase management power over one group of stakeholders may be limited by the
constraints imposed by other stakeholder groups. Most significantly, the board of directors (as the
representative of stockholders), is in theory well positioned to limit managerial actions that it
perceives as being contrary to stockholder interests (Fama and Jensen, 1983). Thus, for example,
management attempts to reduce customer power by building a more diverse customer base
through diversification may be blocked by the board, precisely because the board might regard
such diversification as being an inefficient use of stockholders’ funds. Whether the board can
impose such constraints in practice is the subject of some debate. Contrary to the argument made
by Mace (197 1) and others that most boards do little more than rubber-stamp management
decisions, Mizruchi (1983) has presented strong arguments in support of the proposition that
board control of management actions in public corporations is still possible. More recently, Lorsch
and MacIver (1989) present case study evidence which suggests that among selected United
States corporations, boards are increasingly exercising their control over top management teams.
148 C.W.L. HILL AND T.M. JONES On the other hand, there is also evidence which suggests that
board control over top management is still relatively weak. For example, Jensen and Murphy
(1990), after finding only a very weak relationship between CEO pay and firm performance,
concluded that most boards may lack the power to impose stockholder objectives on
management. Similarly, Burrough and Helyar (1990) have described in detail how one CEO, Ross
Johnson of RJR Nabisco, handed out lucrative consulting contracts to outside directors in a
successful attempt to keep them from criticizing management policies that were clearly
inconsistent with stockholders’ best interests. The issue of how strong boards actually are,
therefore, remains an open one. However, one factor which suggests that tighter control over
management actions may become the rule rather than the exception has been the dramatic rise of
financial institutions as major providers of capital. On both sides of the Atlantic pension funds,
insurance companies, mutual funds, and investment banks have rapidly been replacing individuals
as the main stockholders in public corporations. For example, in the United States, Hanson and Hill
(1991) present evidence which suggests that among Fortune 500 companies the percentage of
common stock held by institutions increased from 24 percent to 50 percent between 1977 and
1986. The growing concentration of stockholding in the hands of a relatively few institutions is
resulting in the evolution of a stock market that bears little resemblance to the fragmented and
dispersed market described by Berle and Means (1932). Instead, the resulting concentration of
stockholdings means that financial institutions are increasingly able to exert direct influence over
management actions, either through (a) the threat to sell their holdings; (b) the threat to fight
proxy votes more aggressively; or (c) by using their voting power to elect their own nominees to
the board of directors. For example, in 1987 a group of financial institutions with major holdings in
General Motors was able to pressure GM management into adopting a bonus pay system for GM
executives that was based upon stock price performance (prior to that time, bonuses had been
awarded automatically, irrespective of the company’s performance). The institutions did this by
threatening to introduce a resolution at the next stockholders’ meeting that would be critical of
management unless the company changed its bonus pay policies (Nussbaum and Dobrzynski,
1987). More generally, Mintz and Schwartz (1985) argue for and present evidence which supports
the view that financial institutions play a key role in the control of large firms. Similarly, Scott
(1979) concludes that large firms and major banks ‘confront one another as equals, each being
constrained by its controlling constellation of interests’ and that ‘banks are able to exercise
considerable influence over the policies of major industrial corporations and so can affect what
happens in companies where they have no direct power’ (p. 175). It should be pointed out,
however, that to a large degree management and major financial institutions share the same
agenda. Although there will undoubtedly be conflict between them, it is reasonable to suppose
that in many cases management actions designed to weaken the power of certain stakeholder
groups (e.g., employees, suppliers, or customers), will be congruent with the interests of major
financial institutions so long as they STAKEHOLDER-AGENCY THEORY 149 increase the profitability
of the corporation. Thus, while important, the potential for conflict between managers and
financial institutions should not be overstated. The Implications of Power Differentials Power
differentials created by the strategies detailed above limit the ability of stakeholders to enforce
implicit or explicit contracts. Diffusion of stakeholder power makes co-ordination between
individual stakeholders more problematic and costly, thereby reducing the ability of stakeholders
to act collectively. In turn, this limits the effectiveness of voice and exit as enforcement
mechanisms. It is more difficult for stakeholders to establish a credible threat when power is
diffused among many individuals and collective action is difficult to achieve. Similarly, the
concentration of management power reduces the choice set of stakeholders, again limiting the
effectiveness of exit and voice as enforcement mechanisms. Stakeholder diffusion also makes
monitoring more difficult. Less powerful stakeholders are less able to demand that management
make itself accountable. They are less able to use the implied threat to exit or exercise voice as a
means of gaining access to insider information or demanding that management regularly provides
them with information concerning its activities. Moreover, the pursuit of diversification strategies
by the firm obscures data relating to the efficiency of individual divisions (firms only have to
publish consolidated accounts). This exacerbates the information asymmetry between
management and stakeholders, making monitoring more problematic. Managers may also take
advantage of power differentials unilaterally to rewrite the terms of the implicit or explicit
contract between managers and stakeholders. Thus, managers may take advantage of power
differentials to revoke warranties, retract hostages posted as bonds, or retract other credible
commitments such reciprocal purchasing agreements, posted prices, or union shops. Similarly,
management may take advantage of a temporary power differential over its employees to rewrite
employment contracts. In all of these cases, the effect of power differentials is to reduce the
effectiveness of existing institutional structures and to increase the residual loss that must be born
by stakeholders. Stakeholder Responses Stakeholder responses to the creation of power
differentials can be analysed by way of figure 2. This shows the marginal benefit and marginal cost
curves underlying figure 1. We start the analysis by accepting the convenient fiction of an initial
equilibrium solution involving institutional structures of C1 complexity, a reduction in utility loss of
0-c, and a remaining utility loss of c6. The effect of a successful attempt by management to create
a power differential will be to reduce the gradient of the benefit function, and hence shift the
marginal benefit function down from MB, to MB2. In other words, power differentials limit the
effectiveness of existing institutional structures and result in a reduction in the utility loss that can
be achieved by stakeholders at each level of institutional complexity. The new equilibrium solution
implied by this shift is to be found at C2. Thus, comparative statics suggest 150 C.W.L. HILL AND
T.M. JONES 0 7 Complexity of institutional structures Figure 2 that when faced with an adverse
power differential it pays stakeholders to reduce institutional complexity to C2 and accept an
increased residual loss of However, in a dynamic sense the existence of d-b can be seen as
providing an incentive to stakeholders to find new ways of economizing on contracting costs (to
develop new institutional structures). As noted earlier, for C1 to be an equilibrium position, the
perceived gross return gained from the innovation of more efficient structures must be equivalent
to the perceived costs of such innovation. Given this, the power shift has created an incentive to
innovate equivalent to the discounted present value of Zc,-d,. Our thesis is that the existence of
such an incentive following the emergence of a power differential has driven much of the
historical evolution of institutional structures. The evolution of labour unions, consumer unions,
special-interest groups, incentive mechanisms and credible commitments, corporate regulation,
and so on, can be traced back to such incentives. For example, the development of the factory
system in nineteenth-century England led to the decline of the ‘putting-out’ system of
subcontracting and upset the balance of power that existed between those who made products to
those who managed the production process, with management benefiting (Landes, 1966). Those
who made products now became ‘employees’ and were at a power disadvantage vis-a-vis those
who managed the process. Traditionally, craft guilds had governed the implicit contract between
‘managers’ and ‘subcontractors’. One consequence of the shift to a factory system was that craft
guilds lost their effectiveness as institutional structures, and declined dramatically in influence
(Landes, 1966). Thus, the decline in the marginal benefit curve from MB, to MB2 (i.e., existing
institutional structures were no longer effective). However, this shift increased the incentive that
those who d-b. STAKEHOLDER-AGENCY THEORY 151 made products (employees) had to develop
new and more effective governance mechanisms. The response was the development of labour
unions with the objective of re-establishing a condition of mutual dependence between those who
made products and those who managed the manufacturing process. The effectiveness of unions
was based on their ability to economize on co-ordination costs between diffused stakeholders and
re-establish exit as a credible threat. Of course, such adjustments are anything but smooth and will
be resisted by the advantaged party. Indeed, there is a long history of management resistance to
the development of union power following the introduction of the factory system. Moreover, the
conflict between management and stakeholders following the development of a power differential
can be expected to spill over into the explicitly political arena. That is, both parties can be
expected to try to use the power of law to further their interests. This is hardly surprising given
that many institutional structures either have a legal component or are supported by the law, but
it does give us a way of explaining the selective use of Political Actidn Committee (PAC) money,
along with more general lobbying by corporate trade associations and public interest groups.
Specifically, at any point in time such monies and lobbying will be devoted to ongoing
management-stakeholder conficts, the amount of activity and money being roughly proportional
to the size of the perceived power differential and the anticipated gains from either changing the
system or maintaining the status quo. Finally, it is important to remember that power differentials
work both ways. Although we have concentrated on the benefits enjoyed by management from
their control over the decision-making apparatus of the firm, and although this control probably
does give management an inbuilt advantage, management may be put on the defensive by
increases in stakeholder power (as may be occurring zlis-u-uis stockholders due to the increase in
the amount of stock held by financial institutions). As with management power, increases in
stakeholder power have their genesis in disequilibrium conditions created either by exogenous
shocks, or by innovations in the way that stakeholders do business. CONCLUSION The objectives of
this article were ambitious. Taking agency and stakeholder perspectives of the firm as our starting
point, we have attempted to construct a paradigm that explains certain aspects of the strategic
behaviour of the firm, the structure of incentive alignment mechanisms, and the institutional
forms that have evolved to police the implicit and explicit contracts between managers and
stakeholders. In doing so, we have drawn on the literatures of business and society, economics,
finance, and organizational theory. The resultant paradigm, stakeholder-agency theory, can be
viewed as a modification of agency theory to accommodate theories of power including resource
dependence theories of organizations. Hitherto, these theories have been seen as offering
mutually exclusive interpretations of. organizational I52 C.W.L. HILL AND T.M. JONES phenomena
(Perrow, 1986). While agency theory assumes efficient markets and rejects the idea of power
differentials between managers and stakeholders, resource dependency theory (e.g., Pfeffer and
Salancik, 1978) implicity assumed inefficient markets which allow for the existence of unequal
resource dependencies (power differentials) between managers and stakeholders. The adoption of
an ‘Austrian’ perspective on market processes allows us to treat notions of power and efficiency
within the framework of the same model. Following the theme of Austrian economics, we accept
that markets are efficient. However, the existence of short-run disequilibrium arising from
exogenous and endogenous change has been argued to give rise to temporary power differentials
between managers and stakeholders. Some of the strategies pursued by managers with respect to
stakeholders can be seen as an attempt to exploit and entrench these power differentials. In turn,
the evolution of new incentive structures and institutional mechanisms for monitoring and
enforcing the contractual relationships between managers and stakeholders can be seen as long-
run market-generated responses to disequilibrium conditions and unequal resource dependencies.
‘ Our contention is that joining together notions of power and efficiency within the same
framework substantially increases the predictive power of the paradigm when compared to earlier
‘theories of the firm’. Unlike earlier theories, the paradigm explicitly focuses on the causes of
conflict between managers and stakeholders following the emergence of disequilibrium
conditions. Stakeholder-agency theory also points the way towards a theory of the adjustment
mechanisms that realign management and stakeholder interests following disruption.