Power, Six Bases of

Though there have been many formal definitions of leadership that did not include social influence and power,
any discussion of leadership must inevitably deal with the means by which a leader gets the members of a
group or organization to act and move in a particular direction. One of the most widely cited analyses of social
power is that proposed by John R. P. French and Bertram H. Raven in 1959. They defined social influence as a
change in the belief, attitude, or behavior of a person (the target of influence) which results from the action of
another person (an influencing agent), and they defined social power as the potential for such influence, that is,
the ability of the agent to bring about such a change using available resources. French and Raven identified five
bases of power: coercive, reward, legitimate, expert, and referent. To this was later added a sixth: information
power.

DEFINING THE BASES OF POWER

The bases of power differ according to the manner in which social changes are implemented, the permanence of
such changes, and the ways in which each basis of power is established and maintained. To illustrate these
distinctions, we will examine a leadership situation involving the power of a supervisor who attempts to
influence the behavior of a subordinate in a work situation.

Power That Leads to Socially Independent Change

One basis of power the supervisor might use is information power. The supervisor carefully explains to the
subordinate how the job should be done differently, using persuasive reasons to explain why that would be a
better and more effective procedure. The subordinate understands and accepts the reasons and changes his
behavior. Information influence, then, brings about a cognitive change in and acceptance by the target. It is
thus called “socially independent change” in that the target now continues the changed behavior without
necessarily referring to, or even remembering, the supervisor as the agent of change.

Power That Results in Socially Dependent Change with Surveillance Necessary

Reward power stems from the ability of the agent to offer a positive incentive for the target to comply (for
example, a raise in pay, a promotion, or special work privileges). An agent who uses coercive power brings
about change by threatening a target who does not comply with undesirable consequences (for example,
demotion, termination, or undesirable work assignments). In both reward power and coercive power, the
influence is clearly socially dependent, since the target relates compliance to the actions of the agent (for
example, “I did it because my supervisor offered me a reward if I complied”). Reward power and coercive power
both from other bases of power in that not only are they socially dependent, but also because their
effectiveness requires surveillance by the influencing agent: If reward and coercion are the only bases of power
used by influencing agents, targets will comply only if they believe that the agents will be able to determine
whether or not they complied. However, coercive power differs from reward power in the ease by which an
agent may maintain surveillance. With reward power, it is to the advantage of targets to let agents know that
they have complied; with coercive power, there may be a tendency for targets to hide the extent of their
noncompliance, so that agents may require targets to demonstrate their compliance. There is also a tendency
for the targets of coercive power to resent the threat of punishment, to resent feeling forced, and to have ill

1
feelings toward the agent, as well as toward the behavior which they feel forced to accept. This is less likely to
be true for reward power; indeed, the positive feelings associated with the reward may lead to a greater
acceptance of the change and a greater liking for the influencing agent.

Power That Leads to Socially Dependent Change with Surveillance Unnecessary

The remaining three bases of power result in change which, initially, is dependent upon the influencing agent,
but where surveillance is not necessary for the influence to occur: Legitimate power stems from the target's
accepting the right of the agent to require the change in behavior (for example, “After all, I should do what my
supervisor requests of me”). Terms such as “obliged” “should,” “ought to,” and “required” may signal the use of
legitimate power. Expert power results from the target's faith that the agent has some superior insight or
knowledge about what behavior is best under the circumstances. (For example, “My supervisor has had a lot of
experience with this sort of thing and so is probably right, even though I don't really understand the reason.”)
“Understanding the reason,” then, is what distinguishes informational power from expert power. Referent power
stems from the identification of a target with an agent. (For example, “I really admire my supervisor, and wish
to be like her. Doing things the way my supervisor believes they should be done gives me some special
satisfaction.”)

FURTHER DIFFERENTIATION
The six bases of power have been further differentiated based on additional research.

Coercive Power and Reward Power: Personal versus Impersonal Forms

In our original statement, coercive power and reward power were presented in terms of tangible rewards and
real threats—for example, the promise of a bonus or promotion or the threat of being fired or fined. However,
the personal approval of someone we respect can also be a very powerful reward, and a threat of rejection or
disapproval from someone we value highly can serve as a source of coercive power.

Legitimate Power: Position, Reciprocity, Equity, Responsibility

Legitimate power stems from social norms requiring that the target of influence comply with the request or
order of the influencing agent. Legitimate position power, the most obvious form of legitimate power, stems
from a social norm that requires that we obey people who are in a superior position in a formal or informal
social structure, such as a supervisor or a higher ranking military officer. Other examples include the right of
parents to influence children, of older people to influence younger ones, and of teachers to influence students.

Other, more subtle forms of legitimate power based on social norms include legitimate power of reciprocity,
legitimate power of equity, and legitimate power of responsibility. The reciprocity norm states that if someone
does something beneficial for us then we should feel obliged to reciprocate. Legitimate power of equity, also
called a “compensatory norm,” would be something like “I have worked hard and suffered and therefore I have
a right to ask you to do something to make up for it.” Finally, the norm of legitimate power of responsibility,
also called the “power of the powerless,” suggests that we have some obligation to help others who cannot help
themselves or who depend on us. A supervisor could conceivably say, “Look, I am not about to force you to
follow my method, but it is absolutely essential to me that you do so in order to get the job done this way. I
really depend upon you to do this for me.”

The Power/Interaction Model of Interpersonal Influence

The bases of power are included within a larger context through the development of a power/interaction model
of interpersonal influence designed in 1988 by Bertram H. Raven. The model begins with a consideration of the
motivation for influence and the use of power, and goes on to explore the factors which lead to an agent's
choice of a power strategy, the preparatory devices for implementing the bases of power, the manner in which a

2
power strategy is utilized, the effective changes or lack of change in the target of influence, the aftereffects, and
the agent's readjusted perceptions and choices of future strategies.

MOTIVATION FOR CHOOSING A BASIS OF POWER

Typically, an agent's motivation for influence will be very obvious, the purpose being to attain some goal or
desirable outcome, and the agent will then use the basis of power that will accomplish that end most
expeditiously and effectively. Often the situation will affect what bases of power will be selected. For example,
different power strategies will be chosen by a supervisor in a supermarket, a warden in a prison, the leader of a
boy scout troop, a parent influencing a child, and a teacher in a classroom.

In addition, the selection of power strategies will depend on how an agent views a target and even more, on
how an agent believes that a target views him. In an analysis in 1960, social psychologist Douglas McGregor
distinguished between “Theory X” supervisors and “Theory Y” supervisors. The former believe that workers
cannot be trusted, that they do not really like their work, and that they try to do as little as they can. In this
model, Theory X supervisors rely on coercive power and legitimate position power and surveillance. By contrast,
Theory Y supervisors, who have a more positive view of workers, are more likely rely on informational and
expert power, and also perhaps on legitimate power of dependence, and to place less emphasis on surveillance.
The basic point is that influencing agents, motivated to achieve the most positive outcome, will select bases of
power based on their perception of what would work best with a specific target.

The choice of power strategies may also depend on more subtle motivations. David McClelland, David Winter,
and their colleagues found three important but subtle motives: the need for power, the need for affiliation, and
the need for achievement. A leader or supervisor with a high need for power will be more likely to select
impersonal coercive power and legitimate position power. Those with strong affiliation needs and a concern that
their subordinates will like them will more likely prefer referent power and reward power, especially personal
reward power. A need for achievement might result in more use of informational and expert power. Other
personality characteristics might also affect the choice of a power strategy, including an agent's having high or
low selfesteem. One reason for this might be that successful influence from informational power tends to be
attributed to the target (for instance, “I gave him good reasons, but he decided to do it”), while successful
influence from coercive power tends to be attributed to the influencing agent (for example, “She did so because
I influenced her to do so”). Influencing agents who have low self-esteem find it satisfying to know that they are
calling the shots and they would be more likely to select a basis of power like coercion.

Another sort of motivation which might affect the choice of power strategies is the attitude of the influencing
agent toward the target of influence. It is, of course, the agent's perception of the target which helps determine
what basis of power would be effective or ineffective, but, in addition, a strong negative feeling toward the
target might lead to a choice of a harsh basis of power, such as impersonal coercion, even when that power
strategy might not be the most efficient or effective. Similarly, a strong positive feeling toward the target might
preclude the use of a harsh basis of power even when, objectively, it might seem most appropriate.

ASSESSMENT OF THE COSTS AND BENEFITS OF DIFFERENT POWER STRATEGIES

The agent might also go through a cost-benefit analysis of the influence strategy. Informational influence or
persuasion would ordinarily be highly desirable, but may require more time and effort than is available.
Coercion, as we had indicated, may result in more rapid compliance, but carries with it the costs of having to
maintain surveillance, the hostility of an unhappy subordinate, and sometimes the violation of one's personal
value system or generally accepted social norms. The legitimacy of dependence (“I need your help”) may lead
to loss of respect and perhaps may imply an obligation to return the favor. Referent power, which emphasizes
similarity, may undermine the target's respect for the agent's superiority in expertise and legitimacy of position.
In addition, as we have noted, powerholders, because of their personalities, experiences, and values, or force of

3
habit, may tend to prefer some bases of power over others.

PREPARATORY DEVICES FOR IMPLEMENTING BASES OF POWER

Though influencing agents may often have immediate access to their bases of power, it is often the case that
some preparation or stage-setting is necessary. To use coercion, it is sometimes necessary to first make the
target realize that the agent has both the means and the will to follow through on the threat. Edward Jones and
Thane Pittman in 1982 and Erving Goffman in 1959 described a number of “selfpresentational strategies” and
“impression management” techniques by which an influencing agent or leader may set the stage for the use of a
particular power strategy.

Establishing Information Power

Agents might carefully rehearse their speeches, examine the logic, and practice the delivery. Or they may first
give a target some background information, which would build a basis for the subsequent persuasion.

Intimidation

To use coercion effectively, it may be important to demonstrate to the target not only that the means are
available for coercion, but also that the agent is ready and willing to pay the costs that coercion implies.
Workers will not be influenced by a threat of dismissal if they do not really feel that their supervisor is ready to
implement the threat. A supervisor, attempting to establish the credibility of coercive power, may launch into an
emotional tirade, or even fire a worker, just to set an example.

Ingratiation

In order to utilize personal reward or coercion, or referent power, the agent may first attempt to ingratiate her-
or himself with the target, with well-placed compliments, flattery, and so on.

Emphasizing Communality

To establish referent power, the agent must also develop a sense of communality with the target. (“Look,” the
supervisor may say, “we are really all one team, trying to get this job done.”)

Self-Promotion

For expert power, a few choice demonstrations of one's superior knowledge would be useful. The supervisor
might tell the worker of the amount of training and experience he or she has had on this and similar jobs.
(Physicians, attorneys, professors, and other professionals go through elaborate stagesetting devices for
expertise—for example, displaying diplomas and extensive libraries or using impressive language.)

Authorization for Legitimate Position Power

To establish his/her formal legitimate position power, the supervisor might subtly mention that he or she is,
after all, the supervisor who is responsible for this job. Similar preparatory devices may be seen in the cases of
the usurper who seizes the throne and then presents evidence that actually heredity justifies his or her
ascendance, or the dictatorial modern ruler who establishes a legitimate position power through rigged
elections.

Doing Favors to Establish Legitimate Reciprocity

To establish this form of legitimacy, the agent may first do a favor for the target, or emphasize the various
favors that he or she has done in the past.

Guilt Induction for Legitimacy of Equity

An agent may induce guilt in order to establish his or her legitimacy of equity. Somehow the agent may
4
convince the target that the target has caused harm or pain, for which the agent is entitled to compensation in
the form of unhesitating compliance.

Demonstrating Effective Surveillance

Since both coercive power and reward power require surveillance, influencing agents who expect to use these
bases of power may find it necessary to establish their ability to determine whether a target has complied. They
may do this by confronting a subordinate regarding an infraction that the subordinate thought had been done in
private.

Implementing the Power Strategy and Assessing Its Effects

Following the influence attempt, the agent will want to assess the effects. Was it successful? Is there evidence
that the target has actually accepted the influence, has actually altered his or her behavior in accordance with
the outcome desired by the influencing agent? Does the target really accept the change personally, or is the
change socially dependent? Is surveillance important for the change to continue—will the target revert to earlier
behavior patterns as soon as the agent cannot continue to check on the degree of compliance? Will the target
subsequently internalize the changes in his/her behavior?

How about secondary effects? How has the influence attempt, successful or not, affected the target's perception
and evaluation of the agent? Has respect for the agent diminished? Is there greater personal liking or disliking?
Have the power bases previously available to the agent increased or decreased in their potency? Agents may
attempt to repair the damage and reassess their relationship with a target. If influence attempts are
unsuccessful, then it is likely that agents will try again. But this time their motivations may change: Whereas
previously they had merely wanted to achieve the extrinsic goal, they now may have developed some hostility
toward the target, which in turn will affect their choice of an influence strategy the second time around. An
agent's success or failure will also lead to a reassessment of the available bases of power and the development
of a very different strategy.

METAMORPHIC EFFECTS OF POWER

David Kipnis and his colleagues have pointed out that the very process of surveillance that goes with coercive
power contributes to the distrust of the influencing agent and further demeans the target of influence.
Demeaning the target allows the agent to feel more powerful and then to use even harsher power strategies—
typically, the use of informational power may diminish, and the use of coercive power and legitimate position
power will increase. In 1976, Kipnis referred to this escalation process as the “metamorphic effects of power.”
There are many examples of forthright leaders who have been transformed into tyrants, leading Lord John
Acton (1834–1902) to observe that “power tends to corrupt and absolute power corrupts absolutely.”

Power/Interaction from the Perspective of the Target

We have thus far examined the power/action model from the perspective of the influencing agent, but it can
also be profitably examined from the perspective of the target of influence. Following an essentially similar
pattern, the target may have various motives to either accept or reject influence from the agent, some of which
may involve personal factors, such as a need for independence, for power, and for selfesteem, and positive or
negative personal feelings toward the influencing agent. Also, like agents, targets may be concerned about how
they would look to third parties if they complied or did not comply. Thus, just as agents or leaders may operate
less effectively because of inappropriate motives, targets may sometimes resist influence inappropriately. For
example, workers might marshal their personal resources in preparation for a verbal assault. They might have
tried to anticipate what bases of power the influencing agent might attempt to use and have prepared to
counter these one by one. Targets might even invoke the powers of third parties to assist in their resistance or
to organize other potential targets so as to resist influence collectively.

5
Subsequent Relations Between Agent and Target

The influence attempt, successful or unsuccessful, has very likely changed both the influencing agent and the
target, changed their perceptions of themselves and changed their perceptions of the other. An unsuccessful
influence attempt may result from a misperception of the available effective power bases, as perceived by the
influencing agent and by the target. Indeed, unsuccessful influencing agents may alter their strategies as the
result of a failed first attempt, adopting new strategies which might now be effective—except for the fact that
the target has also changed. The target may now be amenable to influence strategies which would not have
worked the first time round, or be more resistant to strategies which might earlier have been effective. The
issues become even more complex as the two participants attempt to influence one another, each serving as
both influencing agent and target with respect to the other. Interpersonal or intergroup conflict can be
examined in terms of mutual influence attempts, using various bases of power and strategies by both parties to
the conflict and analyzing the effects which these have on one another.

MEASURING THE BASES OF POWER

Several investigators have developed questionnaires to measure the bases of power, as perceived both by the
influencing agent and by the target of influence. In most cases, these questionnaires have measured only the
original six bases of power. The Interpersonal Power Inventory, introduced by Bertram Raven, Joseph
Schwarzwald, and Meni Koslowsky in 1998, measures the eleven power strategies in a model they call “the
power/interaction model”: coercion (personal and impersonal); reward (personal and impersonal); legitimate
(position, reciprocity, equity, and dependence or responsibility); referent; expert; and informational. The thirty-
threeitem scale includes three items that measure the likelihood or effectiveness of each of these strategies.
Factor analyses have indicated that for some situations, these power strategies can be differentiated into two
major components, a “soft” group consisting of Expert, Information, Referent, and Legitimate Dependence, and
a “harsh” group comprised of the other seven strategies. Job satisfaction was found to be more positive when
supervisors used the “soft” strategies and more negative when “harsh” strategies were used. Power/interaction
profiles for individual respondents illustrate the ways in which leaders may differ in their choice of strategies and
show how their influence may be made more effective.

THE EFFECTIVENESS OF VARIOUS BASES OF POWER

It is of particular practical interest to know what bases of power or which power strategies are most likely to be
effective, but it is clear that there is no simple answer. For example, a power strategy that works immediately
but relies on surveillance (for example, reward power or coercive power) may not last once surveillance ends.
One organizational study found that reward power tended to lead to greater satisfaction on the part of
employees, which means that it might increase influence in a broad range of situations. Coercive power was
more effective in influencing a subordinate who jeopardized the success of the overall organization or
threatened the leader's authority, even though in the short term it also led to resentment on the part of the
target. A power strategy that ultimately leads to private acceptance and long-lasting change (for example,
information power) may be difficult to implement, and consume considerable time and energy. In the short
term, complete reliance on information power might even be dangerous (for example, telling a small child not to
run into the street unattended). A military officer leading his troops into combat might be severely handicapped
if he had to give complete explanations for each move. Instead, he would want to rely on unquestioned
legitimate position power, backed up by coercive power. Power resources which may be effective for one leader,
dealing with one target or follower, may not work for a different leader and follower. The manner in which the
power strategy is utilized will also affect its success or failure. Where coercion is deemed necessary, a leader
might soften its negative effects with a touch of humor. There have been studies indicating that cultural factors
may determine the effectiveness of power strategies.

As long as humans have had to interact with one another, they have utilized power strategies, and in various

6
degrees the strategies have been effective even when an agent has had no formal knowledge of a
power/interaction model or of the bases of social power. The model is simply an attempt to understand how this
process operates and the conditions under which social influence is more or less effective. It is reasonable to
conclude that a leader who is more aware, either formally or informally, of the various options in social power
strategies will be more successful and effective.

—Bertram H. Raven

Further Reading
Berkowitz, L. and Daniels, L. R. Responsibility and dependency. Journal of Personality and Social Psychology vol.
3 (1963) pp. 429–436

French, J. R. P., Jr., & Raven, B. H. (1959). The bases of social power. In D. Cartwright (Ed.), Studies in Social
Power (pp. 150–167). Ann Arbor, MI: Institute for Social Research.

Frost, D. E. and Stahelski, A. J. The systematic measurement of French & Raven's bases of social power in
workgroups. Journal of Applied Social Psychology vol. 18 (1988) pp. 375–389

Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday.

Gold, G. J. and Raven, B. H. Interpersonal influence strategies in the Churchill Roosevelt bases for destroyers
exchange. Journal of Social Behavior and Personality vol. 7 (1992) pp. 245–272

Goodstadt, B. and Kipnis, D. Situational influences on the use of power. Journal of Applied Psychology vol. 54
(1970) pp. 201–207

Graen, G., & Wakahayashi, M. (1993). Cross-cultural leadership-making: Bridging American and Japanese
diversity for team advantage. In H. C. Triandis, M.D. Dunette, ed. , & J. M. Hough (Eds.), Handbook of industrial
and organizational psychology. New York: Consulting Psychology Press.

Hinkin, T. R. and Schreischeim, C. A. Development and application of new scales to measure the French and
Raven (1959) bases of power. Journal of Applied Psychology vol. 74 (1989) pp. 561–567

Johnson, P. (1976). Women and interpersonal power. In I. Frieze, P. Johnson, J. E. Parsons, D. N. Ruble, ed. , &
G. L. Zellman (Eds.), Women and sex roles: A social psychological perspective. New York: Norton.

Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic self presentation. In J. Suls (Ed.),
Psychological perspectives on the self, volume 1. Hillsdale, NJ: Erlbaum.

Kipnis, D., Castell, P. J., Gergen, M., and Mauch, D. Metamorphic effects of power. Journal of Applied
Psychology vol. 61 (1976) pp. 127–135

Koslowski, M., Schwarzwald, J., and Ashuri, S. On the relationship between subordinates compliance to power
sources and organizational attitudes. Applied Psychology: An International Review vol. 50 (2001) pp. 455–476

McClelland, D. C. and Boyatsis, R. E. Leadership motive pattern and long-term success in management. Journal
of Applied Psychology vol. 67 (1982) pp. 737–743

McGregor, D. (1960). The human side of enterprise. New York: McGraw-Hill.

Meliá, J. L., Oliver, A., and Tomás, J. M. A bifactorial theory of organizational power: Hypothesis and
measurement. Revista de Psicologia Social Applicada vol. 3 (1993) pp. 25–41

7
Podsakoff, P. M. and Schriescheim, C. A. Field studies of French and Raven's bases of social power: Critique,
reanalysis, and suggestions for future research. Psychological Bulletin vol. 97 (1985) pp. 387–411

Raven, B. H. (1965). Social influence and power. In I. D. Steiner, ed. & M. Fishbein (Eds.), Current Studies in
Social Psychology. New York: Holt, Rinehart, Winston.

Raven, B. H. (1988). A power/interaction model of interpersonal influence and social power. Presented at the
XXIV International Congress of Psychology, Sydney, Australia.

Raven, B. H. (1988). Social power and compliance in health care. In S. Maes, C. D. Spielberger, P. B. Defares,
ed. , & I. G. Sarason (Eds.), Topics in health psychology. London and New York: Wiley.

Raven, B. H. Political applications of the psychology of interpersonal influence and social power. Political
Psychology vol. 11 (1990) pp. 493–520

Raven, B. H. A power/interaction model of interpersonal influence: French and Raven 30 years later. Journal of
Social Behavior and Personality vol. 7 (1992) pp. 217–244

Raven, B. H., & Litman-Adizes, T. (1986). Interpersonal influence and social power in health promotion. In Z.
Salisbury, S. Kar, ed. , & J. Zapka (Eds.), Advances in Health Education and Promotion. Greenwich, CT: JAI
Press.

Raven, B. H., Schwarzwald, J., and Koslowsky, M. Conceptualizing and measuring a power/interaction model of
interpersonal influence. Journal of Applied Social Psychology vol. 28 (1998) pp. 307–332

Stahelski, A. J., Frost, D. E., and Patch, M. E. Use of socially dependent bases of power: French and Raven's
theory applied to workgroup leadership. Journal of Applied Social Psychology vol. 19 (1989) pp. 283–297

Winter, D. G. (1973). The power motive. New York: Free Press.

Winter, D. G. Leader appeal, leader performance, and the motive profiles of leaders and followers: A study of
American presidents and elections. Journal of Personality and Social Psychology vol. 52 (1987) pp. 196–202

Yukl, G. A. and Falbe, C. M. Importance of different power sources in downward and lateral relations. Journal of
Applied Psychology vol. 76 (1991) pp. 416–423

Entry Citation:
Raven, Bertram H. "Power, Six Bases of." Encyclopedia of Leadership. Ed. . Thousand Oaks, CA: SAGE, 2004. 1242-49. SAGE
Reference Online. Web. 30 Jan. 2012.

© SAGE Publications, Inc.

Brought to you by: SAGE