Bases of Social Power: French and Raven’s Six Types of Power Explained
Author: Ukrainian Psychological Hub · Published: September 24, 2026 · Editorial Policy
French and Raven’s bases of social power are one of the most influential frameworks for explaining why one person can change another person’s behavior, judgment, or choices. The familiar six-type version distinguishes reward, coercive, legitimate, expert, referent, and informational power. Each base answers a different question: what resource, relationship, norm, or belief gives an influencing agent the potential to produce change in a target?
The framework is useful because it does not reduce power to rank, force, charisma, or leadership. A manager may have legitimate power because of a role, expert power because of specialized knowledge, reward power because of control over valued outcomes, and referent power because others identify with them. A person with no formal office may still have considerable expert, referent, or informational power. This is why the framework belongs inside the broader psychology of social power, rather than being treated as a list of leadership styles.
The six bases of social power at a glance
The six bases are:
1. Reward power: influence grounded in the perceived ability to provide valued benefits or remove unwanted conditions.
2. Coercive power: influence grounded in the perceived ability to impose costs, punishment, rejection, or other aversive outcomes.
3. Legitimate power: influence grounded in a belief that the agent has a recognized right to make a request and that the target has some obligation to accept it.
4. Expert power: influence grounded in the target’s belief that the agent has relevant knowledge, skill, or expertise.
5. Referent power: influence grounded in identification, attraction, admiration, affiliation, or a desire to maintain a valued relationship with the agent.
6. Informational power: influence grounded in the content of information, reasons, evidence, or argument that the target accepts as persuasive.
The first five were set out in John R. P. French Jr. and Bertram Raven’s 1959 chapter “The Bases of Social Power” in Dorwin Cartwright’s Studies in Social Power. Raven later added informational power, producing the six-base framework now commonly associated with French and Raven. Raven’s later scholarly retrospectives describe how the model developed well beyond the original taxonomy (Raven, 1993; Raven, 2008).
What does “a base of power” mean?
In the French–Raven tradition, power is a potential for social influence. Influence is the realized change: a shift in a person’s behavior, attitude, belief, judgment, or action associated with another person’s intervention. A base of power is the psychological or social foundation that makes such influence possible.
That distinction matters. Possessing a resource does not guarantee successful influence. A supervisor can formally control bonuses, but reward power is weak if employees do not value the available rewards or believe the supervisor cannot actually deliver them. A physician can have extensive expertise, but expert power is limited when a patient doubts the physician’s competence in the relevant domain. A celebrity can have referent power among admirers and almost none among people who do not identify with them.
Power in this framework is therefore relational and domain-specific. It depends on what the target perceives, values, accepts, fears, trusts, or recognizes. The same agent can hold different bases of power over different targets and in different situations.
This also separates the framework from status, authority, leadership, and personal control. Those concepts overlap with power, but they describe different psychological or social structures.
Where the model came from: five bases in 1959, six after Raven
French and Raven developed their framework within the mid-twentieth-century social-psychological study of influence associated with Kurt Lewin and the Research Center for Group Dynamics. Their 1959 chapter focused on five bases: reward, coercive, legitimate, referent, and expert power.
Informational power was added later by Raven. The addition was theoretically important because persuasion based on accepted reasons works differently from influence that continues to depend on the influencing person. If someone changes a procedure because a supervisor can punish noncompliance, the supervisor remains central to the behavior. If someone changes the procedure because they have accepted evidence that the new method is safer, the reason itself can continue to guide behavior after the supervisor leaves.
Raven’s later Power/Interaction Model also made the framework more dynamic. It considered why an agent chooses a particular influence strategy, what power bases are available, how the agent prepares the situation, how the target responds, and what positive or negative aftereffects follow (Raven, 1993). By 1998, Raven, Schwarzwald, and Koslowsky were measuring 11 differentiated power bases rather than treating the familiar six as indivisible categories (Raven et al., 1998).
The six types are therefore best understood as the canonical map of the framework, not as the final word on every form of human influence.
1. Reward power
Reward power exists when the target believes that the influencing agent can provide something the target values. The reward may be material, social, symbolic, informational, or experiential.
In a workplace, salary increases, bonuses, desirable assignments, promotion opportunities, schedule flexibility, access to resources, and public recognition can all support reward power. In a friendship or family, approval, affection, attention, inclusion, help, or access to shared activities can function as rewards. In online communities, visibility, recommendations, invitations, badges, or access can play the same role.
The mechanism is not simply “giving rewards.” The target must connect the desired outcome to the agent and to the requested behavior. A manager who praises everyone regardless of performance may be liked, but the praise has little contingent reward power over a specific behavior.
Reward power also depends on credibility and control. If an agent repeatedly promises benefits and fails to deliver, perceived reward power declines. If the reward is unwanted, it has little motivational value. If the reward is available from many alternative sources, dependence on the particular agent can also decrease.
Later versions of the model distinguished impersonal rewards, such as pay or promotion, from personal rewards, such as approval or acceptance (Raven et al., 1998). This matters because the psychological meaning of “reward” depends on what kind of relationship is involved.
Reward power can produce compliance without producing agreement. A person may perform a task to obtain a bonus while privately believing the task is pointless. For that reason, reward power should not be equated with persuasion, identification, commitment, or intrinsic motivation.
2. Coercive power
Coercive power exists when the target believes that the agent can impose an unwanted consequence or withdraw something valued if the target does not comply.
Examples include formal discipline, fines, loss of privileges, demotion, exclusion, criticism, rejection, social retaliation, or credible threats. Coercion can be institutional or interpersonal. A supervisor may control disciplinary procedures; a peer group may control inclusion; a partner may use withdrawal or intimidation; an organization may restrict access to resources.
French and Raven treated coercive power as conceptually distinct from legitimate power. A request can be obeyed because the target recognizes a right to command, because the target fears consequences, or because both processes operate at once. The existence of a punishment does not itself make an order legitimate, and legitimacy does not require punishment.
Coercive power often depends strongly on monitoring and credibility. If the agent cannot detect noncompliance, cannot impose the threatened consequence, or is unlikely to act, coercive leverage falls. That helps explain why surveillance is especially relevant to coercive influence in the classic framework.
Coercion can be effective at producing immediate behavioral compliance. Its broader effects depend heavily on context, proportionality, legitimacy, alternatives, and the relationship between agent and target. It can also produce concealment, resentment, resistance, or attempts to escape the relationship. Field research on French and Raven’s taxonomy has long faced measurement problems, so simple claims that one base always causes one outcome are not warranted. A major review by Podsakoff and Schriesheim found that much of the early field literature had serious methodological limitations (Podsakoff & Schriesheim, 1985).
3. Legitimate power
Legitimate power comes from the target’s belief that the agent has a recognized right to make a request and that the target has a corresponding obligation to accept some range of influence.
Formal role is the clearest example. Judges, teachers, managers, referees, elected officers, clinicians, parents, or team leaders may be granted legitimate power within specific boundaries. The power belongs to the role relationship and the norms surrounding it, not merely to the personality of the person occupying the role.
This is the point where legitimate power overlaps most directly with authority. Authority normally implies some recognized legitimacy, whereas social power is broader: an actor can influence another through expertise, rewards, coercion, information, or identification without possessing authority.
Legitimate power is bounded. A professor may legitimately set requirements for a course but has no general right to direct a student’s private life. A physician can make recommendations within a professional domain but cannot assume unlimited authority over a patient’s unrelated choices. A manager can allocate work within an organizational role while lacking legitimate control over an employee’s personal relationships.
Raven later differentiated legitimate power beyond formal position. His expanded model included legitimacy based on reciprocity, equity, and dependence or social responsibility, alongside position-based legitimacy (Raven et al., 1998). These extensions show that perceived obligation can arise from social norms as well as formal office.
Legitimacy also varies across targets and contexts. The same rule can be accepted as legitimate by one person and rejected by another. This makes legitimacy a psychological perception as well as an institutional property, a distinction explored more fully in Legitimacy and Authority.
4. Expert power
Expert power comes from the target’s belief that the agent has relevant knowledge, competence, skill, or experience.
A cybersecurity specialist can influence a chief executive’s decision about a security breach. A mechanic can influence a car owner’s repair choice. A nurse can influence a physician or administrator on a process the nurse understands better through direct experience. A junior employee can hold expert power over a senior manager when the junior employee controls specialized knowledge the manager lacks.
Expert power is therefore not the same as rank. It can flow upward, downward, or laterally. In an influential study of managers, legitimate power, expert power, and perceived persuasiveness were among the important reasons people reported for doing what a boss or peer requested, and personal power was strongly related to commitment and managerial effectiveness ratings (Yukl & Falbe, 1991).
Expert power is domain-specific. A brilliant surgeon does not acquire expert power over tax law merely by being brilliant at surgery. Expertise can also be misperceived. Credentials, confidence, reputation, institutional affiliation, or technical language can create an impression of expertise that exceeds actual competence.
Trust matters because expertise must be accepted by the target. If the target doubts the expert’s knowledge, independence, or relevance to the problem, expert power weakens even when the agent is objectively well qualified.
Expert power is closely related to informational power but remains distinct. Expert power rests on who the source is perceived to be; informational power rests on the persuasive force of what is communicated.
5. Referent power
Referent power comes from identification with, attraction to, admiration for, or a valued relationship with the influencing agent. The target is influenced because the agent is a meaningful reference point for identity, affiliation, aspiration, or belonging.
A respected mentor may have referent power because a protégé wants to emulate them. A popular peer may influence group norms because others want affiliation or approval. A public figure may shape preferences among followers who identify with them. A friend can have referent power when maintaining closeness matters to the target.
Referent power helps explain why influence can occur without formal authority, material rewards, punishment capacity, or superior expertise. The relationship itself becomes a resource.
Charisma can contribute to referent power, but the concepts are not identical. Referent power is defined by the target’s identification or relational orientation toward the agent. A person can be charismatic yet fail to become a meaningful reference point for a particular target, and referent power can grow through trust, loyalty, shared identity, or long-term affiliation without dramatic charisma.
Referent power can also become negative in later elaborations of the framework: a person may deliberately distance themselves from someone they dislike or reject, producing influence through disidentification. The canonical six-type model usually presents the positive form because it is the clearest starting point.
Like expert power, referent power is often grouped with “personal power” in organizational literature. That shorthand is useful, but it should not erase the mechanism. Expert influence depends on perceived competence; referent influence depends on identification or relationship.
6. Informational power
Informational power is influence produced by reasons, evidence, explanation, or information that the target accepts as valid.
Suppose a safety engineer shows that a proposed procedure creates a measurable hazard, explains the mechanism, and provides convincing evidence. If the decision maker changes course because the argument makes sense, the influence is informational. The engineer may also have expert power, but the informational mechanism lies in the accepted content of the case.
This distinction is one of the most important parts of the six-base model. Expert power says, in effect, “I accept this because I trust the source’s expertise.” Informational power says, “I accept this because the reasons or evidence persuade me.”
Once information is genuinely understood and accepted, the resulting change can become less dependent on the continuing presence of the original agent. This is why informational power was described as capable of producing socially independent change. A clear scholarly overview of the six bases emphasizes this feature of informational influence (SAGE Encyclopedia of Leadership).
Informational power is not identical to possessing secret information. Merely controlling access to facts can create leverage, but Raven’s informational base is specifically concerned with influence through the content and acceptance of information. A database administrator who can withhold a report may have resource control; an analyst who persuades a team by showing what the report demonstrates is exercising informational influence.
Expert power vs informational power
These two bases are frequently confused because expertise usually involves information.
The simplest test is to ask what carries the influence.
If the target accepts a recommendation mainly because “this person knows what they are talking about,” expert power is central.
If the target accepts the recommendation mainly because “the evidence and reasoning convince me,” informational power is central.
Both can operate simultaneously. A scientist may gain an audience because of expertise and then persuade that audience through data. A novice can also have informational power when the evidence is transparent and persuasive even though the novice has little recognized expertise.
This distinction is useful in modern information environments because apparent expertise and evidential quality can diverge. Reputation can produce expert power even when a specific argument is weak; strong evidence can carry informational power even when the source has low status.
Legitimate power vs coercive power
Legitimate and coercive power can also produce similar outward behavior while relying on different psychological mechanisms.
A person who follows a workplace safety instruction because the safety officer has recognized authority is responding to legitimate power. A person who follows the same instruction because they fear disciplinary action is responding to coercive power. A person may experience both at once.
The distinction matters because observable compliance does not reveal why the person complied. This is one reason the psychology of obedience to authority cannot be reduced to the existence of punishment. Obedience can involve legitimacy, identification, norms, situational commitments, perceived expertise, and other processes.
Position power and personal power
French and Raven’s taxonomy is often simplified into “position power” and “personal power.” That can be useful as a secondary grouping.
Legitimate, reward, and coercive power often accompany formal positions because organizations assign roles control over rules, benefits, and sanctions.
Expert and referent power often depend more on the individual person’s competence, reputation, relationships, or identification.
Informational power cuts across the distinction because persuasive information can be supplied by people with or without formal position.
The grouping is therefore a practical summary rather than a replacement for the six bases. A formal role does not automatically create strong reward or coercive power, and a high-status individual does not automatically possess relevant expertise. For a broader treatment of formal and informal resources, see Personal Power vs Positional Power.
Power bases are not influence tactics
A base of power is a resource or foundation for influence. An influence tactic is a behavior used to try to influence someone.
Rational persuasion, consultation, exchange, pressure, coalition building, ingratiation, inspirational appeals, and personal appeals are examples of tactics discussed in organizational influence research. The same tactic can draw on different power bases, and the same power base can support different tactics.
For example, a manager may use an exchange tactic because they have reward power. An engineer may use rational persuasion while drawing on informational and expert power. A political leader may use an inspirational appeal drawing partly on referent power. A supervisor may use pressure supported by coercive and legitimate power.
Confusing bases with tactics makes the model look more rigid than it is. Raven’s later Power/Interaction Model explicitly treated influence as a process in which agents assess available power resources and choose strategies in context (Raven, 2008).
People usually use several bases of power at once
Real influence rarely arrives in a pure laboratory category.
A physician can possess legitimate role authority, expert knowledge, referent power created by trust, and informational power when explaining the evidence behind a treatment recommendation.
A manager can combine legitimate power with reward power, expert power, and referent power. A parent can combine legitimate family authority, rewards, sanctions, expertise, identification, and explanation.
The framework therefore works best as an analytic map. It helps identify which mechanisms are operating, how heavily the target depends on the agent, and what may happen if one resource disappears.
This is also why no base can be labeled universally “strongest.” Effectiveness depends on the target, objective, relationship, time horizon, credibility, alternatives, norms, and consequences of the influence attempt.
How later research expanded the six bases
The six-base framework became more differentiated as researchers tried to measure it more precisely.
Raven, Schwarzwald, and Koslowsky’s 1998 instrument distinguished 11 forms: personal and impersonal reward; personal and impersonal coercion; legitimate power based on position, reciprocity, equity, and dependence; expert power; referent power; and informational power. Their factor analysis did not simply reproduce 11 perfectly separate factors; it yielded seven factors and broader “harsh” and “soft” categories, illustrating that the empirical structure is more complex than a six-box diagram (Raven et al., 1998).
Earlier measurement work also showed why careful operationalization matters. Frost and Stahelski developed Likert measures designed to address scale problems in prior research and reported factorially identifiable measures of the original five bases in workgroups (Frost & Stahelski, 1988).
The evolution of the measurement literature is a useful warning: the taxonomy is a theoretical framework whose categories require valid measurement. A familiar label such as “expert power” is not itself evidence that a study has measured expertise-based influence well.
What does the evidence say about outcomes?
The evidence does not support a single rule such as “personal power is good and position power is bad.”
Research has linked different power bases to outcomes including satisfaction, commitment, performance, and perceptions of leader effectiveness, but results depend on setting and measurement. Carson, Carson, and Roe used meta-analytic correlations to examine relationships among the classic bases and outcomes such as satisfaction with supervision, job satisfaction, and performance (Carson et al., 1993). Their work also tested a revised structural model rather than assuming that French and Raven’s original propositions would automatically fit accumulated data.
At the same time, Podsakoff and Schriesheim’s major 1985 review concluded that methodological weaknesses made substantial parts of the earlier field literature difficult to interpret (Podsakoff & Schriesheim, 1985). That critique remains important when popular summaries claim that a particular base “always works” or inevitably creates a particular emotional response.
Later studies and applications demonstrate that the framework can be operationalized and used across organizational, health, and educational settings, while also showing that context matters. The model remains influential as a language for analyzing influence resources; it is not a universal predictive law.
Why coercive and reward power often depend on surveillance
Reward and coercive power commonly depend on whether the agent can observe relevant behavior and connect outcomes to it.
A bonus cannot reinforce a performance criterion if performance is invisible or the person granting the bonus cannot determine who met the criterion. A threatened penalty loses force when violations cannot be detected or sanctions cannot be imposed.
This makes these bases especially dependent on credible contingencies. The target must believe that the agent can recognize what happened and connect behavior to reward or punishment.
Expert, referent, and legitimate power can continue to influence without continuous monitoring because the target may act in accordance with accepted expertise, identification, or normative obligation. Informational influence can become even more independent of the agent once the target has accepted the reasoning itself.
These are tendencies in the theoretical model, not mechanical rules. Real situations mix motives and power bases, and targets can comply publicly while privately rejecting the basis for the request.
Power can work without agreement
One of the most useful lessons from the model is that behavioral change, private acceptance, and long-term commitment are different outcomes.
Coercive or reward power can produce behavior because consequences matter even when the target disagrees. Legitimate power can produce behavior because a rule or role is accepted. Expert power can produce deference to a source. Referent power can produce alignment with an admired person or group. Informational power can produce internal acceptance of reasons.
The same observable act can therefore have different psychological origins. Two employees may follow the same instruction, one because of authority and another because of evidence. Two patients may follow the same clinical recommendation, one because they trust the physician’s expertise and another because they understand the explanation.
This is why power cannot be inferred reliably from surface behavior alone.
The model is not a ranking of “good” and “bad” power
Popular management summaries sometimes turn French and Raven’s framework into a moral ranking. The original theory does not justify that simplification.
Coercive power can be abusive, but coercive capacity also exists in legitimate safety, legal, and institutional systems. Reward power can support constructive goals, but rewards can also manipulate behavior or distort motivation. Expert power can improve decisions, but experts can be wrong or operate outside their competence. Referent power can support trust and cooperation, but identification can also make people less critical. Legitimate power can coordinate institutions, but legitimacy can be contested or bounded. Informational power can support informed judgment, but selective, misleading, or false information can create apparent persuasion without sound epistemic grounds.
Ethical evaluation therefore requires additional questions: What is the goal? Is the request legitimate? Are the claims true? Are incentives transparent? Are sanctions proportionate? Can the target meaningfully question or refuse? What alternatives exist? Who bears the risks and benefits?
The framework explains bases of influence. It does not itself determine whether an influence attempt is ethical.
Power, authority, status, dominance, prestige, and control are different concepts
The bases of social power become much clearer when adjacent concepts are kept separate.
Social power is the broader capacity to influence others or affect outcomes.
Authority is power recognized as having some legitimate basis. Legitimate power is therefore especially relevant to authority, but social power includes many non-authoritative forms.
Status concerns social rank, respect, or standing. A high-status person may have referent or expert power, but status is not itself identical to a French–Raven power base.
Dominance is one route through which people can obtain or maintain rank and influence, often involving intimidation or force. It can overlap with coercive power while remaining a distinct construct.
Prestige is influence and status associated with valued competence or freely conferred respect. It often overlaps with expert and referent power, but it belongs to a different theoretical literature.
Personal control is a person’s perceived or actual capacity to influence their own outcomes. It differs from having power over other people. The distinction is developed in Power and Control.
Leadership is a process of influence in groups and organizations. Leaders can draw on several bases of power, but a base of power is not a leadership style.
How to identify the active base of power
A useful way to analyze a situation is to ask what the target believes would happen if the agent made a request.
Ask these questions:
1. Does the target expect a valued benefit from compliance? Reward power may be active.
2. Does the target expect punishment, loss, rejection, or another cost for noncompliance? Coercive power may be active.
3. Does the target believe the agent has a right to make the request? Legitimate power may be active.
4. Does the target believe the agent has relevant superior knowledge or skill? Expert power may be active.
5. Does the target want identification, affiliation, approval, or closeness with the agent? Referent power may be active.
6. Does the target accept the reasons, evidence, or information supporting the request? Informational power may be active.
Then ask whether several answers are true at once. They often are.
This method focuses on the psychological basis of influence instead of classifying people globally. Someone does not simply “have expert power” in every situation. They have expert power to the extent that particular targets attribute relevant expertise to them in a particular domain.
Examples across everyday settings
Workplace
A department head assigns a task under the authority of the role: legitimate power. They offer a performance bonus: reward power. They can initiate disciplinary action: coercive power. They are trusted for technical expertise: expert power. Team members admire and identify with them: referent power. They present evidence showing why a new process reduces errors: informational power.
The same interaction can contain all six.
Healthcare
A clinician may have expert power because of medical training and experience. Professional role can contribute legitimate power within defined clinical boundaries. A trusted therapeutic relationship can create referent influence. Explanation of risks, benefits, and evidence can create informational power.
The distinction matters because informed decision-making should not be confused with mere deference. A patient can accept information while retaining agency over the decision.
Education
A teacher has legitimate power within the educational role, reward power through grades or recognition, and sometimes coercive power through permitted sanctions. Subject-matter expertise creates expert power. Trust and identification can create referent power. Explanations and evidence create informational power.
Peer groups and communities
Peers often lack formal legitimate authority but can have strong referent, informational, expert, reward, and coercive power. Approval, inclusion, reputation, ostracism, specialized knowledge, and persuasive arguments can all become resources for influence.
These examples show why formal hierarchy captures only part of social power.
Common mistakes when using French and Raven’s framework
Mistake 1: Saying French and Raven originally proposed six bases in 1959
The 1959 chapter presented five: reward, coercive, legitimate, expert, and referent power. Informational power was added later by Raven. Later retrospectives sometimes discuss the six-base model as the French–Raven framework while citing both 1959 and Raven’s later work. The historical sequence should still be stated clearly.
Mistake 2: Calling informational power “control of information”
Control over access to information can certainly create power, but informational power in Raven’s model refers to persuasion through accepted information, reasons, and evidence.
Mistake 3: Treating legitimate power as a synonym for formal rank
Formal position is an important source of legitimacy, yet Raven’s expanded model also recognized normative forms based on reciprocity, equity, and dependence or social responsibility.
Mistake 4: Treating expert power as objective expertise
Expert power depends on perceived relevant expertise. Objective competence and attributed competence can diverge.
Mistake 5: Treating referent power as charisma only
Charisma can contribute to referent power, but referent power is broader and can arise from identification, trust, admiration, affiliation, shared identity, or valued relationships.
Mistake 6: Assuming coercive power always fails
Coercive power can produce compliance, especially when sanctions are credible. The more useful questions concern what kind of compliance occurs, how durable it is, what costs accompany it, and whether the coercion is legitimate and proportionate.
Mistake 7: Assuming the six bases are mutually exclusive
They often operate together. The model separates mechanisms for analysis; it does not require real relationships to use one mechanism at a time.
Mistake 8: Treating the six bases as a complete taxonomy of every kind of power
Modern psychology distinguishes social power from status, dominance, prestige, personal control, power motivation, authority, dependence, and other constructs. Raven himself expanded the original model. The six bases remain a foundational framework, not an exhaustive ontology of power.
Measurement and methodological limits
Research on the bases of power has repeatedly confronted a basic problem: power is difficult to measure without confusing the source of influence, the target’s perception, the agent’s behavior, and the outcome.
Early field studies often relied on single-item or ipsative measures that made it difficult to interpret correlations among the bases. Podsakoff and Schriesheim’s review argued that methodological shortcomings substantially limited conclusions from much of that literature (Podsakoff & Schriesheim, 1985).
Later measurement efforts improved scale construction. Frost and Stahelski developed a 23-item questionnaire and reported support for the separability of the five original bases in workgroup data (Frost & Stahelski, 1988). Raven and colleagues later built measures for 11 differentiated bases, finding a more complex factor structure than a simple one-factor-per-label scheme (Raven et al., 1998).
The scientific takeaway is straightforward: French and Raven’s categories are theoretically influential and empirically researchable, while findings depend on how power is operationalized and on the context in which it is measured.
Is there a “best” base of power?
There is no universally best base.
The answer depends on the objective and the relationship. Immediate compliance under an emergency can require different resources from long-term commitment. A technical decision may depend heavily on expertise and information. A formal institution may require legitimate authority. A mentoring relationship may rely strongly on expert and referent power. Reward systems can coordinate incentives. Coercive capacity can enforce boundaries or rules.
Research in organizations has often associated expert and referent forms with favorable attitudes and commitment, while coercive power is frequently associated with less favorable reactions. Yet the evidence is contextual, and the early literature contains important measurement limitations. Even apparently “soft” forms of power can be misused, and “hard” forms can have legitimate functions under constrained conditions.
A stronger question is: which base is actually operating, is it appropriate to the goal and relationship, and what behavioral and relational consequences does its use produce?
Can power exist without authority?
Yes. This is one of the clearest implications of the framework.
A person can have expert power without any formal office. They can have referent power because others identify with them. They can have informational power because their arguments are persuasive. They can have reward or coercive power through control of resources even when they lack recognized authority.
Authority is therefore one route through which power can be organized, especially through legitimacy, rather than a synonym for power itself. See Power vs Authority for the full comparison.
Can someone have power without high status?
Yes. Status and power often correlate, but they are distinct.
A low-status technician may be indispensable because of unique expertise. A child can influence a parent through affection, information, or control over cooperation. A new employee can have informational power when they discover decisive evidence. A person with high prestige may lack formal authority, while a senior officeholder may have formal legitimate power but little referent power.
For the broader distinction among rank, dominance, prestige, and influence, see Social Hierarchy.
Frequently asked questions
What are French and Raven’s six bases of power?
The six are reward, coercive, legitimate, expert, referent, and informational power.
Who created the bases of social power theory?
John R. P. French Jr. and Bertram H. Raven published the foundational five-base framework in 1959. Raven later added informational power and substantially expanded the model.
Were there originally five or six bases of power?
There were five in the 1959 chapter: reward, coercive, legitimate, expert, and referent. Informational power was added later, so the mature canonical framework is commonly presented as six bases.
What is the difference between expert and referent power?
Expert power depends on perceived relevant knowledge or competence. Referent power depends on identification, admiration, affiliation, or a valued relationship. A person can have one without the other.
What is the difference between expert and informational power?
Expert power depends primarily on trust in the source’s expertise. Informational power depends on acceptance of the reasons or evidence communicated. They often operate together.
What is the difference between legitimate and coercive power?
Legitimate power depends on a recognized right to make a request and an obligation to comply within some domain. Coercive power depends on the ability to impose an unwanted consequence. The same person can possess both.
Is reward power the same as motivation?
No. Reward power is a social influence resource based on control of valued outcomes. Motivation is a broader set of processes that energize, direct, and sustain behavior.
Is referent power the same as charisma?
No. Charisma can contribute to referent power, but referent power is broader and can arise from identification, trust, admiration, affiliation, shared identity, or valued relationships.
Is informational power just having access to information?
No. In the French–Raven tradition, informational power concerns influence through information, evidence, or reasoning that the target accepts. Exclusive access to information can create other forms of resource dependence, but access alone is not the defining mechanism.
Is “connection power” a seventh French and Raven base?
Connection power appears in some later management and training taxonomies, but it is not one of the canonical six French–Raven bases. The established six are reward, coercive, legitimate, expert, referent, and informational power.
Can one person use several bases at the same time?
Yes. This is common. A manager, clinician, teacher, parent, peer, or public figure can draw on several bases in the same interaction.
Which base of power is most effective?
No base is universally most effective. Effectiveness depends on the goal, target, relationship, credibility, context, time horizon, alternatives, and what outcome counts as success.
Are the bases of power personality traits?
No. They are relational bases of potential influence. A person’s traits can affect how they acquire or use power, but reward, coercive, legitimate, expert, referent, and informational power are not personality traits.
Is French and Raven’s model still used?
Yes. It remains a foundational framework in research and teaching on social and organizational power, and it has been repeatedly revised, measured, criticized, and applied. Contemporary psychology also uses many other constructs, so the model should be treated as one important framework within a broader science of power.
Related Articles
References
Carson, P. P., Carson, K. D., & Roe, C. W. (1993). Social power bases: A meta-analytic examination of interrelationships and outcomes. Journal of Applied Social Psychology, 23(14), 1150–1169. https://doi.org/10.1111/j.1559-1816.1993.tb01026.x
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). Institute for Social Research, University of Michigan. https://books.google.com/books?id=5AoFAAAAMAAJ
Frost, D. E., & Stahelski, A. J. (1988). The systematic measurement of French and Raven’s bases of social power in workgroups. Journal of Applied Social Psychology, 18(5), 375–389. https://doi.org/10.1111/j.1559-1816.1988.tb00023.x
Podsakoff, P. M., & Schriesheim, C. A. (1985). Field studies of French and Raven’s bases of power: Critique, reanalysis, and suggestions for future research. Psychological Bulletin, 97(3), 387–411. https://doi.org/10.1037/0033-2909.97.3.387
Raven, B. H. (1993). The bases of power: Origins and recent developments. Journal of Social Issues, 49(4), 227–251. https://doi.org/10.1111/j.1540-4560.1993.tb01191.x
Raven, B. H. (2004). Power, six bases of. In G. R. Goethals, G. J. Sorenson, & J. M. Burns (Eds.), Encyclopedia of leadership (Vol. 4, pp. 1242–1248). SAGE Publications. https://doi.org/10.4135/9781412952392.n285
Raven, B. H. (2008). The bases of power and the Power/Interaction Model of Interpersonal Influence. Analyses of Social Issues and Public Policy, 8(1), 1–22. https://doi.org/10.1111/j.1530-2415.2008.00159.x
Raven, B. H., Schwarzwald, J., & Koslowsky, M. (1998). Conceptualizing and measuring a Power/Interaction Model of Interpersonal Influence. Journal of Applied Social Psychology, 28(4), 307–332. https://doi.org/10.1111/j.1559-1816.1998.tb01708.x
Yukl, G., & Falbe, C. M. (1991). Importance of different power sources in downward and lateral relations. Journal of Applied Psychology, 76(3), 416–423. https://doi.org/10.1037/0021-9010.76.3.416