What is libertarian paternalism?

The paternalistic element is an intention to influence a person's decisions in that person's interest. The libertarian element preserves the possibility of choosing another option. Thaler and Sunstein set out the proposal in two publications in 2003, including an article in the American Economic Review. Their 2008 book, Nudge, brought the approach to a wider audience.

The argument begins with how choices are presented. Ordering a list, preselecting an option, or setting a reference point can influence decisions. Since a decision environment already shapes choice, the authors argue that it can be arranged deliberately to promote people's welfare. In their account, that welfare is assessed by reference to the people concerned and their own judgments.

Two questions therefore need separate answers. One concerns the decision process: does the arrangement influence behavior? The other concerns justification: is this a desirable use of that influence? Evidence that a default changes enrollment establishes an effect on behavior. Assessing the default's purpose and consequences requires an account of whose interests it serves.

How is it related to nudging?

Libertarian paternalism offers a justification for intervention. A nudge changes the choice environment while preserving options and avoiding substantial changes to economic incentives. Defaults, reminders, and changes in the visibility of information can operate this way. A requirement enforced by a penalty or a large financial incentive uses a different form of intervention.

A technique does not reveal its beneficiary. Making an additional purchase easier may benefit a seller while offering little to a buyer. The paternalistic claim requires an explanation of the chooser's welfare rather than the choice architect's good intentions. Sunstein's short guide to nudging (2014) also places choice architecture within questions about transparency and appropriate uses.

The same distinction applies to organizational transformation. Tool adoption may reduce duplicate entry and make work easier, or it may mainly improve management reporting while increasing entry time for employees. The behavior count does not settle that difference. A credible justification considers the effects on the people whose practices the intervention seeks to change.

What do workplace examples reveal?

Automatic enrollment in a retirement plan

Madrian and Shea (2001) studied automatic enrollment in a US employer's 401(k) plan. Participation rose substantially, and many employees retained the default contribution rate and investment allocation. Their analysis highlights inertia and the possibility that employees interpret a default as an employer recommendation. The findings show that the initial setting can shape more than the decision to enroll.

That influence gives the designer a responsibility to examine the default itself. Different employees face different financial circumstances and preferences. Making enrollment easier and choosing an appropriate contribution setting are related but distinct tasks. The evaluation needs to consider what participants retain and how the arrangement serves them.

Making food options more visible

Consider a hypothetical workplace cafeteria placing fruit at eye level and sweet desserts in a less prominent location, with both available. The arrangement seeks to influence choice through presentation. Its paternalistic justification depends on the intended health benefit, users' preferences, and practical access to alternatives. The physical change supplies an intervention to test, while its effects and legitimacy still require examination.

Reserving a place in optional training

Imagine an employer automatically reserving employees a place in an optional course, with a simple cancellation route. The arrangement could make wanted learning easier to access. Its consequences depend on whether the content fits employees' needs and whether they have time to participate. The meaning of “optional” also depends on how managers respond when someone declines.

A cancellation button addresses one part of choice. Employees may still anticipate that refusal will be visible or interpreted as low commitment. The intervention therefore needs an account of the social and managerial conditions around declining. The design can be reviewed when those conditions make the alternative difficult to exercise.

What are the limits in employment relationships?

Employer and employee interests need separate examination

Many changes offer shared benefits, including less re-entry, clearer decisions, or better information access. Others shift costs toward particular roles. Better reporting for management may require more time from the teams supplying the data. The organizational benefit cannot automatically stand in for an individual benefit.

The diagnosis identifies who gains, who carries additional requirements, and which trade-offs are open to revision. Those findings inform the justification for encouraging adoption. Employees may value benefits differently from the sponsor, and the design needs a way to hear that disagreement. A declared intention to help becomes more credible when it is open to scrutiny by the people concerned.

Formal choice can be difficult to exercise

An employee may be allowed to decline while anticipating damage to a relationship or evaluation. The practical questions concern whether refusal requires an explanation, who sees it, and whether alternatives are equally accessible. An intervention affects a relationship in which one party has employment authority. The meaning of choice therefore extends beyond the interface or form.

Companies also establish mandatory practices. An explicit requirement creates a different discussion about its purpose, conditions, and consequences. Describing the same requirement as a freely chosen option obscures that discussion. The organization needs to identify the kind of arrangement it is actually implementing.

More of the target behavior does not settle the outcome

Enrollment records establish enrollment, and login records establish access. For training, the relevant inquiry also concerns learning and use in work. For a tool, quality, time, and difficulties may matter as much as frequency. The evaluation follows the intended benefit rather than stopping at the most convenient count.

Differences across groups also matter. An arrangement may work well for employees with predictable schedules and poorly for those with limited control over time. Recording unwanted effects and unequal consequences gives management evidence for revision. The ability to discover such results belongs in the evaluation from the start.

Can behavioral interventions strengthen autonomy?

Hertwig and Grüne-Yanoff (2017) distinguish nudges from boosts: interventions intended to develop people's own decision-making competence. The approach expands attention beyond steering an immediate choice. It asks whether people can acquire skills they can use independently. The relevant outcome concerns what remains available when direct assistance is removed.

Workplace candidates include learning to compare uncertain information, use an explicit decision rule, or examine alternative explanations. These are design possibilities whose effectiveness needs evaluation in the work concerned. A boost needs to be specified as carefully as a nudge. Giving a technique an autonomy-oriented label does not establish that participants gain an independently usable skill.

The COM-B model provides a wider diagnostic view of capability, opportunity, and motivation (Michie et al., 2011). The Behaviour Change Wheel includes functions such as training, environmental restructuring, and enablement. If the problem concerns time, information access, or authority, the intervention needs to address those conditions. Another reminder leaves that explanation largely untreated.

The link between behavioral science and working conditions is consequential here. A person may know what to do and want to do it while lacking the means. An intervention that changes the work arrangement addresses a different problem from one intended to influence a choice. The diagnosis gives the organization a reason for selecting between them.

How can an organization use these approaches responsibly?

The OECD's BASIC toolkit (2019) integrates ethical questions across behavioral intervention stages. It addresses transparency, stakeholder involvement, opportunities for challenge, and monitoring of longer-term or unexpected effects. These recommendations were developed for public policy. Their application in a company needs to take account of employment authority and the people affected by the proposed change.

UNREST uses five questions to structure that discussion:

  • Whose interest is being served? Separate intended organizational outcomes from benefits expected for employees.
  • How can people challenge the proposal? Identify a route for objections and a decision process capable of revising the design.
  • What does choosing otherwise involve? Examine effort, visibility of refusal, and anticipated consequences.
  • Which other interventions fit the problem? Consider skills, rules, resources, and work organization before selecting a technique.
  • What findings would change the decision? Define observations that would justify adjustment or stopping.

These questions express a professional and editorial position rather than a validated ethical scale. They make the reasoning available for discussion by a sponsor and affected employees. Cultural and behavioral transformation includes responsibility for those choices. Intentions need to be discussable, effects observable, and decisions open to revision.

Calling a practice optional while treating refusal as disloyal undermines the freedom the approach claims to preserve. A legitimate intervention needs a credible account of both welfare and choice in the employment relationship. Employees should be able to contribute to defining the problem and evaluating the proposed solution. That participation gives the organization information a higher adoption rate cannot supply.

Frequently asked questions

Is libertarian paternalism the same as manipulation?

Critics raise manipulation concerns when influence is difficult to recognize or challenge. Preserving formal options answers only part of the autonomy question. The purpose, transparency, and practical conditions for choosing otherwise need examination. A description of how the intervention works helps make that discussion possible.

Do all behavioral interventions rely on libertarian paternalism?

Behavioral science supports many types of intervention. Some develop decision skills, while others revise environments or implementation conditions. Libertarian paternalism is one justification for some uses of influence. Each intervention needs its own account of purpose and consequences.

Is a reminder always a nudge?

A reminder can function as a nudge when it changes the decision environment without removing options or substantially changing incentives. Its classification depends on the arrangement in which it operates. A reminder attached to a mandatory requirement may be part of a broader intervention with different functions.

How can an employer tell whether an opt-out is genuine?

Examine the process and what happens when people use it. Relevant evidence includes effort required, who sees the refusal, treatment by managers, and access to alternatives. A policy statement and an interface option need to be considered alongside that experience.

About the author

Maxime Rabéchault is the founder and president of UNREST, a change enablement and organizational development consultancy grounded in applied behavioral science. Trained at Université Paris Dauphine-PSL, the Conservatoire national des arts et métiers (CNAM), University College London (UCL), and Sciences Po Paris, he also teaches as an adjunct at IAE Paris-Sorbonne. His work connects observable professional practices with the organizational conditions that make transformation possible.

References

Thaler, R. H., & Sunstein, C. R. (2003). Libertarian Paternalism. American Economic Review, 93(2), 175–179. https://doi.org/10.1257/000282803321947001

Sunstein, C. R., & Thaler, R. H. (2003). Libertarian Paternalism Is Not an Oxymoron. University of Chicago Law Review, 70(4), 1159–1202. https://chicagounbound.uchicago.edu/uclrev/vol70/iss4/1/

Sunstein, C. R. (2014). Nudging: A Very Short Guide. Journal of Consumer Policy, 37, 583–588. https://dash.harvard.edu/entities/publication/73120378-d6a8-6bd4-e053-0100007fdf3b

Madrian, B. C., & Shea, D. F. (2001). The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior. The Quarterly Journal of Economics, 116(4), 1149–1187. https://doi.org/10.1162/003355301753265543

Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42

Hertwig, R., & Grüne-Yanoff, T. (2017). Nudging and Boosting: Steering or Empowering Good Decisions. Perspectives on Psychological Science, 12(6), 973–986. https://doi.org/10.1177/1745691617702496

OECD (2019). Tools and Ethics for Applied Behavioural Insights: The BASIC Toolkit. OECD Publishing. https://doi.org/10.1787/9ea76a8f-en