What Is Libertarian Paternalism? Preserving Choice While Guiding Behaviour

There is a tension at the heart of behavioural design that I encounter again and again. You want to help people make better choices - eat more healthily, save more money, live more sustainably. But who are you to decide what a better choice is? And is it not manipulative to design environments so that people drift in that direction without quite noticing?

This is precisely the tension that Richard Thaler and Cass Sunstein confronted in 2003. Their answer - libertarian paternalism - is perhaps the most influential idea in applied behavioural science of the past thirty years.^[1]

What is libertarian paternalism?

The term sounds contradictory and that is intentional. Libertarian refers to preserving freedom of choice: people may do as they wish without the government or an organisation forcing or penalising them. Paternalistic refers to the idea that choice architects - those who design environments - try to promote the wellbeing of the chooser, even when the chooser has not actively asked for help.

The central claim of Thaler and Sunstein is simple but provocative: there is no such thing as neutral choice architecture. Every environment has a structure. Every form has a sequence. Every menu has a layout. Every pension plan has a default option. That structure influences behaviour whether you intend it to or not. The question is therefore not whether you steer. The question is whether you do so deliberately and in the interest of the person choosing.

This insight is foundational. It makes the classic libertarian position - “just let people choose for themselves” - untenable. Because “just choosing for yourself” does not exist. There is always a setup that makes some choices easier than others. The only question is whether that setup is accidental or deliberate.^[2]

There is no neutral choice architecture. The question is not whether you steer - the question is whether you do so deliberately.

Three core principles that distinguish it from coercion

The opt-out must always be available

This is the hardest criterion. A libertarian paternalist intervention always leaves an easy way out. Automatic pension enrolment is the textbook example: you are enrolled, but you can opt out at any time. The default nudges you toward saving. But if you decide you need the money now, nothing stands in your way.

Compare this with a mandatory pension system. There is no opt-out there. That is paternalistic but not libertarian. And compare it with having no pension system at all and hoping people save on their own. That is libertarian but not paternalistic. Libertarian paternalism sits deliberately in the middle.

In the UK, the automatic enrolment reforms introduced from 2012 under the Pensions Act are a textbook case: participation rates rose from around 55 per cent to above 90 per cent, with opt-out rates remaining low. The default did the heavy lifting. Nobody was forced.^[1]

In cafeterias, the same principle applies. A healthy layout places vegetables and fruit at eye level and at the start of the service line. Crisps and sugary drinks are still available, but they are no longer the first thing you reach for. Nobody is forbidden from choosing anything. The environment simply makes healthy choices easier.

Transparency, not deception

Libertarian paternalism does not operate in the shadows. The choice architecture may be visible. It may even be explained. This is what separates nudging from manipulation. An energy company that silently switches all customers to green electricity without communicating it is manipulating. An energy company that does the same thing, explains it in the welcome letter and offers a straightforward opt-out process is nudging.

This is also why libertarian paternalism works better when the institution behind it commands trust. The tax authority that sends letters informing taxpayers that most people in their area pay on time - a social norm intervention - works precisely because the tax authority is seen as a credible source. The same message from an advertiser would be dismissed as a gimmick.

In the interest of the chooser, not the architect

This is where it gets interesting. Thaler and Sunstein argue that libertarian paternalism is legitimate only when the intervention promotes the welfare of the chooser based on that person’s own preferences - not on what the architect thinks is best, but on what the chooser would want if they were fully informed and thinking carefully.

This principle guards against abuse. A supermarket placing its most profitable products at eye level uses choice architecture in the interest of itself, not the customer. That falls outside the libertarian paternalist definition. It is simply smart shelf positioning - which it always was, just now with a scientific name.

The policy applications are far-reaching: default energy-saving settings on new appliances, automatic enrolment in vaccination programmes, smokers actively choosing smoking rooms rather than non-smokers actively choosing smoke-free zones. All cases where the default promotes the outcome people say they want when you ask them.

The Influence Framework applied

When we analyse libertarian paternalism through the SUE Influence Framework, we immediately understand why it is so effective. Changing behaviour is not a matter of giving people more information. It is a matter of balancing four forces.

Pains (what pushes people away from current behaviour): the pain of the poor choices they are currently making - saving too little, eating too unhealthily, wasting too much energy. But that pain is often abstract and distant. Tomorrow there is always a tomorrow.

Gains (what attracts people toward better behaviour): being healthier, having more saved, a lower energy bill. Those benefits are also abstract. They sit in the future and compete with the immediate needs of right now.

Comforts (what holds people in their current behaviour): the comfort of the default. If you do nothing, you do what the environment has set up for you. This is precisely the force libertarian paternalism harnesses: instead of fighting inertia, it places the default on the right side.

Anxieties (what holds people back from changing): the cognitive load of an active choice. Many people do not enrol in a pension plan because it seems complicated, not because they do not want to save. By making enrolment the default, that anxiety disappears. The decision has already been made, so you can direct your energy elsewhere.

Libertarian paternalism works precisely because it converts the Comfort of inertia - normally a blocking force - into a driving force. And it reduces the Anxiety of change by making the better choice the path of least resistance.

Critique and the limits of libertarian paternalism

Libertarian paternalism is not without critics. The most serious objections come from two opposite directions at the same time.

Strict libertarians argue that any steering is an infringement on autonomy, even if opt-out is available. Setting the default is making a political judgment about what is good for people, and that judgment belongs to the individual.

Strict paternalists argue the opposite: if something is demonstrably good for people - vaccination, seatbelts, pension saving - why not simply mandate it rather than settle for half-measures?

Thaler and Sunstein accept both critiques as internally consistent but point to a third way: in a pluralist society there are domains where coercion is politically infeasible and laissez-faire is proven insufficient. Libertarian paternalism is the pragmatic middle ground. Not perfect, but workable.

The practical limit lies with choices where people have no coherent preferences across time. Pension saving is a good case for libertarian paternalism: most people do want to have enough when they are old, they simply fall short because of present bias. The default helps them achieve what they themselves want. But for choices where people fundamentally disagree about what is good - political choices, religious choices - libertarian paternalism is the wrong tool.

Application in organisations

For anyone who designs environments - HR professionals, marketers, policymakers, managers - libertarian paternalism offers a clear compass. Three questions help determine whether an intervention qualifies:

Question 1: Is opt-out simple and penalty-free? If your organisation automatically enrols employees in a sustainable commuting benefit but opting out involves administrative friction or implicit social pressure, the opt-out is not truly free.

Question 2: Does the intervention benefit the chooser? A cafeteria that places healthy options at the front because that reflects the preference most employees express when asked is libertarian paternalist. A cafeteria that places its most profitable options at the front under the guise of promoting health is not.

Question 3: Is the architecture transparent? Employees may know that their environment has been designed to encourage certain behaviours. Transparency is not a weakness of libertarian paternalism. It is a feature.

In practice this means: set pension contributions high as the default. Make the healthy lunch option the standard in your catering. Set energy-efficient modes as default on all equipment. Make the standard meeting 45 minutes rather than 60. Small adjustments, structural effects.

Frequently asked questions

What is the difference between libertarian paternalism and ordinary government regulation?

Ordinary regulation mandates or prohibits behaviour. Libertarian paternalism always preserves freedom to choose otherwise. Automatic pension enrolment is libertarian paternalist: you can opt out. A mandatory pension scheme is not. The difference lies in the preservation of choice.

Who invented libertarian paternalism?

Richard Thaler (behavioural economist, Nobel Prize 2017) and Cass Sunstein (legal scholar) introduced the term in an influential 2003 paper in The American Economic Review. In 2008 they developed it into their book Nudge, which became the foundation for behavioural policy worldwide.

Is libertarian paternalism manipulation?

That is the central objection. Thaler and Sunstein respond that choice architecture is unavoidable: there is no neutral design of a choice environment. The default must sit somewhere. The question is therefore not whether you steer, but whether you do so consciously and transparently in the interest of the chooser.

How does libertarian paternalism differ from coercion?

With coercion there is no alternative: you have no choice. With libertarian paternalism the opt-out is always available, even if it is not the path of least resistance. A healthy cafeteria layout that places salads at the front is libertarian paternalist. The crisps are still there, you simply do not reach for them first.

What is a practical example of libertarian paternalism in organisations?

Automatic enrolment in workplace pension schemes is the most cited example. But smaller interventions count too: a calendar invite defaulting to 45 minutes rather than 60, an email template that shows sustainable options first, or a restaurant menu that lists plant-based dishes at the top.

Conclusion

Libertarian paternalism resolves a fundamental problem every designer of behavioural interventions faces: you cannot not steer. The environment influences behaviour, always. The question is only whether you know that and whether you act on it deliberately. If you want to learn how to answer that question in practice, the Behavioural Design Fundamentals Course is the best place to start. Rated 9.7 out of 10 by more than 10,000 professionals from 45 countries.

PS

At SUE we work daily with organisations wrestling with exactly this tension: we want to help people, but we do not want to be paternalistic about it. Libertarian paternalism offers not only a philosophical exit from that tension but also a practical handle. Design the default on the right side. Make opt-out easy. Be transparent about what you are doing. That is not manipulation. That is good design.