You just moved into a new apartment. It’s clean and cool, and the smart thermostat is pre-set to an energy-efficient temperature. It’s a bit lower than you usually set it, but it’s actually pretty comfortable (and it’s fun to bundle up and be cozy), so you never end up changing it. Months later, you realize your energy bill is lower, and your carbon footprint is too. You weren’t forced to conserve energy. You could’ve cranked up the heat, but the default setting nudged you to do the right thing—for yourself and the planet.

That’s libertarian paternalism in action, a policy approach that seeks to influence individuals’ behavior in ways that will improve their lives, as judged by themselves, while preserving their freedom to choose. It might seem strange to imagine how those two words could ever coexist, but they do. The concept weaves together economics, psychology, and public policy, and the ethics of its implementation are widely debated. It uses insights from behavioral economics, such as cognitive biases and decision-making heuristics, to design nudges that steer people toward better outcomes without restricting any options or imposing mandates. It's about designing a world where the healthy, smart choices are the easiest ones, but never the only ones, to make.

Libertarianism: A political philosophy emphasizing individual freedom, limited government, and voluntary choice. In this context, it stresses that people should be free to make their own decisions without coercive interference.¹

Paternalism: An approach in which authorities or institutions intervene in people's decision-making for their own good, often by overriding personal preferences. It typically assumes that the intervener knows what’s best for the individual, even if the individual disagrees.¹

Behavioral Economics: A field that combines insights from psychology and economics to better understand how people make decisions, especially when they deviate from rational, utility-maximizing behavior. It provides the empirical foundation for many nudges and policy interventions.

Nudges: Subtle changes in the way choices are presented that steer people toward better decisions without restricting their options or significantly altering economic incentives. They preserve freedom of choice while influencing behavior in predictable ways.

Choice Architecture: The design and organization of the context in which people make decisions. By structuring choices—such as setting defaults, ordering options, or framing information—choice architects can influence outcomes without eliminating any options.

Default Options: Pre-set choices that take effect if an individual does not actively make a selection. Defaults are powerful nudges because many people stick with them due to inertia, perceived endorsement, or decision fatigue.

Framing Effect: When different presentations of the same information lead to different choices. Effective nudges often rely on framing to highlight benefits or risks in a way that guides better decision-making.

Bounded Rationality: The idea, introduced by Herbert Simon, that people aim to make rational decisions but are limited by cognitive capacity, available information, and time constraints. Instead of optimizing, individuals often satisfice, seeking a good-enough solution rather than the best one.

Bounded Self-Control: The tendency for individuals to struggle with following through on their long-term goals due to internal limitations like temptation, procrastination, or lack of willpower. It explains why people may act against their own intentions, such as failing to save money or maintain a workout routine.

Hyperbolic Discounting: A behavioral bias where people disproportionately prefer smaller, immediate rewards over larger, delayed ones. This leads to time-inconsistent decisions, such as prioritizing short-term gratification despite long-term costs, and is central to understanding behaviors like addiction or under-saving.

System 1 and System 2 Thinking: From Daniel Kahneman’s dual-process theory, System 1 thinking is fast, automatic, intuitive, and emotional, while System 2 thinking is slow, effortful, logical, and deliberative. Many cognitive biases and decision errors arise from overreliance on System 1, which nudges often seek to influence.

Like many stories in the world of behavioral economics, the history of libertarian paternalism begins with two powerful minds joining forces. In this case, it was Richard Thaler, a behavioral economist who would later win the Nobel Prize in Economics, and Cass Sunstein, a renowned legal scholar. In the early 2000s, they noticed something that may seem obvious to anyone who’s ever observed (or been) a human: people often make decisions that aren't quite aligned with their own long-term best interests. This insight, born from the growing field of behavioral economics and forming one of its core concepts, highlighted that human decision-making isn't always the perfectly rational process that traditional economic models assumed. Instead, we’re influenced by a whole host of psychological biases. 

Recognizing these very human tendencies, Thaler and Sunstein proposed a seemingly oxymoronic idea: "libertarian paternalism." They first formally introduced the term in a 2003 article in the American Economic Review, soon followed by a more in-depth exploration in the University of Chicago Law Review that same year.¹ Essentially, they argued that it's both possible and legitimate for institutions, whether public or private, to "nudge" people towards better outcomes without restricting their freedom of choice. 

Their famous example involves a cafeteria director deciding how to arrange food. In any case, the director has to make a choice about where to place the different food items. By strategically placing healthier options at eye level or at the beginning of the line, the director is nudging people towards healthier eating, but no one is forced to choose the salad over the chocolate cake. Thus, people are free to make their own choices (the libertarian component), but the thoughtful setup makes it easier for people to choose what may be better for them in the long run (paternalism).¹ 

This concept of libertarian paternalism really hit its stride with the publication of Thaler and Sunstein's bestselling book, Nudge: Improving Decisions About Health, Wealth, and Happiness, in 2008. The book’s accessible language and relatable examples brought the ideas to a much wider audience, moving them beyond academic circles and into the mainstream public discourse. This included policymakers around the world whose interests were piqued by the idea that small, low-cost interventions could significantly improve outcomes, from increasing retirement savings to boosting organ donation rates. This is when we saw the rise of "Nudge Units" in governments, perhaps most famously in the UK with the Behavioral Insights Team (BIT), established in 2010.² The Obama administration in the US also embraced behavioral science, with Sunstein himself serving as Administrator of the Office of Information and Regulatory Affairs from 2009 to 2012, helping to integrate these insights into policy.³

As libertarian paternalism gained traction, it also sparked vigorous academic and public debate. Critics, naturally, had some strong opinions, and one of the primary concerns revolved around the very word "paternalism." For many, paternalism evokes images of an overreaching state, dictating choices rather than enabling them. This concern is certainly valid: historically, when governing bodies have made decisions on behalf of citizens as if they were children, it perpetuated unequal power dynamics and discrimination. Paternalism has even been used to justify slavery, arguing that enslaved people needed to be cared for and controlled, and that ownership of another human somehow protected them.⁴ In medical situations, the paternalistic model of the physician-patient relationship can discourage patients from actively participating in their own healthcare decisions.⁵ 

This dark side of paternalism has critics worried that even "soft" paternalism, if not carefully constrained, could slide down a slippery slope towards more coercive forms of intervention. Some argued that nudges, through subtly influencing choices, could be manipulative or undermine individual autonomy, even if freedom of choice was technically preserved.⁶ If people are unaware they are being nudged, are their choices truly free?

Another critique focused on the "libertarian" aspect. Some libertarians argued that any form of government intervention, even a nudge, is an infringement on individual liberty. They questioned whether it's truly "libertarian" if the government is intentionally shaping choices, regardless of how gently.⁷ Furthermore, questions were raised about the legitimacy of those who would be choice architects, deciding what constitutes a "better off" outcome. Who decides what's good for people, and on what basis? Is it always clear what an individual's "true" preferences would be if they were fully rational and informed?⁷ (Hint: the answer is no, people’s preferences are often a mystery, even to themselves). This opened up discussions about the potential for errors in judgment by governments or institutions themselves.

Thaler and Sunstein, along with other proponents, consistently responded to these critiques. Mainly, they emphasized that paternalism is often unavoidable because choice environments are never neutral. Every arrangement of options, every default setting, and every way information is presented inherently influences choices. Since someone has to design these environments, why not design them in a way that promotes welfare? They also reiterated that the "libertarian" qualifier is crucial, ensuring that individuals always have the option to opt out or choose differently, which is what distinguishes it from traditional, coercive paternalism.¹ 

Despite the critiques (or perhaps even because of them), the concept of libertarian paternalism continued to evolve. Researchers began to explore its application in a wider array of domains, from environmental policy and public health to education and even international development. The focus broadened beyond just nudges to a more comprehensive understanding of choice architecture, the idea that the design of the environment has a profound impact on behavior. This also led to a deeper look into the ethical implications of the concept, with scholars grappling with questions of transparency, manipulation, and the potential for unintended consequences.^6,7,8 

Today, the concept of libertarian paternalism is booming, and evidence-based policymaking and behavioral insights units are now commonplace in governments and organizations worldwide. But the debate around the ethics of libertarian paternalism is also as alive as ever. With increasingly sophisticated algorithms and vast amounts of data, the ability to influence choices through "AI-powered nudging" is becoming ever more potent. This might help users understand how they’re being guided or give them more choices because only after identifying a specific behavior they want to change do they actively select the technology and the intervention to use.⁹ This gives users a greater sense of control and awareness over their behaviors and may help them recognize how other agents’ choices may be influencing them. 

Some researchers suggest that the use of VR (virtual reality) and AR (augmented reality) could help users alter the external environment in a way that prompts actions consistent with their goals, values, and beliefs, and therefore override unwanted contextual influences imposed by others.⁹ Unfortunately, the integration of wearable technology, data-tracking, and the inescapability of smart devices also raises new concerns about transparency, privacy, and the potential for manipulation on a massive scale.⁹ 

How do we ensure that these sophisticated nudges genuinely respect individual autonomy and don't lead to a loss of reason-based decision-making? The concept of "dark nudges"—design choices that exploit biases for commercial gain or to lead individuals to make decisions that are not in their best interest—has become a significant area of concern in domains like online shopping or app design. This puts renewed pressure on policymakers and designers to adhere to ethical guidelines. While many behavioral biases are universal, the effectiveness and ethical acceptance of nudges can vary across different cultures and political systems. Research is expanding to understand how libertarian paternalism can be applied effectively and ethically in diverse contexts.⁹

Looking towards the future, some scholars argue that as behavioral science advances, the notion of "libertarian" in libertarian paternalism might need to be re-evaluated. If our choices are so pervasively shaped by our environment, how much true "freedom" do we possess? This pushes the discussion towards deeper philosophical questions about free will and responsibility. As our understanding of human behavior grows and as technology offers new ways to shape choice environments, the ethical and practical challenges will continue to evolve.⁹ The goal, as Thaler and Sunstein originally envisioned, remains to leverage these insights to make it easier for people to live healthier, wealthier, and happier lives, all while preserving the crucial element of individual freedom. It’s a delicate balance, but one that continues to captivate researchers and policymakers alike.

Cass Sunstein

A legal scholar and co-author of Nudge, who helped shape the theoretical and ethical foundations of libertarian paternalism. He has also advised governments on applying nudges in public policy and emphasized the importance of maintaining individual autonomy within regulatory frameworks.

Richard Thaler

A Nobel Prize–winning behavioral economist who co-authored Nudge and is widely regarded as the intellectual father of libertarian paternalism. His research integrates psychology with economics to explain why people often make irrational financial decisions, laying the groundwork for choice architecture and policy nudges.

Daniel Kahneman

While not directly involved in coining libertarian paternalism, Kahneman’s pioneering work in cognitive biases and decision-making under uncertainty provided the empirical foundation for behavioral interventions. His research influenced Thaler and others by highlighting the limits of rational choice theory.

Amos Tversky

A long-time collaborator with Kahneman who contributed to the development of prospect theory and the study of heuristics and biases. His work paved the way to understanding why people deviate from rational decision-making, which is central to the rationale behind libertarian paternalism.

Gregory Mitchell

A legal scholar and psychologist known for his critical analysis of behavioral law and economics. He challenges the practical relevance of laboratory findings for real-world policy, arguing that nudges may be less effective or predictable outside controlled settings, and that policymakers often overstate behavioral insights.⁶

Riccardo Rebonato

An economist and physicist who critiques libertarian paternalism on philosophical and methodological grounds. In Taking Liberties, he argues that nudging undermines individual autonomy and places excessive faith in policymakers' ability to know what is best for individuals.⁷

Glen Whitman

An economist and a prominent critic of libertarian paternalism, particularly its potential to justify overreaching paternalistic policies. He argues that policymakers are also prone to cognitive biases and that nudging risks slipping into coercive or manipulative territory under the guise of benevolence.⁸

Relative to many other economic concepts, libertarian paternalism is incredibly new. Because Cass Sunstein and Richard Thaler’s original paper established the idea quite thoroughly, it may be helpful to focus on unpacking their description of libertarian paternalism and its benefits. Although there have been additional writings on the topic since the original paper’s publication in 2003, these are the core ideas presented in their piece. They argue that their approach is not only possible but also legitimate, rooted in observations about human behavior and the nature of decision-making contexts.

Inevitability of influence

Perhaps (what many consider to be) the strongest argument presented by Sunstein and Thaler is that some form of paternalism isn't just an option, it's often unavoidable.¹ Think about it this way: whenever we design a system, whether it’s how retirement savings are set up, how a website displays privacy options, or even (to return to their main example) how food is arranged in a cafeteria, someone, somewhere, has to make a decision about the defaults, the order, the framing. These choices aren't neutral—they inherently influence the decisions people make.

In their cafeteria example, imagine you’re the director of the lunchroom. You've got all sorts of delicious (and maybe not-so-delicious) options: crispy fries, a vibrant salad bar, chocolate cake, fresh fruit, and so on. Now, you have to put each item somewhere, right? Does it make a difference if you put the salad bar right at the beginning of the line, brightly lit and overflowing with colorful veggies, but leave the fries tucked away in a dimly lit corner at the very end?

What if the default drink option in the soda machine is water, and you have to specifically select soda? Or vice-versa? In both scenarios, you, as the director and choice architect, are influencing behavior. You're not forcing anyone to eat salad or drink water; the fries and soda are still there, available to be selected at will. But by simply arranging the environment, you're making certain choices easier, more prominent, or more appealing. This subtle influence, even if unintentional, is a form of paternalism. You're shaping the environment in a way that guides people, even if you’re not explicitly telling them what to do.

This is the crux of Thaler and Sunstein's argument: since someone always has to design these choice environments, and since these designs always have an impact on behavior, the real question isn't whether we should be paternalistic, but rather, how we can be paternalistic in a beneficial way. For the authors, this means helping people make choices that genuinely serve their long-term interests, without stripping away their fundamental freedom to choose.¹ With this argument established, the conversation shifts from a rigid "should we or shouldn't we" nudge people to a more pragmatic "how can we do this well?"

Bounded rationality and bounded self-control

So, we've now established that choice environments are never truly neutral, and that designing them well is a legitimate endeavor. If designed thoughtfully, Thaler and Sunstein posit, the environments can alleviate some of the limitations we face when trying to make choices that are in our best interest. The first central concept in this argument, bounded rationality, suggests that our rationality isn't infinite; it's inherently constrained. We don't have perfect information, and even if we did, our brains have limited capacity to process it all. 

Think about the struggle of trying to choose a new phone plan, a mortgage, or even just deciding what to have for dinner from a massive restaurant menu. The sheer volume of information, the different features, and the pricing structures can be terribly overwhelming. We often end up "satisficing" rather than optimizing, meaning we pick an option that's "good enough" rather than spending endless hours trying to find the absolute best one. We might miss crucial details, get confused by jargon, or simply suffer from information overload.

Secondly, even when we know what's good for us, and even when we intend to do it, our present self often caves to immediate gratification, leaving our future self to deal with the consequences (also known as hyperbolic discounting). It’s why we hit snooze instead of getting up even if we know it will make us late, or have one more glass of champagne even though we’re well aware of the headache that will follow. So, when libertarian paternalism comes into play, it's not because Thaler and Sunstein think people are stupid or incapable. Instead, they recognize that these inherent human limitations, such as the difficulty in processing complex information, the tendency to stick with defaults, and the struggle with willpower, often lead us to make choices that, upon reflection, even we would agree weren't in our own best interest.¹

This departure from the "rational agent" assumption is the key motivation for libertarian paternalism interventions. Nudges aren't about forcing us to do something we don't want to do, but instead about helping people overcome their own cognitive biases and self-control challenges, making it easier for them to choose what they really want and what will truly make them "better off, as judged by themselves.”¹ 

Steering choices towards welfare

Thus, if we accept that some form of influence is indeed unavoidable and that our human brains are subject to the constraints of bounded rationality and sometimes limited self-control, then the next goal for Thaler and Sunstein is figuring out how we go from passive influence to deliberate, thoughtful guidance with the ultimate goal of genuinely improving people's well-being.

This is where the "paternalistic" part of the equation truly shines, but hopefully with a very important caveat: it's a benevolent, well-intentioned paternalism. The aim isn't to control or dictate, but to gently steer choices towards outcomes that promote welfare, just as the best parents do. It's like a really good GPS, pointing you towards the most efficient route, but still letting you take that scenic detour if you really want to. 

In their outline, libertarian paternalism seeks to guide us toward the decisions we would make for ourselves if only we were perfect.¹ This would mean having complete information (every single fact, every nuance, every long-term consequence laid out clearly before you, without any hidden clauses or confusing jargon), unlimited cognitive abilities (the mental capacity to process all that information instantly, weigh every pro and con flawlessly, and never get bogged down by analysis paralysis), and perfect self-control (always choosing the savings account over the impulse buy, or the early morning train over the cozy bed, without an ounce of internal struggle).

In that ideal world, we'd consistently choose wisely, always aligning our present actions with our long-term goals. Despite what some influencers may want you to believe, no one lives in that world. Libertarian paternalism aims to bridge that gap, not by forcing you to make those "perfect" decisions, but by designing the choice environment so that those "perfect" decisions are simply easier, more likely, or the default option.

You might have already sensed a potential critique here: How do we know what constitutes "welfare" or a "better off" outcome? While this is a legitimate question (which we’ll explore more later), Thaler and Sunstein have tried to establish that this isn't about some distant authority imposing their values. Instead, it's rooted in the idea of what individuals would choose for themselves if they were fully informed and perfectly rational.¹ This often involves respecting revealed preferences, because while traditional economics takes the choices people make at face value, libertarian paternalism acknowledges that what people do choose isn't always what they would choose under ideal conditions. 

They also use proxies for welfare, which could mean things like increased retirement savings, higher vaccination rates, healthier eating habits, or reduced energy consumption, because these are generally accepted as welfare-enhancing outcomes that most people would agree are good for them.¹ The researchers also focus on measurable impacts, researching whether, for example, a particular nudge actually leads to higher savings rates or less food waste, and by how much. Focusing on these measurable outcomes makes the paternalistic aspect more tangible, rather than just an abstract notion of what's "good."

Respecting freedom of choice

While the full, in-depth academic explorations by Thaler and Sunstein weave through many intricate and fascinating layers, for the purposes of this reference guide, the essence of the "libertarian" element boils down to one critical principle: interventions must never eliminate freedom of choice. That means that no matter how cleverly designed a "nudge" is, and no matter how much it steers people towards a demonstrably "better" outcome, individuals must always retain the ability to opt out or choose differently. 

To take one of the authors’ key examples, imagine a company decides that, instead of employees having to actively sign up for their retirement savings plan, they'll be automatically enrolled when they start their job. Right from day one, a percentage of their paycheck is funneled into their future security. This is paternalistic because it steers people towards saving, acknowledging that left to their own devices (or facing complex paperwork), many might procrastinate or simply never get around to it, even if they know saving is vital.¹ The nudge then leverages the power of the default to encourage a choice that will benefit employees in the long run.

The libertarian part of this equation means that while employees are automatically enrolled, they’re always given the explicit option to opt out. They can fill out a form, click a button, or make a phone call to stop the contributions. They can choose to save less, save more, or not save at all through the company plan. The money is still theirs, and the choice is always available, even if the default has been thoughtfully set. 

This right to opt out is the mechanism that differentiates libertarian paternalism from more coercive forms of government or institutional intervention. It respects individual autonomy, acknowledging that people's circumstances are unique and that they, ultimately, are the best judges of their own lives. It ensures that the nudge isn't a shove, and the guidance isn't a mandate.¹ So, when we talk about "libertarian paternalism," the "libertarian" isn't just an afterthought or a political nod, but a fundamental design constraint that ensures these powerful behavioral insights are used to enable freedom, not diminish it. 

Given that the topic is so complex, you probably won’t be surprised to find out that there was significant criticism from other economists and philosophers in the field in response to Thaler and Sunstein’s original paper. For simplicity, let’s look at the biggest critiques from three of the most vocal responders: Riccardo Rebonato, Gregory Mitchell, and Glen Whitman. 

Social and cultural context

Economist Riccardo Rebonato outlines a significant line of criticism: he argues that Thaler and Sunstein’s focus largely neglects the fundamental, pervasive role of broader social and cultural context in shaping our preferences and choices. In essence, he suggests that libertarian paternalism is too focused on the immediate choice architecture and overlooks the very foundations upon which our preferences are built.⁷

He brings in the example of "Mike," a hypothetical individual making what, from a libertarian paternalism perspective, might look like "unhealthy" or "irrational" choices, perhaps opting for sugary drinks, processed snacks, or high-fat foods. If you're looking through the lens of libertarian paternalism, you might explain Mike's choices as simply a result of "irrational impulsiveness" or a series of cognitive biases, like choosing the easily accessible option or being swayed by clever marketing. The solution, then, would be to nudge Mike towards "better" choices by, say, putting healthier options front and center, or making sugary drinks less visible.⁷ 

However, Rebonato contends that this view is far too simplistic and argues that Mike's choices are much more convincingly explained not by mere cognitive missteps, but because these choices are deeply embedded in his cultural, social, and perhaps even “class” identity. Those "unhealthy" items aren't just random selections; they're consumed by his family, shared with his workmates, and enjoyed by his friends. They're part of his traditions, his comfort food, his way of belonging. These communal habits and shared experiences effectively define and narrow the range of what Mike even considers a possible choice or preference within what is socially and culturally acceptable and available to him.⁷ 

On a personal note, I find the constant “healthy eating” examples used by all of the men, both critics and proponents, a bit exhausting. The need to label various foods and eating behaviors as good or bad can be problematic for many reasons and ignores the systemic and often class-related barriers to nutrition, not to mention perpetuates a diet culture that often villainizes any food with a high caloric content. 

Putting this aside, Rebonato views libertarian paternalism's primary focus of manipulating the immediate choice context to exploit cognitive biases as problematic because it essentially treats a "poor choice" as if it were just a "technical accident" in an individual's thinking process. He argues that this approach completely ignores the deeper social and cultural roots of preference formation. By framing the issue purely in terms of individual cognitive errors, libertarian paternalism interventions risk being ineffective, or even worse, harmful, because they fail to address the underlying societal reasons why people form the preferences they do in the first place. Treating competent adults like children uses a narrow view of rationality, which allows choice architects to intervene without adequately understanding or justifying their actions.⁷

Instead of libertarian paternalism's "context manipulation," Rebonato suggests that a far more valuable and respectful approach would be to focus on favoring the social conditions that actually facilitate the formation of an informed and well-considered choice. This would include examining the social or cultural conditions under which people form their preferences.⁷ What are the societal norms, the marketing influences, the community resources that shape what we desire and deem possible? It also means providing appropriate information that is truly accessible and understandable, creating opportunities to experience alternative ways of life, and protecting individuals from distorting influences that actively undermine their ability to make free and informed choices. 

Individual differences 

In contrast to Rebonato’s argument that we should focus more on social and cultural influences, Gregory Mitchell zeroes in on the inherent difficulty in defining welfare and how we can compare it across diverse individuals.⁶ He argues that libertarian paternalism can’t genuinely claim to enhance welfare when it's so incredibly hard to pinpoint what welfare means. This is because it ultimately requires making what economists call "interpersonal comparisons of utility"—trying to figure out if Nudge A made person X better off than Nudge B made person Y, and by how much. This is notoriously difficult, if not downright impossible, especially when we're typically relying on ordinal utility measures (meaning we can say X prefers A to B, but we don’t know by how much more, or if X prefers A more than Y prefers C).⁶ 

Libertarian paternalism often implicitly (or explicitly) favors one type of preference, which assumes a universal best way of being. That definition is probably not going to be consistent across the world, as everyone experiences and defines their own well-being. To go back to the example of Mike at the convenience store, how could we compare him to someone else (say, Sarah) who has had a completely different life, with different goals and tastes? 

Beyond just free choices, perhaps Sarah has different nutritional needs or dietary restrictions. What if she were gluten-free, with an autoimmune disorder preventing her body from being able to process the “healthy whole grains” in wheat products? What if she were a marathon runner, whose caloric needs far surpassed other women of her cohort? Perhaps finding calorie-dense foods with simple sugars and a high fat content would be the optimal choice for her training. Mitchell argues that it would be nearly impossible to define welfare in this way because the differences between people are so complex.⁶ 

Slippery slope 

Meanwhile, Glen Whitman's central claim is that any paternalism, including soft or libertarian paternalism, carries a serious risk of expansion.⁸ While he might agree with other critics' concerns, his primary and most urgent fear is that if we, as a society, start condoning even a little bit of paternalistic intervention, we're placing ourselves on a slippery slope from soft to hard paternalism. For Whitman, this risk of expansion must be factored in as a significant cost when we evaluate any nudging policy. It's a real-world concern that policymakers will, over time, inevitably gravitate towards more and more intrusive forms of intervention.⁸

Whitman provides a compelling array of reasons why policymakers might be tempted down this path. He argues that the perceived success of initial, gentle nudges could breed overconfidence or a desire for even greater impact, leading to incrementally stronger interventions. Furthermore, the very distinction between "soft" and "hard" paternalism, which Thaler and Sunstein emphasize, is, in Whitman's view, incredibly difficult to maintain in practice. Where exactly is the line between subtly steering and overtly compelling? This blurring of boundaries, combined with political pressures and the natural inclination of bureaucracies to expand their reach, creates a powerful gravitational pull towards more intrusive measures. For Whitman, the very existence of soft paternalism risks normalizing intervention, making increasingly forceful policies seem less objectionable over time.⁸

Public choice, benevolent planners, and lack of transparency 

Each of the critics brings up concerns about what happens if (or when) those in charge lean too far into the paternalistic. Mitchell highlights a significant concern right off the bat: compared to "hard" paternalism, which overtly limits our choices (think mandatory helmet laws), libertarian paternalism can be much less transparent.⁶ When nudges are nearly invisible, how do we hold the choice architects accountable? This lack of clarity creates what Mitchell calls an accountability deficit. Rebonato actually builds on this, suggesting that “hard" paternalism, precisely because it’s more visible (and often more annoying), might paradoxically be less dangerous politically.⁷ Its overt nature makes it easier to monitor and challenge democratically because we can vote out governments that overreach. Most libertarian paternalism literature, however, says very little about how we plan to monitor those who are monitoring us and seems to assume our leaders will always have our best interests in mind. 

Rebonato strongly criticizes this apparent assumption of a perfectly altruistic government (or whoever is designing our choice environment). He argues that everyone, including government officials, pursues their own interests, and these probably won’t always align with the public's welfare. Policymakers will inevitably, he argues, start nudging us not for our own good, but for their own political gain or to subtly advance their own agendas. If voters are susceptible to System 1 thinking and manipulation, as Thaler and Sunstein themselves suggest, then the democratic process of holding planners accountable through voting is undermined.⁷ 

Adding another layer of concern, Whitman warns that the slippery slope danger is especially great if policymakers themselves exhibit the same cognitive biases they attribute to the very people they're trying to help. Imagine policymakers falling prey to extremeness aversion, where people avoid extremes, making moderate options seem more attractive. If we introduce some amount of paternalism, even a mild nudge, that might make more paternalism appear to be the plausible middle ground, rather than less. If policymakers are prone to bias, they might unconsciously push for ever-increasing intrusiveness, rationalizing it as a balanced approach when it's actually driven by their own cognitive blind spots.⁸ 

Are biases immutable?

Finally, Mitchell and Rebonato challenge the core assumption of libertarian paternalism, which suggests that our biases are hard-wired and unchangeable. Rebonato, for his part, strongly believes that individuals can actually be trained to become better thinkers.⁷ He argues that fostering true rationality through debiasing (reducing the impact of biases) or direct cognitive training might be a more empowering and sustainable path. He even suggests that if we constantly rely on nudges that exploit our biases rather than encouraging conscious, deliberate thinking, we might inadvertently reinforce those System 1 neural pathways. Therefore, if we're constantly being steered by subconscious cues, we might lose some of our ability for rational deliberation, and our feared rational deficiencies become a self-fulfilling prophecy. 

Returning to the example of the cafeteria, as an alternative to simply reorganizing where food items are located, Rebonato might suggest providing nutrition facts and information about important components of a balanced diet. With each dish labeled, people may be more aware of their choices, helping them move away from (theoretically) mindlessly selecting the most convenient option and move towards mindful choices that fit their needs, whatever those may be. 

Mitchell expands on this, highlighting that, contrary to libertarian paternalism's apparent approach of largely working around cognitive biases, methods for diminishing the effects of choice frames and other biases actually exist. He points to techniques like actively considering alternative or opposing arguments, engaging in critical self-reflection, or receiving other specific training designed to ameliorate our inherent biases.⁶ This isn’t just a way to improve our thinking in general, but it could actually be a direct route to fostering individual freedom of choice. This focus on the value of debiasing is often overlooked, but it’s an important point. We ultimately need to consider whether our goal should be to empower individuals to make better choices on their own terms or to create the best predetermined path (and ensure it comes from the best intentions).^6,7 It’s a powerful call for investing in human cognitive capacity, rather than just designing environments to accommodate our cognitive shortcuts.

Save More Tomorrow 

Thaler and economist Shlomo Benartzi designed a program to help employees who want to save more money but lack the willpower to act on this desire. Chalking it up to willpower may be controversial, as there are clearly additional social, cultural, and economic factors at play, but this program focused on overcoming mental barriers in particular.¹⁰ The essence of the program is straightforward: people commit in advance to allocate a portion of their future salary increases toward retirement savings. 

Employees are invited to sign up for the Save More Tomorrow program, in which their contributions to the savings plan are increased annually whenever they get a raise. They’re approached about these increasing contribution rates approximately three months before their scheduled pay increase, so the timing leverages the fact that people's self-control intentions about the future often exceed their ability to implement them in the present (which relates to the influence of hyperbolic discounting). Once they join, their contribution to the plan is increased (up to a preset maximum), beginning with the first paycheck after a raise, and because the increase is linked to a future raise, it feels less like a loss. Employees can opt out at any time, but the impacts of procrastination and inertia suggest that once enrolled, employees should remain in until they opt out.¹⁰

In the first company to use this plan, the vast majority of those enrolled (80 percent) remained in it through the fourth pay raise. The employees who joined increased their savings rates from 3.5 percent to 11.6 percent in a little over two years (three raises), which amounts to a very real difference in retirement savings.¹⁰ Even though the plan is paternalistic in that it attempts to influence choices to promote welfare, it’s considered libertarian because choices aren’t blocked off and it imposes minimal costs on those who don’t want to take part. It’s also even less intrusive than automatic enrollment, since participants have to take some action to enroll, and it’s even more successful at getting people to save. Proponents of the plan favor this model because it may be just as effective as some of the recent (and far more expensive) proposals for doing so by decreasing or eliminating taxes on savings.¹⁰

On the other hand, critics argue that if nudges are effective and exploit decisional inertia, the nominal right to opt out amounts to very little. They also argue that libertarian paternalism, through programs like the Save More Tomorrow plan, may lead to redistribution of resources from rational to irrational individuals, especially in plans with employer contributions tied to employee contributions, because the rational are less likely to need or use the nudge to maximize contributions and employer matches, while the irrational are steered into doing so. But it’s worth asking, even if despite these concerns, the value of encouraging savings improves the liberty of everyone’s potential future selves.¹⁰

Cigarette availability 

How many consumer products can you think of that, if used as intended, will kill you? Tobacco use causes the deaths of hundreds of thousands of people every year and is the number one cause of preventable death in many high-income countries. Despite the lethal effects of their products, tobacco companies manipulate store displays to make their products as tempting as possible, even though frequent exposure to retail tobacco marketing is associated with a 50% increase in the odds of ever smoking at all.¹¹ That effect is as big as the impact of having someone in your household who smokes. Research shows that the store displays we’re exposed to can have a huge effect on our tobacco consumption, especially for children (almost 90% of smokers start before age 18). Particularly for people trying to quit, being bombarded with smoking paraphernalia and advertisements can make it much harder for people to stick to their goals.¹¹ 

Because smoking and tobacco use are widely seen as an objectively negative externality (as it’s bad for the individual, produces dangerous secondhand smoke, and pollutes the environment, all with very little, if any, positive value), few people object to government regulation of tobacco products. This is what makes cigarettes an interesting example of libertarian paternalism. On one end of the spectrum, regulation can be very libertarian, for example, providing the raw information about tobacco’s effects and posting that “smoking causes cancer” on the website.¹¹ This involves no interference. On the opposite end of the spectrum, there is very coercive paternalism to discourage smoking in many places (most restaurants, airlines, and other public spaces don’t allow smoking).¹¹ 

Somewhere in the middle, though, there’s libertarian paternalism, which nudges people in the right direction without any mandates. This includes graphic warning labels on the packages, making the health effects more salient, and changing the format of cigarette displays and where they’re located within a store. The harder they are to get to, the more the simplification bias is at stake, reducing the amount of products purchased.¹¹ Now the question is, would (and should) the limitations be different if the product were something other than tobacco? 

Nudge Theory 

Libertarian paternalism often uses nudging as a tool to guide people. To better understand nudging and its potential criticisms, read this article, which delves into the definition of nudging, an explanation of different categories of nudges, and some important case studies. 

Why We Need More Than Just a Nudge 

In this perspective piece, Andrew Lewis unpacks some of the biggest limitations of nudge interventions, partly due to the inconsistent messaging coming from different countries’ intervention styles. This article takes a closer look at some of the most pertinent nudging examples from the UK’s COVID-19 response policy.

1. Sunstein, C. R., & Thaler, R. H. (2003). Libertarian Paternalism Is Not an Oxymoron. The University of Chicago Law Review, 70(4), 1159–1202. https://doi.org/10.2307/1600573 
2. Behavioural Insights Team. (n.d.). Our history. https://www.bi.team/about-us/our-history/
3. Sunstein, C. R. (n.d.). Cass Sunstein. Obama White House Archives. https://obamawhitehouse.archives.gov/blog/author/cass-sunstein
4. Porter, J. D. (2021). The archaic roots of paternalism: Continuity in attitudes towards slaves and slavery in The Odyssey, Xenophon’s Oeconomicus, and beyond. Greece and Rome, 68(2), 255–277. https://doi.org/10.1017/S0017383521000061 
5. Wilkinson, D. (2023). The harm principle, personal identity and identity-relative paternalism. Journal of Medical Ethics, 49(6), 393–402.  
6. Mitchell, G. (2005). Libertarian paternalism is an oxymoron (FSU College of Law, Public Law Research Paper No. 136; Law and Economics Paper No. 05-02). Florida State University College of Law. https://ssrn.com/abstract=615562 
7. Rebonato, R. (2013). A critical assessment of libertarian paternalism. SSRN. https://doi.org/10.2139/ssrn.2346212 
8. Whitman, G. (2010, April 5). The rise of the new paternalism. Cato Unbound. https://www.cato-unbound.org/2010/04/05/glen-whitman/rise-new-paternalism 
9. Krpan, D., & Urbaník, M. (2021). From libertarian paternalism to liberalism: Behavioural science and policy in an age of new technology. Behavioural Public Policy, 8(2), 300–326. https://doi.org/10.1017/bpp.2021.40
10. Thaler, R. H., & Benartzi, S. (2004). Save More TomorrowTM: Using Behavioral Economics to Increase Employee Saving. Journal of Political Economy, 112(S1), S164–S187. https://doi.org/10.1086/380085 
11. Berman, M., Miura, M., & Bergstresser, J. (2013). Tobacco product display bans: Policy synopsis. Public Health Law Center. https://www.publichealthlawcenter.org/sites/default/files/resources/nycenter-syn-tobproductdisplaybans-2013.pdf