Nudge theory – how does it hold up, 15 years later?

A revolutionary theory of how governments can get us to change our behaviour for our own good without interfering with freedom of choice is now 15 years old. Does it stand up?

Nudge theory has had a pretty good run. Ever since it was first set out by behavioural economist Richard Thaler and law professor Cass Sunstein in their 2008 book Nudge: Improving Decisions about Health, Wealth and Happiness, it has progressively caught the attention first of readers – selling more than two million copies – then of politicians and governments. 

Former US president Barack Obama and former prime minister David Cameron set up “nudge units” tasked with turning the theory into practice. As Simon Ruda, a co-founder of the Downing Street unit, pointed out in an essay last year for UnHerd, the revolutionary fervour soon spread. An initial team of seven was established in Number 10 during Cameron’s reign. By 2021, there were more than 400 such units globally. Downing Street’s team was “spun out” of government in 2014 and became a for-profit company selling its services to the public sector and any other government or organisation “seeking to improve people’s lives”. The original team of seven grew organically to more than 250 people with offices in nine major cities around the world, working on projects in more than 35 countries. 

Subscribe to MoneyWeek

Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE

Get 6 issues free

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

“In the early days,” says Ruda, “we felt as if we had discovered a secret sauce and our mission was to share it as widely as possible.” 

What is nudge theory?

The secret sauce: What’s the recipe? Take one bowl of cashew nuts, for starters. Thaler is fond of an anecdote that sheds light on the basic idea, as he told the Financial Times in an interview in 2019. One evening he served a bowl of cashew nuts to guests while waiting for dinner to cook, and they all compulsively grazed on the snack, gradually ruining their appetites. Thaler eventually removed the bowl and hid it in the kitchen, and his guests thanked him for taking it away. This tale would be unremarkable except that such behaviour does not fit with economic models which assume that we are all rational utility maximisers. It might make sense for us to eat the cashews or not, but why would we thank someone who moves them out of reach? Reflecting on such puzzles inspired Thaler to create behavioural economics. 

The basic idea is that humans make consistent errors in judgement when faced with choices. We do not make the decisions that are best for us, nor the ones that we would make ourselves if we reflected on them rationally and summoned up enough willpower. Rather, we do what we are led to by the “ biases” in our cognitive makeup – the rules of thumb and habits that make decisions easier, but not necessarily wiser. 

A “nudge” can help us make better decisions without forcing us to do anything or by changing the economic incentives, but merely by changing the “choice architecture” – the way in which choices are presented to us. In the example, the guests were not forced to eat or not eat the nuts, and when taken away only had to ask if they wanted them back. But the way the offer is made makes all the difference to whether we stuff our faces or not. The approach has been called “libertarian paternalism” – libertarian in that we retain our free choice, paternalist in that we are helped, “nudged”, to do what is considered to be good for us and for society. Beyond Thaler’s dining table, this insight has, proponents say, led to many small changes in government policy that have nudged people into making better decisions. 

Once upon a time we had to decide for ourselves whether we wanted to set up a private pension plan, for example. Now we are enrolled into one by default and given the choice to opt out if we like. With just that one small nudge, an ambition of both Labour and Conservative governments was quickly achieved, as Stuart Mills points out on The Conversation. 

More than 90% of eligible private sector workers are now members of workplace pension schemes. Small changes in presentation have led to increases in prompt tax payments. Health-related nudges have led people to make healthier choices in diet. Amsterdam’s Schiphol International Airport placed fly-shaped sticks in the men’s urinals to improve aim and brought down the costs of cleaning by 80%. Other examples abound (see the Behavioural Insights Team website). The theory has triumphed. 

So, what's the issue with nudge theory?

From the book’s first publication, some critics have feared that nudge would turn into little more than a psyops operation on behalf of an overweening state with the aim of tricking us into doing what others think is good for us, whether we want to or not. Nudge is not, they say, “libertarian paternalism” at all, nor could it be as this is, obviously, a contradiction in terms. It is simply paternalism – rule by an authority that thinks it is a better judge of what we need or desire than we are ourselves. Worse, it risks tipping over into totalitarianism as every aspect of our lives is nudged and prodded into a shape more pleasing to bureaucrats. This seemed like a stretch to the ideologically uncommitted, even to fellow libertarians. 

Christopher Snowdon, writing for the libertarian-leaning Institute of Economic Affairs (IEA) in 2017, argued that there was nothing to worry about and that a state that overturned its legislative programme to embrace nudge would, in fact, be more libertarian than any country currently in existence. Ruda was of a similar mind and thought the criticisms were “quite frankly nuts” given how well-intentioned his colleagues in the Downing Street nudge unit were, how committed to the scientific method and to doing good. Then came Covid. 

In addition to mandates and punishments, the government deployed nudge tactics to increase compliance with their pandemic-related guidance. Government propaganda posters sought to raise the levels of fear of the disease in the population, for example. Text messages were written in such a way as to boost vaccine uptake. The government made mistakes during Covid that were a direct result of the nudge unit’s success in changing attitudes, says Ruda. 

Using behaviour-focused models might, for example, give rise to messaging that improves parental involvement in a child’s education, and educational attainment. But using those same models to invoke fear to convince people to stay at home during a pandemic is not appropriate because the negative long-term consequences of this will be missed in a typical trial and effects that are real, but hard to measure, ignored – an erosion of social relations and trust in institutions, for example. The fear thus stoked then drove policy-making decisions in a worrying feedback loop. The nudge unit was also responsible for encouraging empiricism at the expense of other fundamental principles of good policy-making, says Ruda, placing all value on data at the expense of reflection, reason and debate, and obscuring the limitations of data as a depiction of reality. 

In short, the kind of mission creep that we should by now have learnt to expect from governments meant that nudge theory was being transformed into an acceptance that the state should use its heft to influence our lives without the accountability of legislative and parliamentary scrutiny. This should have been eminently predictable from the start. After all, if nanny really does know best, then why stop with a nudge when a shove would work better?

Do nudges really work?

Still, the whole debate over whether nudges are authoritarian or not, useful or not, becomes somewhat moot if they don’t actually work anyway. 

One paper among many on this theme from academics Stefano DellaVigna and Elizabeth Linos in the journal Econometrica, “RCTs [randomised controlled trials] to scale: comprehensive evidence from two nudge units”, shows that nudge policies often yield small or null results, as economist Timothy Taylor reports on his blog, The Conversable Economist. The authors found that the impact of policies pursued in the real world is much smaller than the effect reported in academic studies. Boffins setting up experiments with their students, in other words, tend to find more support for their theories than practitioners operating in the real world, where “the average effect is pretty close to zero”. 

Another major meta-analysis, which combines multiple studies to get the most reliable data, reported finding no evidence of nudges working, once “publication bias” – that is, the tendency of researchers to cherry-pick the good studies to publish – is taken into account, says Mills. 

Perhaps this shouldn’t be a surprise. Nudge is a theory based on a science that came into being when economists, faced with the fact that real human beings don’t act in accordance with their theory, decided that there must be something wrong not with their theory, but with the people, and set out to correct them. 

Behavioural economics is, as John Staddon points out on Mises.org, a pseudoscience in that it does not present a theory of causation from which predictions can be deduced and tested, but rather a list of observed effects that are assumed to apply universally but, in fact, do not. 

The plain truth is that we simply do not know the processes by which people arrive at their decisions. Behavioural economics is, in Hayek’s phrase, a pretence of knowledge. The results obtained by behavioural economics, and the theory that describes them, are a property of groups of people operating under very restricted conditions. They do not represent general properties of human nature. 

“The method is just opinion polling, albeit with a clever set of questions,” and nudge policy, to the extent that it relies on behavioural economics, “relies upon a fallacy.” 

But don’t let that put you off – nudge policy will almost certainly “continue to prosper” because it is a “way of gaining political capital from cheap solutions to difficult problems”, as Mills points out. 

Fundamental reform and radical change is politically challenging and expensive. Nudges are cheap and likely to be less controversial to the electorate than new bans or mandates. “So perhaps nudges do not even need to work to continue to have a role in modern society, because politicians will always demand a tool like them for their political ends. It is an argument that could nudge nudging along for at least another 15 years.”

This article was first published in MoneyWeek's magazine. Enjoy exclusive early access to news, opinion and analysis from our team of financial experts with a MoneyWeek subscription.