Can we ‘nudge’ our way to higher growth?

Can we “nudge” our way to a higher rate of economic growth? In a recent speech, David Halpern argued that we should at least try. Halpern was the founder of the Behavioural Insight Team (BIT) that was so enthusiastically championed by then UK prime minister David Cameron, so it is no surprise to find him suggesting that policymaking informed by behavioural science might raise the UK’s growth rate. But is he right? 

Behavioural public policy has come to mean two distinct things. The first is a recognition that people do not necessarily conform to the simple model of human behaviour in an economic textbook. For example, workplace pensions could be set up for anyone who decides to opt in, or for everyone who decides not to opt out. The distinction seems trivial, yet in a real-world setting full of fallible real-world people, the change in the default option makes a huge difference to what people actually choose. 

The second is that new policy ideas are worth testing rigorously, ideally through randomised controlled trials. There is nothing particularly “behavioural” about this idea, but such trials have tended to provide strong evidence that behavioural economics is worth taking seriously, and so the advocates of randomised trials often tend to be the same people as the advocates of using ideas from behavioural economics. 

So, could more psychologically realistic, rigorously evaluated policy lead to higher economic growth? It might. The trouble is that the kind of policies in question tend to be rather small in scope. For example, a decade ago BIT ran an experiment that revealed people tended to miss fewer appointments with the NHS if they were sent timely text message reminders explaining that each missed appointment typically cost the NHS about £160. 

Taking the BIT numbers at face value, such well-aimed texts could prevent more than a million missed appointments and save about £220mn a year. Nobody could object to that, but saving £220mn — about £3.25 per UK resident — is not a growth strategy. 

One approach, then, is to focus on scaling up that sort of experimentation so that schools, hospitals, police and prisons endlessly keep finding improved ways to do things, delivering better results for less. Discovering a new hack like those text messages would be nice. Discovering one a month would be nicer. Discover one a week, and the growth numbers might actually start to improve. 

Cabinet minister Pat McFadden may have had this in mind last December when he announced a small fund to support a “test and learn culture” in government. A good idea, but not a new one. Ubiquitous, evidence-based learning and improvement is essential, but it would have been good to hear McFadden explain how his proposal would advance the “test, learn, adapt” approach advocated by the Cabinet Office back in 2012. 

So far, behavioural insights and rapid experiments have been solutionproducers looking for problems. If one growth strategy is to find and solve thousands of small problems, an alternative is to start with the biggest problems and ask if behavioural science can help. 

In the UK, those problems include: poor skills education, especially for people not bound for university; bad infrastructure; low business investment; a long tail of unproductive firms; and burdensome planning restrictions. 

A behavioural approach might offer some hope here. For example, Halpern praises the apprenticeship levy, which effectively levies a tax on businesses to fund apprenticeships, but offers a tax rebate to those that spend the money on their own training schemes. 

This seems — and is — more fussy than a simple tax break for training, but is structured to take advantage of our tendency to compartmentalise money into arbitrary categories depending on how we plan to use it. The technical term for this is “mental accounting”. The quintessential example is the person who keeps one jar for rent money, a different one for food, another for savings, and a fourth for having fun. The apprenticeship levy harnessed mental accounting by suggesting to firms that they had a chunk of training money sitting in a jar, and if it wasn’t spent the government would take it. (The Treasury, says Halpern, was caught by surprise at how effective the scheme was.) 

A similar approach might be used to encourage firms to invest in capital. Or, to take advantage of an old idea from Nobel laureate Paul Romer, the government could write a law allowing industries to vote for a sector-wide compulsory levy that would fund relevant R&D. The widget industry could collectively vote for a 1 per cent levy on all widget firms, which would fund research projects chosen by each firm, with fruits available to every firm in the sector.

Romer’s idea is a clever piece of economic engineering, but it also makes psychological sense: it funds the public good of innovation, while giving each firm the sense of ownership of a pot of money, and real control over where it goes. 

More broadly, Halpern wants to use this idea of mental jam jars to build support for investment. Perhaps a new tax, or a cut in day-to-day spending, would seem less painful if the benefits were explicitly earmarked as investments for the future of the nation. 

As for infrastructure and planning, I do not know. Maybe what is called for here is not behavioural science, but leadership.

Yet on the subject of leadership, Halpern does hint at another interesting approach: perhaps the government should be a little more upbeat? There is much more to an economy’s performance than Keynes’s “animal spirits”, but there is ample evidence to suggest that animal spirits do matter. 

Businesses are more likely to invest when they feel confident, and while some of that confidence will depend on hard facts about policy and the economy, some of it is really just vibes. It would probably be good politics for Keir Starmer and his team to start sounding a bit more cheerful about the UK’s prospects. It might be good economics too. 

Authored by Tim Harford.

Copyright The Financial Times Limited 2025 

© 2025 The Financial Times Ltd. All rights reserved. Please do not copy and paste FT articles and redistribute by email or post to the web.