Nudge theory: How good intentions can backfire

There’s a restaurant called El Tintero in Malaga. It’s a fun place. Waiters wander around with plates of fish shouting out what they have to offer. Customers put their hand in the air when they see something they fancy and (a wee bit like in an auction house) try to avoid all inadvertent eye contact. All the plates are charged at €7. On the face of it, this makes decision-making pretty straightforward: have what you want and the price will take care of itself. It’s simple and great fun but decidedly mediocre in terms of value. You pay €7 for a dish that locally would be €3-4 and you inevitably end up eating twice as much as you would anywhere else, swept away in the simplicity of uniform pricing and fear of missing out.

Despite its drawbacks, simplicity is one of the key components of nudge theory, described by the behavioural scientist Cass Sunstein (pictured) as “liberty-preserving approaches that steer people in particular directions, but that also allow them to go their own way". Since the publication of his book Nudge in 2008, co-written with Richard Thaler, at least 51 countries have initiated behaviourally-influenced policies at the national level, including the UK, France, The Netherlands, the US and Australia. These behavioural insights are most commonly applied as part of policy implementation and are used to change the behaviour of citizens affected by particular policies. Some of the influences that these policies have had are discussed in a 2014 report by the Economics & Social Research Council and a 2017 report from the OECD.

Perhaps the best example of a 'simple' behavioural intervention is changing a default option. Since 2012, auto-enrolment into pensions savings plans has made the most desirable choice (to enrol) the simplest one to pick, precisely because you don’t have to do anything at all. Can a user experience be any more frictionless than one in which you don’t even have to act?

When issuing rules at the Office of Regulatory Affairs, Sunstein himself promoted simplification “including the use of plain language, reductions in red tape, readable summaries of complex rules, and the elimination of costly, unjustified requirements”.

Nudge theory might be popular with governments and NGOs, but critics have voiced both ethical and practical objections:

• Some people may passively accept defaults that result in unwanted outcomes – like inadvertently ending up with money locked away in pension savings that you can’t access for 35 years.
• Loss of autonomy: People may not reject defaults that would be harmful to them or where there is a significant burden to opt out of unwanted defaults. This is arguably the case in the UK where you have to opt out of pension plan enrolment after you’ve been opted in, and even if you do opt out, you get opted back in again after three years.
• Learning opportunities could be lost: if you don’t have to choose, how do you learn from mistakes?

The debate continues as to how a nudge should be defined (or redefined) (Hansen and Mongin & Cozic) but Thaler and Sunstein effectively rebut the arguments outlined above by putting it this way:

"A nudge is any aspect of the choice architecture that alters peoples’ behaviour in a predictable way without forbidding any options or significantly changing their economic consequences. To count as a mere nudge, the intervention must be easy and cheap to avoid."

Defaults don’t forbid an option or change the economic incentives so they would count as a nudge. However, because defaults are usually easy to opt out of, many of the ethical objections above become more tenuous. Can you claim to have lost your autonomy when you chose not to exercise it? Have you been coerced when you had a low-cost alternative but chose not to take it?

Still, other ethical objections are a little harder to overcome. For example, some interventions may have hidden motives while others may simply be wrong because there isn't enough information to make a good judgement about what the best outcome is. The argument could be made that a default option to invest in retirement may not always be in an individual’s best interests, depending on their individual circumstances.

There is also the potential for manipulation where someone’s decision-making is subverted or perverted by unconscious factors arising from the manipulation, for example the use of subliminal advertising to get people to save more. Sunstein wrote against these objections in 2015, defending seven propositions, which you can read in his essay, Nudging and Choice Architecture: Ethical Considerations.

Whichever side of the ethical argument you land on, there’s evidence that people, although by no means all of them, are happy enough to be nudged by their governments provided there’s no suspicion about the motives of those doing the nudging. However, the severity of the nudge (whether it is a default-change or a simple reminder) can affect attitudes towards it, as can individual characteristics (degree of conservatism, desire for control, level of empathy, and more). There are also international differences in acceptability: in one study German participants were found to have more negative attitudes to nudges than American or Israeli participants.

So, if in general we’re quite happy with governments nudging us, is there a difference between policymakers using frictionless experiences to change behaviour and corporates doing the same?

I, for one, love a simple, frictionless customer experience. I’m a sucker for the €7 plate where I don’t have to think too hard about the trade-offs between cost and value. Sign me up for 1-click ordering so I can get that book just a few seconds faster. Pull me in on a $20 for 12-week introductory deal and just wait to see me flop effortlessly onto the rolling annual contract at $200 a year. I, like you, am only human. My attention is limited, as is my time, and having things just happen without me even having to think about them feels like a win – I’m buying more time for more interesting things.

This is probably a post-rationalisation. I’m fairly sure that at the time, I don’t even notice them to the point where I’m inured to the win and only notice the lack of frictionless experiences. You know, that stuff that happens in the real world, where you end up standing at a restaurant bar for half an hour trying to pay the bill, or you have to show up in the venue to pick up your concert tickets, or your banking app makes you scroll through everyone you’ve ever paid in your life before letting you select the person you pay every week.

Frictionless experiences are noticeable by their absence as much as their presence. Lots of companies are spending lots of time on making their experiences as positive as they can be – it’s a new commercial imperative.

But a frictionless customer experience isn’t just about the beauty of its design, its simplicity, its form or even its function, it's also about the intention of its use. Here, we come back to the ethical objections to nudging: the bias of the nudger or the intention to manipulate someone into a course of action which they would never have chosen of their own free will. Intentions matter.

A few years ago, I was working with an organisation that asked me to find ways for them to be able to offer a large number of their own low-income workers low cost loans. They felt that it would be good for their staff to be able to access some form of low cost credit without being exposed to high cost short term credit providers (payday lenders). In all lending activity, the trick is to get the money you lent paid back. It’s a particularly pressing problem if you’ve lent money at rates that don’t necessarily reflect the credit profile of the customers in question, as was the intention here.

I came up with the bright idea that as an employer you can, with their permission, take the money before it even hits your employee’s payslip so there’s no real credit risk and you can offer loans to your employees at rates they’d never normally qualify for. I had high hopes for this one – there were plenty of good intentions, some positive welfare-enhancements (no late fees, no missed payments and no having to go to payday lenders) and after running some numbers it looked like a no-brainer. Until I worked through the proposal with a loans expert.

It turned out that in the UK, there are hierarchies of household debt, which makes some debts a priority for repayment (because not paying them back can have repercussions including bailiff visits, bankruptcy or even, in the case of council tax in England, a prison sentence). Even though the intention was good, by leap-frogging our place in the credit risk hierarchy by taking the money before it ever became “income”, we’d put our employees at risk of some terrible consequences. The idea was killed that afternoon.

As I discovered, unintended consequences can result from even the most well-intentioned concepts, so while intentionality is a necessary condition for a frictionless customer experience to be ethical, adapting to the specific situation is also required to avoid unintended consequences.

Simple, well-designed customer experiences (and in particular online user experiences) such as 1-click ordering make using a trusted retailer super-easy and convenient. Having an online supermarket shop that auto-fills your favourite items cuts down on the time it takes to do a weekly chore. It’s hard to argue with things that save us time.

But frictionless experiences do change our behaviour. Would you really have bought a third copy of the Blues Brother Soundtrack at 2.30am if you’d had to find your physical credit card and read the numbers? Would I have bought 24 pints of milk when I’m the only one home if my basket hadn’t been pre-populated? Would you buy fewer cups of coffee if you didn’t use contactless payment?

Sometimes a little bit of sludge in a system can lead to a welfare-enhancing outcome. Earlier, I wrote about falling off an introductory rate into a high-price annual contract. A strict opt-in to the higher price at the end of the introductory period would have guaranteed that I didn’t sign up. Even a simple reminder of the change would have been sufficient for me to have hit the cancel button. In fact, if the whole signup process had been a bit harder or had made the annual cost a little more salient (rather than hiding it at the bottom of the scroll in six point light grey text), I would likely not have bought it at all. And therein lies the tension. For the company in question, they hope that I will continue to subscribe, perhaps form a habit, or even not notice (!) the rollover on the annual contract. It fits with their profit motive. And maybe I get what I deserve, a punishment for being lulled into a false sense of security by a frictionless buying experience.

This is where frictionless customer experiences can end up being less consumer-friendly, causing issues like overspending. As more private enterprises use behavioural insights to inform their user experiences, we are starting to see them fall under the scrutiny of regulatory bodies like the Competition and Markets Authority (CMA) in the UK. It is no secret that some hotel booking sites use certain misnamed “nudges”, but as a result of a CMA investigation focused on scarcity messages, hidden charges, promotions and commission-based search rankings, six major hotel booking sites have undertaken formal commitments to change their practices by 1 September 2019. Should they fail to make the necessary changes, the CMA will take further action under Consumer Protection law.

There’s a balance to be struck between the competing elements of the consumer/company relationship. Well-intentioned frictionless customer experiences are likely to be rewarded – there is no legislating against experiences that we enjoy, that make our lives easier or improve our welfare. After all, the seductiveness of frictionless customer experiences lies in their very simplicity – they’re attractive because we are fundamentally attention poor and they take up little of our cognitive bandwidth.

A couple of weeks ago I was back in Malaga, sitting on the wall opposite El Tintero, watching the Sunday lunchtime rush as I waited for my reservation somewhere else. I was going to have to make the difficult decisions about how much to pay, but I was happy to. Sometimes, simplicity just isn’t enough.