Three reasons why nudges fail
Pensions policy in a pickle
Since 2012, people in the UK have been 'nudged' into saving for their retirement, by being automatically enrolled into a pension scheme.
The results are in, based on a large scale new study (Beshears et al - link below).
Autoenrollment…
We'd suspected autoenrollment might cause some people to borrow a bit more. Now know that this is indeed the case.
All in all, an ambiguous result. The autoenrollment nudge isn't quite as good for consumers as hoped.
It's not just pensions - the wider nudge narrative is downbeat
The narrative around nudge remedies is considerably more downbeat than a few years ago. On the whole, nudges haven't lived up to the hype.
Sure, nudges may still be net positive for society (especially as they are typically a low cost remedy to implement). But nudges don't tend to radically improve consumer outcomes or competition.
There are many reasons why regulatory and policy interventions can fail. When it comes to nudges, there are three common reasons...
Fail reason #1: binding constraints
There is only so much time in the day, and money in the bank.
Some nudges compete in the attention economy, and the attention economy operates within a 24-hour day. If I carefully read and responded to all the messages I received from my energy supplier, broadband provider, mobile network, water company, car insurer, home insurer, and bank… I wouldn’t have any time for Netflix. And I actually enjoy Netflix.
Other nudges compete for the contents of our wallets, such as pensions autoenrollment, or the FCA's failed attempt to nudge people into paying off their credit card debt more quickly. People face tough choices about how to spend and save their money. You can't nudge someone into having more disposable income.
Fail reason #2: strong self-correcting mechanisms
People and markets adjust and adapt to nudge remedies. These self-correcting mechanisms are also known as negative feedback loops.
People acclimatise to warnings. What makes us pause for thought today, will just be noise tomorrow. This should cause some pause for thought when it comes to nudges which involve sending consumers risk or fraud warnings.
Firms re-optimise around the nudge remedy, finding new ways to influence consumer behaviour. If the nudge changes one element of the choice architecture, firms may be able to change the rest of the choice architecture to neutralise the effectiveness of the nudge. For example, US banks re-optimised and successfully undermined a nudge remedy aimed at reducing the likelihood of consumers dipping into their overdrafts.
This kind of re-optimisation can happen with all remedies, not just nudges. (Of course, the one regulation that shouldn’t fall into this trap is the Consumer Duty – as it requires firms to deliver good consumer outcomes.)
领英推荐
Fail reason #3: weak self-reinforcing mechanisms
In an ideal world, the nudge will change consumer behaviour in the short term.
Then, in the long run, consumers will learn that this new behaviour leads to better outcomes. And so consumers want to continue to follow the new behaviour. This is an example of a self-reinforcing mechanism, or positive feedback loop.
The problem is that it is hard for consumers to learn when it comes to financial services (see here: https://www.fairerfinance.com/insights/blog/when-will-they-learn).
Self-reinforcing mechanisms are sometimes too weak to multiply the impact of the nudge remedy.
What next? Behavioural systems thinking
Markets are populated by real people, who are subject to cognitive limitations and thus different biases, emotional responses, and ways of thinking. Firms are using behavioural insights for commercial ends (e.g. deceptive patterns). So, to understand competition, you have to understand these behavioural dynamics.
Ditching behavioural analysis is not the right answer for policymakers. But having identified the problem, it is perhaps too easy to reach for the nudge lever, when other levers would be more impactful.
This is exactly the criticism of behavioural science levied by the academics Nick Chater and George Lowenstein. They argue for a bit less focus on the individual, and a bit more focus on the system.
In many cases 'the system' is what economists would describe as the market dynamics, populated by networks of agents with varying characteristics and objectives.
'Systems thinking' is not new. But behavioural science is only starting to wake up to its implications. Expect to see more references to behavioural systems thinking in the years to come. In the meantime...
Three practical steps for policymakers
Nudging the pensions narrative
Overall, the pensions autoenrollment nudge is probably still net positive for people. It's just that it didn't turn out to be a slam dunk.
But when has policy ever been simple? There are always trade-offs. Winners and losers.
I'm looking forward to seeing more analysis exploring the impact of the autoenrollment nudge for those on low incomes, and whether there are ways of mitigating any unintended consequences.
Sources
?
Founder and Chief Executive, Plain Numbers. Veteran.
9 个月Many thanks for this great article, Tim - and I couldn't agree more with the need for #systemsthinking rather than suggesting that some tinkering at the edges will solve what are often structural problems. What we're seeing at Plain Numbers is that applying our Approach (which is not just about behavioural science but does incorporate BI) can lead to a substantial increase in customer comprehension from circa 1 in 3 showing a reasonable level of understanding to circa 2 in 3 - so far more positive impact than any 'pure' BI interventions we've seen. However, we fully recognise that parallel system-change work is needed if we are to enable that remaining third of customers to understand, make informed choices and achieve good outcomes.