Confronting COVID-19: nudge and sludge

Eminent political scientist Professor Cass Sunstein FBA joins Hetan Shah to discuss nudge and sludge in the age of COVID-19. Please note that the timestamps on the transcript refer to the podcast recording of the event, rather than the YouTube video.

Transcript

Hetan Shah: Welcome to this British Academy podcast on Confronting COVID-19: nudge and sludge. I'm Hetan Shah, Chief Executive of the British Academy. We're the UK's national academy for the humanities and social sciences. This event, part of our Shape the Future programme, which brings together researchers to examine a range of issues around COVID-19, took place on 10 June 2020. I was delighted to be joined by the eminent political scientist and Fellow of the Academy, Professor Cass Sunstein, to discuss nudge theory in the context of a global pandemic and what behavioural barriers, or “sludge”, need to be removed if we're expected to nudge our way out of a crisis.

Cass, for those in the audience today who may be a bit less familiar with the concept of “nudge”, can you say a little bit more to explain it, especially in the context of COVID-19?

Professor Cass Sunstein FBA: [00:00:54] Great. So the basic idea is that sometimes it's helpful to have an intervention that preserves freedom of choice, but that helps us get where we want to go. So a GPS device is a nudge: a reminder, saying, for example, there's an event starting at a certain time, given five minutes in advance, that's a nudge. Something saying a bill is due, or you have a doctor's appointment, that's a nudge. A warning is definitely a nudge. A bit of information, say about calories connected with your lunch, that's a nudge. If you're automatically enrolled in a programme, it might be a pension programme, it might be some other programme that your employer just puts you in but you can get out if you want to, those things are nudges.

Reminders of social norms are nudges. If the norm is mask-wearing, you're told that, that's a nudge. The idea is these are interventions that preserve freedom of choice and that's crucial. People can go their own way if they don't like what the GPS suggests, but they do a little steering of people in a direction, which we hope is going to be good for them and if it isn't, then we should change it. There's nudging all over the world right now in connection with COVID-19. In fact, this has been a worldwide effort. There might be a nudge to wear a mask, which involves something like a clear communication that says, “wear a mask, the life you save may be your own,” or “wear a mask, the life you save may be your elderly neighbours”. That's a nudge. These things in stores now, all over the world, which show the two metres, or the one-and-a-half metre distance people should maintain, that might be on the floor of a store, that's emphatically a nudge. New Zealand maybe wins Olympic gold – New Zealand, I don't think wins a lot of Olympic gold – but they might win Olympic gold for 'Nudging 2020'. It's not all they're doing, but they've done a ton of things including nudges that say such things as, “united”, which is signalling a social norm, and “be kind”, which is a very simple way of saying treat others with respect but also, don't get other people sick.

Hetan Shah: [00:03: 14] This must have been a fascinating time to see different governments using behavioural insights in different ways. Can you share your thoughts about how different countries have incorporated social science and behavioural insight, alongside medical or more traditional scientific advice in responding to the pandemic?

Professor Sunstein: [00:03:35] Yes, so thank you for that. There's a shared understanding in diverse countries that our species, homo sapiens, has many incredibly impressive attributes compared to Neanderthals and homo erectus who went extinct. We have some really good attributes, but we have some things that can get us into trouble. For example, we tend to be present-biased, that is, today and tomorrow really matter; the long-term, maybe not so much. We tend to be unrealistically optimistic, which can be a good thing. We tend to be overconfident and many governments have people, sometimes really at the top, who are aware of these, let's say, imperfections, and they've taken steps to try to counteract them. I think that in Australia and New Zealand, both of which have flourishing behavioural science communities, including people in the government, to try to counteract unrealistic optimism by signalling the seriousness of the threat, has been a prime idea. Another idea in behavioural science is, if you want people to do something, the first thing to do is make it easy, rather than shoving them or pushing them, figuring out why aren't they doing it anyway and removing the obstacle.

Maybe a simple obstacle, like you can't find a mask, or if you're staying at home, the economic situation you face is going to be horrific and governments have been responding to that; again, Australia and New Zealand have done extremely well. In the United States, my own country, recently, in particular, states and localities have been alert to the keen importance of social norms and have signalled not, “you're going to be fined if you don't wear a mask”, or “you're going to be fined if you open your store,” but "this is what everyone else is doing, don't you want to join them and be a member of your community?"

These are small things, but in aggregate, they can produce very large effects. Not only that – they have. In nations, Taiwan's another example, where they're really alert to the behavioural framework, to say a little bit about what it concretely entails, but they've been very alert to it in Taiwan and some of their good results are a direct outgrowth of behaviour change, smart behaviour change.

Hetan Shah: [00:06:08] I've certainly noticed the issue around social norms before we went into lockdown. Really quite early on in the pandemic, I saw somebody wearing a mask on the train. I remember thinking to myself, “gosh, that's a bit strange, I wouldn't do that”. Whereas now, of course, a few months on, this is now much more of a social norm and much easier to do, so I think we can all relate to that. How do you see the role of nudging within, I suppose, the armoury of the policymaker, especially when they're currently faced with something that is so life-and-death as a pandemic? Isn't their instinctive response to reach for regulation and law because this matters so much, we couldn't just leave it to something as amorphous as a nudge. How does it all fit together?

Professor Sunstein: [00:06:59] I think if you would have policies that familiarly involve mandates, like don't commit violent crime, or don't sexually harass people, some of the mandates may be accompanied by criminal penalties, some may be accompanied by civil penalties. In the context of violent crime and sexual harassment, we know, particularly for the latter, that if all we relied on was the mandate itself, we'd have a crazy amount, as we sometimes have had, of sexual harassment and violent crime. In both of those contexts, if we want to reduce the incidence, we need to compliment the prohibition and enforcement of the prohibition with nudges.

Now, they can take the form of pretty stern warnings, or they can take the form of efforts to inculcate values of a certain kind, which hopefully, will be part of a democratic process rather than a top-down imposition. Some of the progress we've seen in communities that have seen big decreases in violent crime and sexual harassment, it's been driven largely by nudges, where people are given a framework, maybe by reference to social norm.

Maybe by reference to just triggering a sense of conscience, and what a good person does, maybe by a sense of triggering a sense of empathy with people who could be hurt. These are not, “you do this, and you're going to have to pay a fine or you might face a jail sentence.” They're instead, “this is what we do here”. That notion of this is what we do here, is just consistent with what you said about masks, where, in the easily remembered past, if you wore a mask, it was a signal, "I'm sick" or "I’m a terrified person". Now, it doesn't have either of those signals, it's just that, that is what one does.

For nations that are drawn to mandates and bands, and surely they have an important role to play, to compliment them with nudges is a really good idea. In places like Denmark and the United States right now, and there are countless others, which are opening up a bit and so the prohibitions are weakened, to have simultaneous nudges is essential. The word nudges is a small word, it's just something little, but we have cases where nudges are producing thousands or even millions of changes in behaviour and that's exactly what the world needs with respect to public health today.

Hetan Shah: [00:09:59] Absolutely, I can see that the cumulative effect of some of these small changes may be very major. One of the things that was certainly within the UK, there was a lot of uncertainty as to whether the public would stay in lockdown, or whether they would tire of it and suffer some behavioural fatigue. In retrospect, it looks as if public compliance has been better than perhaps many people had predicted. What do you think we could learn from this episode?

Professor Sunstein: [00:10:32] The word behavioural fatigue is one to be very cautious about. Think of brushing your teeth, where many people do it every day and they don't feel, “I'm tired of brushing my teeth.” They feel that's what they do. Or, of engaging in eating behaviour – it's an elaborate word for lunch, breakfast, dinner – and people don't get tired. Maybe they do, but generally, lunch, breakfast and dinner is a habit. What I think is important for COVID-19 is to notice that the flip side of behavioural fatigue is habit creation and that's, I think, a more robust phenomenon than behavioural fatigue, if the underlying behaviour is one that it is relatively easy to make customary in your life.

The idea of wearing a mask every day, like the idea of bringing out an umbrella when it's raining, “it's raining – umbrella.” That's what the mask is. That doesn't seem fatiguing. That just seems either what one does, or safe. It's raining a lot. I wouldn't worry terribly much about behavioural fatigue. If a nation is closed down in a way that doesn't fit with the data and it's causing severe economic stress, that's self-evidently a problem. It's not behavioural fatigue. It’s how am I going to get through my year? A different point.

Hetan Shah: [00:12:16] I think the point about habits is very powerful and I suppose the interesting question would be how this has played out with different kinds of groups. For somebody like me that can relatively work from home, I've created new habits and this is working pretty well. If you're stuck, I suppose, in a relatively small flat with no outdoor space and you've got young children, it may not be possible to create a habit and therefore, that becomes more difficult.

Professor Sunstein: [00:12:48] It's incumbent on nations to tend to the distributional points, which are signalling, where the reason some groups may face more hardship than others could be purely economic. It could be just a physical small space or it could be kids. These aren't things that can be solved on a dime, but the problems can be reduced pretty quickly through financial support, through accommodations of various kinds that help people with kids and that is, there's an underlying unfairness that's very severe, but there's also a spotlight on the unfairness, which creates an opportunity.

Hetan Shah: [00:13:37] The pandemic has certainly revealed again a whole series of inequalities, which were, for some people, lying beneath the surface. What surprised you most about human behaviour during this period?

Professor Sunstein: [00:13:50] The extraordinary speed of adaptation by people of wildly diverse situations. We've seen, and this is inspiring, we've seen people who are self-important and wealthy, who are not poor but they're staying at home and they're not going into work, about whom it might've been said, “good luck with that.” There are people who are really struggling, single parents who are staying at home and doing an amazing job with their family and often with their work from home. There are emergency room doctors, I have some friends who are emergency room doctors, who are not saying, at the height of the pandemic – and it's pretty bad now but it was worse where I am a couple of months ago – who were saying, “we are raring to go, this is what we trained for, let's go.” These are emergency room doctors who are surrounded by really sick people. The attitude is this, “this is what we're on the planet for” and it wasn't with despair. It was with a passion for saving people. To see that the diversity of people who are doing that, it's something about the human spirit, but in terms of predictions from behavioural science, the diversity and rapidity of that is a surprise.

Hetan Shah: [00:15:25] Turning to the second part of the name we've given our session. We talked about nudge and sludge. "Sludge" might be a term that fewer of the people who have tuned in today may be familiar with, in this context of behavioural psychology, but it's increasingly being used. Can you explain the term and give us some examples?

Professor Sunstein: [00:15:46] The basic idea is that human life is pervaded by frictions and paperwork burdens or administrative burdens, which are imposed sometimes in order to torment people and drive them crazy and that is a successful project sometimes. Sometimes they're imposed by self-interest, so if you want to subscribe to something online, sometimes that's really easy, but to unsubscribe, you have to navigate sludge, meaning you have to talk to somebody, you have to answer 13 emails, you have to say something about exactly why you don't want to subscribe anymore. By the time that question has been asked four times, you might say, “I'll keep subscribing. It's not that bad." That's sludge. Think of it as friction, it can be waiting time. Very recently, in my country, voting in Georgia required four to seven hours of waiting time. That's terrible sludge. It might be waiting time on the phone. It might be a requirement of an interview. It might be that you have to fill out some forms to get an occupational license. It may be you have to deal with the authorities in a way that is tough in order to avoid something bad happening to you, which might be a fine or a really not-good encounter with the criminal authorities. Think of it as a full universe of frictions that sometimes come from the telephone company, sometimes come from the utility company, sometimes come from public officials and there's too much of that.

Hetan Shah: [00:17:27] You've spoken positively about how some public officials around the world have been working hard to get rid of sludge. Can you say a bit about what you're seeing? It'd be particularly interesting to hear UK examples if you have any, but any from around the world of success in de-sludgfying?

Professor Sunstein: [00:17:42] What we've seen is an unnamed war on sludge. I'd want to study the UK situation carefully before calling out examples, but I'll give you the feel of examples and I'm sure there are UK equivalents, where to get certain kinds of benefits, people have to go in for an interview. That's eliminated. Hospitals and doctors have reporting and paperwork requirements in order to do things that are being waived. There are requirements that poor people have to meet, sometimes every month, sometimes every six months, in order to get benefits to which they are normally entitled. Those are being reduced.

There are benefits that governments are providing specifically in the context of COVID-19 that would ordinarily have sludge associated with it, maybe an application requirement, and some countries, including Canada prominently and in one context, the United States, we have north American examples, they don't make you apply at all. They just send the stuff to you. It's going to be in your bank account because the government knows you're entitled to it. It may seem like a small thing, but in a period in which people's health and economic situations are extremely fragile, the burdens imposed by sludge in terms of the costs are really high compared to, let's say, six months ago and are somewhat more subtle. I think a more crucial point all over the world is that all of us have limited mental bandwidth.

Our ability to do the 50 things that we're supposed to do today, we just don't have that. Maybe we'll get through 21 of them. It appeared in a pandemic for many people, the cognitive bandwidth is just even less. They're dealing with their kids. They're dealing with economic fear. They're dealing with their employer. Some have had an increase in bandwidth, but many have had less. Insofar as we're talking about populations that are elderly, that have a mental health issue, that are dealing with economic terribleness, poor people, the bandwidth problem is very, very severe. Then sludge can be literally a killer – it can be metaphorically a killer, but literally a killer – and governments, in a bottom-up way, this isn't so far as I'm aware of any country, a directive from a prime minister or the president though I hope that's coming in a hurry. If anyone is listening that can make that happen, that would be a gift to help make it happen. A sludge reduction directive from the prime minister or the president would be a terrific thing to see. It saves time, which is the most precious thing that people have. It gives them access to things, often, that can turn their lives around and this is a period in which we need more of that.

Hetan Shah: [00:20:54] There's an interesting point that rises from that. The desludging agenda sounds similar to the casual [unintelligible 00:21:04] although I think it's different to the deregulatory agenda. They're motivated probably by slightly different things, and might take us into slightly different places but one could imagine a government confusing the two. Do you just want to clarify the distinction between them or where they do overlap?

Professor Sunstein: [00:21:19] The deregulatory agenda, on reflection, is something we should be ambivalent about. Not negative, but ambivalent about. I'll give you some data that will capture the point. The Trump administration has imposed the lowest costs of any administration annually, on the standard accounting, of any administration for which we have numbers. It only goes back to about 2000, but that's 20 years. That's good. The Trump administration also has had the lowest benefits from regulation of any administration going back 20 years. If you subtract benefits minus costs, which is what you want, that's the net benefits, the record of the Trump administration has been the worst. And the deregulatory action deserves approval insofar as its costs reduction that can spur economic growth, it can take costs away from consumers and workers, so not negative, ambivalent about the deregulatory agenda. But if we're talking about taking away things that make air clean or that make workers safe, or that get resources in the hands of people who a) need them and b) deserve them, then that's not good deregulation. To make, particulate matter is a very bad air pollutant, to increase levels of air particulate matter in the air is to kill significant numbers of people, and that actually is interacting right now in a not ideal way with COVID-19. Deregulatory agenda, yes; when the costs of regulation exceed the benefits; more regulation, yes when the benefits of regulation exceed the costs. In the context of sludge, we could, for just simplicity, define it as excessive frictions, in which case we want to get rid of all excessive frictions. That would be a good agenda or we could define it more neutrally as frictions. Then the ones we'd want to get rid of are the ones that are excessive, which wouldn't be a deregulatory agenda. It would be a sludge audit agenda.

What we're seeing, mostly in informal ways but it's just starting, in companies as well as in governments, is more formal sludge audits where there's an accounting: how much sludge do we impose and is it in the end justified by the legitimate goals that sometimes animate sludge? Like making sure people who get benefits are eligible for them, or keeping good records to make sure that a program is working? Those are good reasons for sludge, but do we have 200% of the amount of sludge we need to accomplish those goals? If so, we can cut the sludge in half.

Hetan Shah: [00:24:15] Thank you. We've got about 300 people tuned in at the moment, so for those regulatory geeks amongst you, I hope you enjoyed that exposition.

Professor Sunstein: [00:24:24] We call them regulatory aficionados. Geeks would be a nudge not to be one – aficionados isn't the best. Regulatory heroes, that's what we call them.

Hetan Shah: [00:24:38] Cass, since you co-authored Nudge in 2008, one of the big changes has been the huge rise in social media. What's the impact of that in terms of the communications side of nudge and sludge?

Professor Sunstein: [00:24:52] It's a great question. Some of the social media platforms are very good at sludge reduction. Facebook has an initiative about helping people on their birthday to give money to charities and helping them encourage their friends. I've participated in this and it's almost sludge-free, it's so easy. The Facebook designers are obviously alert to the behavioural finding that if you make something easy, sometimes the results will be spectacularly large in size. Some social media providers are using nudges constructively to benefit both donors and donees.

The recent controversy over Twitter’s signalling that a tweet from President Trump was, I think the word “pretend” was potentially misleading, it's just a nudge. What Twitter did was to say, “to get the facts on this issue, go here”. That's just a nudge, no ban. So we're seeing very agile uses of social media architecture to nudge people in various directions. Some of these aren't clearly in the public interest, insofar as people are nudged to see things that are basically louder echoes of their own voice. That's probably not the best thing for democracy. It may be worse than that. If people are being nudged, and this is in social media platforms but it's a lot of website designers, are using behavioural principles in a fiendish way, it's called dark patterns, automatically to enroll people in programmes that they didn't clearly consent to, or pressing them repeatedly to buy a product that they have no interest in, knowing that if they keep being pressed on it they'll eventually give up. Or using social norms in a way that is in the economic self-interest of the seller, and not clearly in any way in the interest of the buyer. Those aren't things, so far as I'm aware, that the social media providers are doing themselves, though they are doing some things that are very much in their interest to attract clicks, but they are nudging people, the one I'm talking about, they are nudging people to part with their money. That's not so good. Especially not good is automatic enrolment in things that mean people will be, for years potentially, giving their credit card to someone for some service that they don't use at all. All of us need to think much harder about the social media platforms' responsibilities – this is a cliché, I'm going to try and make it a little more concrete – both to promote public health and to make democracies work better, for the community standards that are used by, say, Twitter and Facebook to be revisited, let's say tonight, in an hour, to think about nudges and sludge and how it's affecting public health, including through misleading or false statements, that would be a good idea.

Hetan Shah: [00:28:21] Thanks, that's fascinating. Turning back to something you raised earlier, which is the speed of change we saw in human behaviour as being something that you found surprising, thinking about other big global challenges we face, notably climate change, what could we learn from how behaviour has changed in the pandemic, and what that might mean for what we could do for other global challenges?

Professor Sunstein: [00:28:46] It's a great question. One way in, so in the White House where I work, the president had a phrase: “admire the problem”. He said “meanwhile, back at the problem", meaning, "think of solutions. I'm going to admire the problem for a moment, and then I'll try to be – meanwhile, back at the problem". For climate change, it's different from the pandemic in the sense that the sense of immediate threat for many people just isn't there. As we recently saw in Paris, if there's a tax, even a relatively modest tax, people might think, "what are you doing here?" Whereas if people are told something much more severe, "you have to stay home" people say, "okay" and that's because death and illness concentrate the mind and climate change is more abstraction.

That's the difference. One thing, that recent data shows is working in Germany that's behavioural, is that companies with, something like approval or spurring of Chancellor Merkel, they are automatically enrolling people in green energy, solar and wind and people can opt-out if they want, if they want to use coal, they can and in some places, coal is cheaper, but automatically, you're in the climate-friendly energy source. The recent data shows that's been a spectacularly effective policy that people all over this large and important country are in green energy, just because that's been the nudge.

I think they think "it's just not worth it to opt out", or "I'd feel like a creep if I opted out". In any case, the policy is working. Not every country can do that, but lots and lots of countries can. For climate change, the basic idea out of that little story from Germany is, if you can make the climate-friendly options easy, then people are more likely to choose them. If it's confusing or hard to figure out how to do something that makes your climate imprint smaller, then people are busy, they're not going to do it. We need to think hard about that. I think with special reference to the large entities like power plants or automobiles, manufacturers, who are the main contributors. For automobiles, consumers are acting in response to nudges like fuel economy information, in a way that's not enough, but that's something and that is helping spur a more climate-friendly fleet.

Hetan Shah: [00:31:41] That's a great example about the opting-in, opting-out of your energy supplier as it were. It emulates what is probably the best-known nudge of all, which is around pension schemes and getting people enrolled in and having to opt-out if they want to opt-out. That's a fascinating one I haven't come across. The final question from me, before we take the Q&A coming in from our virtual audiences, is, looking a little bit ahead, it's clear that we're going to face a whole series of new challenges as we're coming out of the pandemic. The economy looms large, in the UK we're likely to face a rise in unemployment, something we've not seen really for a very long time, but also changes in the patterns that we live and work. What's your advice to leaders when they're starting to think about how to apply behavioural science lens to the future challenges that are coming down the track, what should they be thinking about?

Professor Sunstein: [00:32:41] That's a large, as well as great, question. I have two ideas. One is, in deciding what policy to choose, to think in the most disciplined possible way about the consequences for human welfare. That's an abstraction. I want to make a plea for cost-benefit analysis. There aren't banners under which people march, saying "cost-benefit analysis now", but it's the best way we have of being really careful about what the human impacts of our policies are. The basic idea is, if you're deciding whether to go to one phase or another, to get the most current information about the incremental health problem, I think there's one, the incremental economic gain and to just be measuring it versus alternatives.

If this isn't bringing smiles to people's faces, I can say in the US government, when I worked there, when things went well rather than less well, it was because this had happened. It resulted in a lot of lives saved, or a lot of economic distress being averted. The staring at numbers, I'm not going to write a book called Staring at Numbers because no one would buy it, but that would be advice for public officials, try to stare at some numbers.

The second is, to keep a little framework in mind, and this is an adaptation of a framework used by the UK's amazing behavioural insight's team, and the framework is called FEAST. I'll tell you the EAST part of the framework, which is Easy: make it easy; make it Attractive, if there's signage, or if there's an opportunity, make it attractive. S is social, use social norms and to try to help nudge social norms in a certain direction, and T is timely, where the idea is, if you tell people at night "keep six feet apart when you go to the grocery store", good luck with that. If you could tell them when they go to the grocery store, keep six feet apart, that's the right time to do it. My little amendment is the word F for FEAST, and F stands for fun.

In New Zealand, the prime minister has been amazingly good at injecting an element of fun into the COVID-19 response. Who would've thought that was possible? She said "we're going to have a big lockdown, but the Easter bunny and the tooth fairy are being exempted. They're getting an exemption", and she did it with a kind of delight and wit, that helped people see they're in a situation that isn't great and it's scary, but we're all going to get over it, we're going to be laughing all the way. I just learned – at least we're going to be laughing as much as we can, people are dying, you're not going to laugh, but as much as we can, we're going to retain that human capacity.

I just learned yesterday that in Thailand the word “fun” – I'm sorry, in Taiwan. In Thailand, I hope, Taiwan I know – the word "fun" is part of the official framework for how to deal with the problem. The FEAST framework is designed to say to public officials, human beings need, especially in the context of a serious challenge, to enjoy their lives and people are doing that.

Hetan Shah: [00:36:07] Thank you. That's great insight. I'm going to turn to questions now that we've been getting in from the audience. The first one is from Justin Parkinson from the BBC, who's asked that "the government here has recently changed its main coronavirus message in England from 'stay at home' to 'stay alert' and that's been criticised as vague, so does it work as a nudge?"

Professor Sunstein: [00:36:33] I hope they're rethinking 'stay alert'. I have a great deal of admiration for many people in the UK government and the behavioural insights team are my friends and colleagues. 'Stay alert', the question is completely right, it's too vague. It's like “be smart”. To help people, you need to give them something like a GPS, “stay at home” is that. It's not indicated anymore, but we need some equivalent to that.

Hetan Shah: [00:37:08] Would you have a suggestion for something better?

Professor Sunstein: [00:37:12] I have a direction, not a specific, because I'm not sure what's meant by “stay alert.” The advice has to be actionable and “stay alert”… maybe there's a certain percentage of the population that is just going through life like a dog I once knew, meaning not alert – a lovely dog, but not an alert dog –but if you're doing that, probably being told to stay alert isn't going to have a big impact. Something like “stay safe” might have enough connotations, [phone rings] sorry about that.

Hetan Shah: [00:37:54] That's the UK governor on the line.

Professor Sunstein: [00:37:57] The governor saying, “why are you objecting to our slogan?" The basic idea is something that maps onto something that people already have in their mind that is concrete. 'Stay safe' is better than “stay alert', or 'be kind.' The New Zealand idea is better than 'stay alert', because it calls up some particular associations. I want to know, if I were working there, what do you want people to do, actually? Then you'll get some answers that can lead to a verbal formulation.

Hetan: [00:38:35] Very good. Thank you. We've got a few questions which are around this cluster of an idea that how do we hold the nudges to account if we're not aware about what they're doing? This is mentioned in a couple of different contexts. One person asking about in autocratic societies, another person is asking about how political parties are using nudges. Could you talk about that in that broad idea?

Professor Sunstein: [00:39:05] Great. For governments, we need a bill of rights for nudging. I actually have studied public opinion in 17 countries, including the UK, the United States, Russia, China, South Korea, Denmark, France. There's something surprisingly like a consensus, not quite that, but close to that, on what rights operate as a check on nudges. What emerges from this is if the government has illicit or self-serving motives, it's violating the morality that all governments should have to respect.

If the government is nudging people in a way that's inconsistent with their interests or values, people rebel against that, in the first countries. If the government is manipulating people, for example, with subliminal advertising, people don't like that very much. If the government is showing religious or racial favouritism through nudges, or even party favouritism through nudges, people don't like that either. What's maybe just short of thrilling in this data, is that in countries that have very different systems, we see a lot of agreement about what's not acceptable and to import that as a series of constraints on what governments can do. Transparency, also, and maybe an opportunity for public input before a nudge is chosen, is a really good idea. In the Obama administration, we did that. Not because we were anything other than pragmatic, which is, we might not know what we're doing, or people might hate it and both of those would be good to know and maybe people have ideas about something that's better, or about why this wouldn't be a good initiative. Plenty of times we learned from the public that we weren't going in the best direction and they helped us devise a better nudge. I'll give you a couple of examples: fuel economy labels and calorie labels. The public was extremely informative about exactly what it made sense to do.

Hetan Shah: [00:41:26] That's really interesting, because in a sense sometimes the criticism of nudge is that it's a very technocratic approach. It doesn't involve, or it's really just a bunch of officials deciding what's good for you. But the examples you're giving there are about something partnership with the public, but then designing a policy instrument that flows from public will in some way.

Professor Sunstein: [00:41:52] There's a lot there. If I were to identify the most technocratic approaches, I think I'd put high on the list mercury regulation, ozone regulation and regulation of chemicals that are potentially dangerous and carcinogenic, that these are extremely technocratic judgments, that entail use of expertise and it's not really accessible to people who aren't trained, or not easily accessible. I wouldn't put nudges at the top of the list of the technocratic. If you go to – there used to be a time when there were these things called airports, I don't know if you remember them, and these big metal things that would go up in the air and you'd all sit close to each other – but if you go to Heathrow, there are nudges everywhere, telling you where the bathroom is and how to get to your gate and where the rental cars are. I haven't been to Heathrow very recently so I don't remember if there are little steps that direct you step-by-step how to walk to various places, many airports have those things. They're really helpful, they're not too technocratic.

Hetan Shah: [00:43:06] We've got a couple of questions around how we get governments to take up a nudge approach. Somebody's asked, "It's clear that there are some governments that have already really adopted this, but how do we get the others who are not really thinking about this to take it on?" Another related question which was very nicely phrased, which is, "How do we nudge towards sludge audit?" How do we make those easy, as it were?

Professor Sunstein: [00:43:31] I'm not sure how many of you with whom you're speaking have connections with France, but France is a country with such a, words fail, such an inspiring history in so many ways and just nudging isn't coming there, and they could do so much on the cheap to help their people. We're just not seeing much, I'm not sure why. The best inspiration for doing nudging is a track record of success. When I would, in the US government in 2009, if I said I co-authored this book on nudging, let's do that, I think people would have said, “well, we have a place you should go and it isn't Washington DC. Not "how?" but "Back to the private sector. Don't talk to us about abstractions, talk to us about problems". If the question is, how can you help people save for retirement, the example you gave, the question is what can we do about climate change, or what can we do about H1N1? That was something I was engaged in that has some parallels to COVID-19. There the behavioural science is relevant to all of those. If it's problem-centred rather than theory-centred and if it's driven by things that actually can be shown to work, David Halpern in the UK has been extremely successful in drawing attention to solving problems rather than to academic theories.

We've seen a burst of activity in the last months, and also in the last years, with respect to use of nudges. I think in a way it's just going to happen, as people who know how to do this work end up in government that are old enough to be hired. Some of them are being trained in economics departments, or psychology departments or somewhere. In terms of getting it there in countries that aren't there, to get somebody excited. If it's the prime minister, fantastic. If it's someone who's a minister, that's also really good.

Hetan Shah: [00:45:50] Great. We've got a series of questions, in a sense connecting back to this question of democracy and nudges. One is about, "How do we make nudging more democratic?" You've talked a little bit about that but it'd be interesting to get your thoughts on that more. Then somebody else had also asked, "Does nudging work equally well when people know that they're being nudged?"

Professor Sunstein: [00:46:21] The second one is great and we have data which shows, so far, that every tested nudge is actually at least as effective when people know they're being nudged and sometimes more effective when people know they're being nudged. On reflection, that's not that surprising, because nudges aren't hidden. If you're given a calorie label, it's not a secret. You go to the Subway, or the McDonald's and there's calorie labels there, so you know you're being nudged. Or if you buy a car and there's a label on it that tells you something about the fuel economy, or a refrigerator that tells you about the energy efficiency, you don’t not know you're being nudged.

In cases in which there's very clear disclosure that people are being nudged, people either don't care or they think, “oh, there must be a really good reason for it, I'll do it". If people are told "the reason you've been enrolled in a pension plan is that this will ensure a larger accumulation of savings when you're older, unless you don't want that, if you opt-out, then you'll have more money now and less money then, and we're making it so it's really easy for you to decide save for retirement but you don't have to". That one, particularly to my knowledge, hasn't been tested in a rigorous way but it's very possible to think people think, “oh, I better do this. It seems like a really good idea". To give people transparency about their being nudged is important. It's respectful. It's part of the underlying ethics of nudging, which is preserving freedom of choice [inaudible] clarity that they're being nudged, then their freedom of choice isn't full. GPS nudges you and that doesn't make it less effective. If anything, it makes it more effective, so too with a sign on a beach that says, “warning, you probably shouldn't swim today because the waves are pretty terrible.’’

In terms of the democratic, that's a really great question. I mentioned calorie labels. I'll say a little bit about that. The requirement came from a law, which was enacted by our legislature. Then the question was how to implement it. There are ways to implement calorie labels that can have more or less aggressiveness and more or less range. Do you include movie theatres? The law isn't clear on that. The way the US government tried to figure it out, is it said, “Here’s our proposal; here are five alternatives,” – I'm making that number five up by the way, I think it was more than that – "five alternative things, in terms of presentation, or in terms of scope and we want to hear your comments." It was a lengthy process of obtaining comments and many of the comments suggested the proposal wasn't a good one and some of them were completely convincing. We ended up doing something different. What's true for democracy and nudging is, I hope, going to be even more true for democracy and mandates, because mandates are more threatening, because they don't allow people to opt-out.

There's an openness to a nudge which creates a safeguard against government mistake, or corruption, or self-interest, but since nudges are impactful, we need these safeguards of democratic engagement.

Hetan Shah: [00:50:09] I think it's a powerful point you make, which is sometimes forgotten, which is that the nudging architecture is all about allowing people choice. It nudges, but I think that the formal term you used originally was liberal paternalism and it's got a liberal element, as well as a paternal element.

Professor Sunstein: [00:50:28] The American term is libertarian paternalism because liberal means left of centre, so think of it as freedom preserving, and that's humility, that government officials need to be alert to the fact that they may get it wrong.

Hetan Shah: [00:50:47] A fascinating question that's come in and a difficult one is "What does nudging have to say about the way that certain public debates are getting more and more polarised at the moment?" We see this in particular in the way that social media spheres operate, but we can see that, I think, particularly in the US at the moment. Does nudge give any insights on that?

Professor Sunstein: [00:51:12] Well, behavioural science does. I'll tell you a little experiment I was involved in a few years ago. We got a bunch of people who were left of centre to talk about an international treaty on climate change and to talk with each other about whether there should be one. We asked them to record their views privately and anonymously before they talked to each other and privately and anonymously after they talked to each other. We also do that for people who are right of centre, that they only talk with like-minded others, but in both groups, there was diversity going in, on an international treaty to control climate change.

What happened was the people who were left of centre ended up more extreme, more confident and more unified. They went whoosh to the left on climate. On the right, there was some diversity before they started to talk. In their anonymous post-deliberation statements of opinion, they went whoosh to the right. They all thought, international agreement, crazy idea, they all thought that, it was unanimous and it wasn't going in. Social media were observing the Colorado experiment. They are often, not always by any means but often, echo chambers, by which people are sorting themselves into like-minded groups and they end up more extreme. Now Facebook, on its newsfeed, can aggravate that problem or reduce that problem. It can encourage people just to hear their own voice, or it can nudge people to hear a diversity of voices. Regrettably, Facebook hasn't gone in the, let's say, ideal direction from the standpoint of democracy, though they've been worrying over the problem, they've been "admiring the problem", but meanwhile, "back at the problem". There can be nudges that would help by having some clarity in your newsfeed about the diverse voices that are out there, or something that would give you a sense that what you're reading right now is actually false. Facebook has taken some steps in that direction, so has Twitter, but let's say more can be done.

Hetan Shah: [00:53:34] Great. We're getting close to the end of our time, but a final set of questions cluster around this idea of, how do we make sure that some of the successes we've had around nudging during this early stage of the pandemic continue through, as we now start emerging out of lockdown. A couple of people make the point that actually there are more people who've now had the infection, but we seem to be relaxing almost at the same time. I'm sure there's different pictures in different countries, but what's your prognosis of how we should be using nudge in this, what might be quite a dangerous period?

Professor Sunstein: [00:54:12] Yes, there's suggestive data, it's not airtight yet, but it's suggestive data, that if you tell people, "if you take precautions you will reduce risk to yourself", that's helpful, but if you tell them, "if you take precautions, you're less likely to endanger your friends and neighbours", that has a bigger impact. New Zealand has been masterful at giving people a sense that, sensible precautions, let's call them, are a way to ensure that you yourself are not responsible for the illness of another member of the human species. There are ways of doing this that are unpleasantly directive and sanctimonious. To do it instead with a sense of a common endeavour and good cheer – the record of success in multiple countries over the last months suggests that's something to build on.

Hetan Shah: [00:55:17] Thank you very much. Well, look, Cass, we will end there. I'd really like to thank you for all of the time that you've given us today. It's been fascinating insights on how we can respond to the pandemic in general, but we've strayed well beyond that as well, to think about the role in democracy, social media and preventing more authoritarian uses of nudge as well. It's been fantastic. I'd like to thank our virtual audience as well, most of whom have stayed with us through the course of an hour and of course, we've been recording this so many more people will be able to hear it as we go into the future.

For those of you who are still with us, do keep an eye on the British Academy website and our social media. You'll be able to see more from our Shape the Future work, which is responding to COVID and our future events. For now, I'd like to thank all of you, and thank you again, Cass.

Professor Sunstein: [00:56:08] Thank you.

This talk originally took place on 10 June 2020, part of the Shape the Future programme, which brings together researchers to examine a range of issues around COVID-19.