What behavioural science tells us about the COVID-19 response

In Southasia, the COVID-19 response will be encumbered by inequalities in access to resources. Not many will have access to running water or the luxury of self-isolation. Resource inequality may determine not only who is at highest risk of infection but also who can adopt recommendations to slow the spread of the disease. Some of the fallout for low-income communities can be mitigated by designing policies tailored to Southasian needs..Himal Southasian speaks to Saugato Datta, a managing director at ideas42, a leading non-profit organisation in the field of behavioural science. Datta designs, tests, and scales the applications of behavioral science to benefit low-income individuals in developing countries. He was previously a correspondent for the Economist and has worked at the World Bank..***This is an unedited transcription from the podcast. Please listen to the corresponding audio before quoting from it.Amita Arudpragasam: This is Amita Arudpragasam from Himal Southasian. Today we will be speaking to Saugato Datta, a behavioural scientist and a managing director at Ideas42. He designs, tests and scales the applications of behavioural science to benefit low-income individuals in developing countries. Welcome Saugato.Saugato Datta: Thank you Amita. Nice to be here..AA: Behavioural economics is a relatively new field. Can you tell us very briefly what it is and perhaps a little bit about what it offers in terms of the COVID-19 response?.SD: I think the best way to think about it is that, or sometimes the way I like to describe it, is that behavioural economics comes about when economists realized that a lot of the ways they describe human behaviour and the way they think about people making decisions, or deciding, or acting on certain things, is not really a complete description. And specifically, economists have long been used to thinking about the fact that we take decisions or we act in certain ways mainly based on a sort of cost-benefit analysis. So that's what economics is fundamentally about..Now, at some point, all this evidence started coming in from psychologists which said 'hang on that's not the only thing that people actually take into consideration.' They are influenced by a lot of other things which don't have anything to do with how valuable something is, or how much it costs or, you know, what we think of as economic factors. And, in fact, those things and those could be simple things like what do you see other people around you doing, or what's the last thing that someone told you about, those have an outsized influence on how people actually behave. And so if we don't take that into consideration – we end up with a very sort of stunted view of human decision-making which actually gives us wrong predictions..AA: COVID-19 requires large-scale behavioural change – whether that is social distancing staying indoors or frequent hand-washing. What wisdom does behavioural economics offer for improving design, response design?.SD: Fundamentally, like I said, behavioural economics and sometimes we call it behavioural science, is about understanding how people act in the ways that they do. And as you mentioned, everything to do with the pandemic response is really fundamentally about behaviour change. So either we are asking people to do things like wash their hands more frequently or in a different way than they would normally have done, we're asking people to really disrupt their patterns of daily living. So things that you are very used to doing, whatever your circumstances may be, will now have to change. There are some broad lessons that behavioural economics gives us about how to make these kinds of behaviour change happen more easily..So even this idea that I just mentioned, which is the idea that we respond to cues in our public environment about what other people are doing – that's a fairly consistent finding in the literature. But how it applies may be quite different..So there's sort of a large literature about things like habit formations. We're really trying to disrupt habits here. For instance you would have seen some of those videos that try to link hand washing lengths to popular songs. That's actually something that behavioural economics says is a good idea because what it does is it takes something abstract, which is very hard for people to comprehend and really follow, which is wash your hands for 20 or 30 seconds, and turns it into something which is a very simple rule of thumb that people can follow..So these are some kinds of ideas that come out of this field which, if you again just take a step back, if you didn't think in this way, you might think 'okay what you really need to do is tell people hand-washing is really important and hand washing is going to save your life and hand-washing is something that you should do in this particular way.' That's that's fine, that's important because people need to understand that it is something that they need to do. But that doesn't usually translate into people actually doing that. And to make that happen behavioural economics says that you might be able to leverage certain other aspects of peoples' psychology – like, you know, I mentioned the rules of thumb, so turning it into a song – or you might be able to more effectively show them that other people that are like them, for instance, are doing what they're doing..AA: And what are the dangers or limitations of applying behavioural insights outside the geographical areas where they are researched and tested?.SD: We can't take things that have been developed in a completely different context and just sort of transport them to a new one. So, you know, behavioural economics as a field and even the psychology studies that underlie it they were mostly done not only in the developed world but on a very specific sub-population in the developed world..You know, most of these were experiments done on college students on campuses in the US and Europe, you know, that's not a representative population. Some of the biases that they may be subject to may not either be the same or may not work in the same ways when you take them to a new setting. And so a lot of work now is really about maybe in some ways retesting those biases, and seeing – do they apply in the same way?.I mean the good news is that there is a fundamental basis which does seem to apply across different settings. But it's important to remember that how it applies is going to vary quite drastically. So even this idea that I just mentioned, which is the idea that we respond to cues in our public environment about what other people are doing – that's a fairly consistent finding in the literature. But how it applies may be quite different. You know there are all these studies, for instance, where you find that telling people that other people have done something, for instance, paid their taxes, leads more people to pay their taxes. Now these were mostly done in places where tax rates, tax payment rates, were very high. And so the social norm already was around paying taxes. Now when you take this to a setting where perhaps people don't trust the government, or they're a bit worried that the tax money might be misappropriated, you can't just take that insight and apply it wholesale. And so I think that applies to COVID-19 responses as well..Because, the simplest way to think about it is, if you have very few financial or monetary resources, you're having to manage many more complex things with a very limited amount at your disposal and that actually makes every decision and everything that you have to do much more complicated..I talked about hand-washing. Clearly hand-washing is important. But you can't have the same kind of hand-washing campaign for people who don't have running water or who don't have access to soap or face other structural barriers. So you've got to think about that context specifically and see what can happen..And actually maybe one example, I think, drives it home and it's not to do with the current crisis. But it's to do with a series of studies about how to get people to chlorinate their drinking water. And the original studies were done, I think, in East Africa. I could be getting this wrong. But they were very successful because they install this little chlorine dispenser at the well. And so when you filled your bucket of water, you pressed it once you got some chlorine in the bucket and you took it home. Now remember, the basic idea is that this is something very beneficial for you, like, it cleans your water, but people still forget to do it, right?.Even if they have chlorine tablets at home they may not use them and this kind of removed some of those barriers and made it easy. But then they try to test the same intervention in the slums of Dhaka, in Bangladesh, and it was a total failure, and people did not use it. And they were trying to understand why, and then they realized that in Dhaka instead of everyone coming with essentially a bucket or some large thing to fill water in, the hand pumps were very close to the houses. So sometimes you would just come with a glass of water to fill water to drink and obviously the same amount of chlorine was going to make that water completely undrinkable, right? Because it would, it was no longer adapted to the context in which this is being done..So even the basic insight is the same that you've got to make it easy for people. You've got to remove hassles, but how you do it is going to be quite different in different contexts, and I think that applies just as well to COVID-19. So it applies for hand washing, it applies to social distancing or physical distancing, as I think we you know in some sense would prefer to call it, because you cannot have the same rules of thumb you cannot have the same modalities for doing this in different kinds of settings and you've got to think very carefully about that..AA: We talked about the difficulties of applying insights from one country to another, but is there anything that is similar about low-income communities and the kinds of resource constraints that they face, that might lend itself to designing policy in a particular way? .SD: I think there is, and I think that the sort of key insight from the literature is that, is this idea of scarcity that you mentioned, which is that, we think, of course, of economic scarcity. But economic scarcity also goes with, you know, sort of essentially, cognitive scarcity. Because, the simplest way to think about it is, if you have very few financial or monetary resources, you're having to manage many more complex things with a very limited amount at your disposal and that actually makes every decision and everything that you have to do much more complicated..And so one thing that comes out of that, is that, if we do indeed want to design policies that are helpful for low income people, we have to make them as simple as possible. We have to avoid, you know, creating all of these sort of hoops that people have to jump through in order to, for instance, access benefits. We also have to do it in a way which sort of tries to minimize additional uncertainty that people might feel at a time like this. So if you're trying to design policies for low-income people, you know, it would veer towards the side of making them universal, making them easy, so that people don't have to, you know, in addition to having to figure out what to do at this time, whether they may be out of work, and their income may have stopped, you don't also want them to have to essentially kind of, you know, cobble together a whole bunch of documentation, for instance, to prove they are eligible for something..So I think in a developing country context governments would do well to think creatively about how they can use the resources they have at their disposal. So how should we, for instance, combine cash assistance and in-kind assistance? What are the easy ways of actually getting things to people in a way that will be useful for them during this crisis?.Because if you do that, what's going to happen and this is actually across the world you see this, you end up leaving out the most vulnerable people, because they are in fact the people who do not have these things, they are in fact the people who have a harder time putting all of this together, right, so they're simply not going to go. And I think that applies also to things like access to health care. You want to remove some of these additional hassles, to the extent possible, because by you know by having them in place, what you're going to end up with is an even more unequal response to what is already something that is playing out in terms of you know a long axis of inequality that already exists..So we know that when you ask people to stay home and you know not go to work that's a very different proposition for somebody with like a stable job, which is going to keep paying him or her, from somebody who is on, let's say, daily wages right. And so you've got to take that into account. And so, for instance, you've got to realize and you've got to actively plan to supplement the income or make up some of the lost income of some of these people..So we've been really thinking about the fact that, some form of close to universal cash transfers, even in small amounts would be, I mean it's too late to institute that obviously once this crisis has hit, but for these kinds of situations, that would be incredibly helpful. Because that's something which sort of everyone gets, and it's something which you don't have to, sort of maybe prove additional eligibility for, it's something you can ramp up quickly, you can also ramp down when you don't need it..But having those systems in place would be incredibly useful because otherwise you are going to further, you know, sort of disadvantage those who are already disadvantaged in this current situation. And this I must say, by the way, applies also to policies in the developed world. So in the developed world also they have to think about the fact that there are a lot of poor people who they actually in fact do typically are asked to do all kinds of things to prove that they're eligible. So if you look at the US, right now they have this big problem with unemployment insurance. It's available, but to get it you have to do all sorts of things people are not being able to do, and which require them to use resources they may not have. Even if it's just having like a stable internet connection to stay on and submit an application which takes hours, right? So if we can remove that step you actually improve access, in a way that you can't do otherwise..AA: What does behavioural science tell us about the most effective ways for governments to provide financial support to low-income communities – whether that's cash or in-kind support, broad or targeted support? Tell us a little bit about what works in developing countries from what you've seen..SD: So again I think we have a sort of, a good body of evidence that, for instance, cash transfers are in general a useful tool. But we have to think about how that works in this particular sort of situation. Now we are in a particular situation where it's not as if money is the only thing that people need, because things may not be available in a certain way. So I think in a developing country context governments would do well to think creatively about how they can use the resources they have at their disposal. So how should we, for instance, combine cash assistance and in-kind assistance? What are the easy ways of actually getting things to people in a way that will be useful for them during this crisis? And I don't think, actually, enough has been done on that..What the literature does tell us, what the sort of field does tell us, is whatever you do you should be doing in a way that's quick and that's easy for people to access, and, if possible, especially when giving cash or other kinds of assistance, you should try and sort of provide some very simple assistance for people to figure out how they should use this money. Because you know, for instance, they may, the government may give you a cash transfer. Now you may not have been directly impacted by this crisis at the time that it comes. So the government should have combined that with something which sort of gives you a rough idea of how you should allocate this money, so that maybe in a month if something does happen, but you're not going to get another tranche of money, you have some of that money saved up, right?.You know more broadly, I would say that there seems to be this idea that sudden announcements or you know sort of shock therapy as it were is a good way to enact policy and I think what behavioural science tells us, is that that's very rarely the case..So a lot of the work we do with cash transfer programs is in a more kind of normal situation asking these questions, to say, like, if the government's giving you money every two months, how do we maximize impact of that money, by giving you tools for yourself to figure out what is the best way for you to allocate it? How much should you put aside, how much should you spend on immediate needs, and so on. Now all of these things have become much more complicated in this situation because some of your options are not available to you, you may not be in a place where you know how to access things. But I think the broader lesson, as I mentioned earlier is, that we should be thinking about in-kind assistance as well..And I was quite struck by something that, you know, for instance, I hadn't even thought about, which is that when you have people who are, you know, workers, who are working in a different part of the country, for instance, away from their families, you know, for some of these people they don't even have a place where they can cook regularly, right? So even in-kind assistance has to be done in a way, if you want them to stay there, then they have to provide them actually with actual meals that they can eat, right?.It's not even it's not enough to give them money because there's nothing to buy right, they can't go out, it's not enough to just give them like raw food because what are they going to do with that? And so these specificities of people's situations have to be taken into account, if we want these policies to actually have the effect that they would have. Otherwise as you've seen, for instance, with India's lockdown, is that, if we don't take the behavioural response into account we get the natural behavioural response, which could in fact exacerbate the problem, or could certainly render whatever it is that you're trying to do, you know, much less effective..AA: Let's talk about India. So you mentioned the need for quick policy response and we saw in India, on 25th of March, Modi announced a 21 day nationwide lockdown. With just hours of notice, he left tens of thousands of urban migrant workers no choice but to walk hundreds of kilometres back home. If he took behavioural science and the insights from this field into account in his policy response – would he have done things as he did?.SD: You know there are just fundamental ways in which the way the lockdown policy was handled in India, which were, I would say, sort of very unmindful of what we know from behavioural science. So for instance when you give people just four days notice and you don't make it clear..AA: four hours notice..SD: Four hours notice, and you don't make it clear what will and will not stay open, you shouldn't be surprised by the panic buying and the people queuing up to go to grocery stores because that is a natural response. And so one of the fundamental things about any policy response like this is, like you have to be very clear. And you have to think through what the responses are going to be, and do it in a graduated way, so that people are able to adapt, right?.I don't think a national government can mandate, you know, all of these finer details. The national government can set the agenda, but it has to consult much more with local and state governments about what's the situation in your state, or your city, and how do we manage that..So that I think was not done clearly in this case. I mean the migrant worker situation is again something that – it's completely foreseeable, right? If you're going to leave people without income, and these are people who do not have stable, you know, kind of flows of money, they're not in their permanent places of residence, like you know, we know what's going to happen. And so, either you have to supplement any measure like a lockdown with, very very active ways to ensure that these people have whatever they need to be able to stay in place, or, you have to do it in a graduated way, and you have to allow them the possibility of making their way back to wherever they are without excessive hardship..Now in this particular case, my rough understanding is that, you did not in fact want every, you know, the 100 million or so workers who work away from their hometowns, to travel back, because there's a possibility that they were already infected, and then, you know, you would spread the infection further in the hinterland. So if you want to avoid that, then you have to realize that, they don't have any other option, and so you've got to give them what they need, to enable them to stay in place. And I do believe that there at the sub-national level there has been some attempt to do this. So I think some of the states in India, for instance, did step in and try to provide migrants from other states who worked in their state with, you know, the essentials that they needed and so on..And I think you've got to, you know, it's a very subtle sort of situation because, for instance, it's not just about food, and so on. People are naturally going to feel scared, isolated, you know, cut off. So things like, are they able to call their families? You know, many of them have mobile phones, but may not have any way to top up, you know. Is there something that you can do around that? You know, these are very sort of small measures that, if done as part of a kind of a comprehensively thought out package, I think could have had a much much much better chance..So it goes back to that behavioural insight about we respond to what we think people like us are doing, and you know 'like us' can be defined in a lot of different ways, and one of those is, you know, people who share a religious religion or a faith, or follow certain practices..You know more broadly, I would say that there seems to be this idea that sudden announcements or you know sort of shock therapy as it were is a good way to enact policy and I think what behavioural science tells us, is that that's very rarely the case. Because sudden announcements and things that are done without adequate warning, or without adequate and it seems that adequate thought to what all the kind of ripple effects might be actually cause a whole lot of behavioural responses that that actually undermine whatever your state policy objectives might be..If you know the context well, you actually can come up with some ways that, you know, are more behaviourally informed. Like you can think about what are the networks of these migrant workers, what are the networks of for instance street vendors, who may not be able to operate in their normal way, but can we leverage them in some way to supply things that, you know these guys need, to them, and stuff like that, which I think wasn't done, and I think has to be actually done… you've actually got to let that be done at the local level. I don't think a national government can mandate, you know, all of these finer details. The national government can set the agenda, but it has to consult much more with local and state governments about what's the situation in your state, or your city, and how do we manage that. And I think that wasn't done sufficiently, or certainly not done in a coordinated way, which I think is unfortunate..AA: We recently received a letter from someone writing from Pakistan who wrote about the psychological difficulties, anxiety and stress, of not being able to fulfill a religious duty in a time of crisis. Talk a little bit about how crisis and religiosity interact and how religious institutions and other community leaders can actually leverage their trust networks among communities to promote public health initiatives. .SD: You have to move early. You have to talk to religious leaders to actually explain to them why in this particular situation, gatherings are the thing that's dangerous. So it's not that it's a religious gathering which is dangerous, any gathering is dangerous, a religious gathering is an example of that. You know get them on board and get them to be the people who are telling people what they could be doing. And I do think we've seen some examples where, for instance, I think that, in fact from Pakistan, perhaps, where you know there was a coordinated attempt to tell people that you know it's okay to do your namaz at home on Fridays. Like that's actually encouraged actively right now, and they changed the announcement that was coming during the azan to say, you know, instead of coming to the mosque, you know, pray at home..So if I do some public messaging, where I only show men saying something, it's not as if, by having men do it, we're saying women shouldn't do this, but subconsciously women will take that message away, that this is a message for men, we don't have to do it, right? So that's why diversity and representation in the makeup of who is spreading these messages is very important, right?.Because you do need, people do need that kind of reassurance and leadership, and I think we've actually seen in the past, religious leaders can play a very important positive role in promoting public health initiatives. So during the polio campaign, in in India, which eventually, you know, succeeded more or less in in wiping out polio, they really leveraged religious leaders leading by example, for instance, showing their own children being administered polio drops, or their grandchildren, to dispel any, you know, lingering fears that people might have about the efficacy of the vaccine, and so on..And so similar things could be done in this situation. So it goes back to that behavioural insight about we respond to what we think people like us are doing, and you know 'like us' can be defined in a lot of different ways, and one of those is, you know, people who share a religious religion or a faith, or follow certain practices. So we need that leadership to just, sort of, essentially tell people that this is okay. And that they are, you know, they're that it's fine to do this. They should not feel bad about not doing things that they would normally do, and then perhaps think about ways in which other aspects of their religious lives which are probably quite important to them at this time, can be maintained, while respecting whatever need there is for people to change their behaviour in other ways..Because, again, you know we suffer from things like confirmation bias, so when we see something that looks like it's happening in a particular way, if it conforms to our existing prejudices about how the world works then we sort of amplify that and we react accordingly, which ends up actually not being very useful in dealing with the virus, right?.I think in a funny way people put, you know, religious activities, in a very different bucket than other activities, but really these are all like human social interactions of one form or the other. And the same principles apply, but we have got to think about their, sort of, significance to people may be different. And I don't think I have seen enough of like people the governments reaching out to religious leadership to say, you know, instead of coming from the government, if the message comes from them, it might actually be more powerful, you know, for a lot of people and this needs to be more of that..AA: And I imagine it's not just religious leaders but also, let's say, members of the opposition that need to be called upon to leverage their trust networks, individuals who, especially in divided countries, countries where there are a lot of political rifts, for example..SD: Yeah absolutely. And, you know, one of the one of the things that that pops up over and over again, is in any kind of messaging or communications, it's the the sort of perceived identity of the messenger is very important. So, and that can be played out around various axes. So if I do some public messaging, where I only show men saying something, it's not as if, by having men do it, we're saying women shouldn't do this, but subconsciously women will take that message away, that this is a message for men, we don't have to do it, right? So that's why diversity and representation in the makeup of who is spreading these messages is very important, right? You can't…if you have only celebrities telling you to wash your hands then it's actually quite natural for normal people to think so maybe this is something for celebrities, right, it's not really for us..AA: That's interesting..SD: So, you've got to do a mixture of things. At the same time, people are very kind of responsive to what celebrities tell them, so you shouldn't not have that either, you see. So it's sort of thinking creatively about how to use these levers. How do you, how do you leverage the fact that people may be responsive and certain to certain kinds of authority, so as you said, religious authority may be one, political authority is another. You know local leadership of various kinds. We've got to find ways that messaging does not seem like it's coming from people who are not like us. That's very important. That's sort of a common feature that we see, and a mistake that we see made over and over again, where you know, if I give you a message from somebody and it doesn't seem like you, then you're not going to follow it, because you don't think it really applies to you. And it's not a consciously thought-out thing, that someone is sitting there and saying this doesn't apply to me, it's just that that's how they process it and then, you know, they don't attend to it..AA: The individuals who were tested positive for COVID-19 and those suspected to carry the virus have been stigmatized — that includes doctors. What advice do you have for governments who want to encourage individuals to report voluntarily for medical testing?.SD: Stigma is actually actively detrimental to dealing with these problems because when somebody is affected by something you actually need them to go, and you need them to be treated, and so on. And that is something again that, you know, unless you actively try and make that happen, you are going to fall into a situation where people are going to hide it, because they are they're going to be afraid of the consequences, and I think this is just something where you need to have a very active campaign of public information and public education to sort of make people understand how, you know, how the spreads, what is and isn't a danger, right?.But I think knowing that these tendencies exist, and knowing that people will tend to over-attribute or misattribute, it's just that what it speaks to then is why are they so susceptible to this particular version of this? And that is because there must be an innate prejudice that already exists and I think, you know, a wise government would seek to actively disabuse people of these ideas, by showing them how this is not the way in which it's actually operating..Because, again, you know we suffer from things like confirmation bias, so when we see something that looks like it's happening in a particular way, if it conforms to our existing prejudices about how the world works then we sort of amplify that and we react accordingly, which ends up actually not being very useful in dealing with the virus, right? So in this particular case, people have these kinds of ideas, about what sort of contact or what kind of person might, you know, be more susceptible. But often, that's not true, right, and so by falling prey to that, you're actually, you're harming yourself, or you're reducing the chance that you can actively protect yourself and you may be making things worse for other people as well. So I think taking these kind of decision-making biases into account in framing policy responses and framing communications is very important..AA: We've also seen a rise of Islamophobia across the region. Mitra Sharafi, a legal historian who recently wrote with Himal, noted that that scapegoating others during epidemics is not a new phenomenon, but she noted that this has happened in the past. Why might that be from a behavioural perspective? What is it about humans that causes us to blame other people and communities we distrust, when we ourselves feel fear?.SD: So starting with the Black Death from Europe you know there were entire countries where they blame the Jews for this and then they expelled communities that because you know they fundamentally didn't understand how this thing spread right. I think it's an interaction between existing beliefs and prejudices and mental models that people have, and this inability or sort of lack of facility with thinking about the way probability and statistics work right..So you see this not just in sort of epidemiological context, you see it around things like crime. So if you have a crime committed by a member of a visible minority, then people overattribute, right, the prevalence of that sort of behavior, among that visible minority, not realizing that in the vast majority of cases, same thing is being done by people who you don't identify as belonging to a particular group simply because they look to your mind like the average, right?.I do think this again speaks the fact that one other thing that it would be good for governments to have done. I think decentralization, more devolved decision-making, more local you know political power, and ability to make decisions all of these things stands you in good stead in a crisis like this, because it's so graduated, it has to be so graduated, and so differentiated, that you cannot expect, you know, a top-down approach to work here..And so I think that's part of what's happening here, with this over-attribution of cases, you know, so people when it comes to religious gatherings… I mean in the Indian case you saw this sort of enormous scare mongering around this gathering that took place in Delhi, in Nizamuddin, for the Tablighi Jamaat, and I think the fundamental point there was if you're only going to be testing and tracing people who went to that particular gathering, that particular gathering is going to look like this huge vector. You're not testing and tracing people who went to, you know, 500 other sort of gatherings of that nature..But I think knowing that these tendencies exist, and knowing that people will tend to over-attribute or misattribute, it's just that what it speaks to then is why are they so susceptible to this particular version of this? And that is because there must be an innate prejudice that already exists and I think, you know, a wise government would seek to actively disabuse people of these ideas, by showing them how this is not the way in which it's actually operating. But, you know, wisdom isn't necessarily what we always associate with our governments. So, you know, these are existing fault lines in society that combined with the way people think about all of these things, is giving rise to an upsurge in this sort of like demonization of particular minorities or particular groups, and it's very unfortunate..And it will in fact reduce the efficacy of any other messaging. So I just talked earlier about how messaging needs to feel like it comes from someone like you. So if you are then, if you're a religious minority which is being kind of portrayed in this negative way, why would you be receptive to government messaging around safe practices, right? You are you're going to think that it doesn't apply to you, and so it's going to just make matters worse..AA: At Himal we've also started to collect notes and journal entries from individuals across the subcontinent. You might expect these individuals to have the same response, because they may be from the same place. But how can governments, you know, effectively communicate in such a crisis when individuals from the same place, you know, same situation, same circumstances, are having such diverse emotional responses to this pandemic? How – what kind of messaging is appropriate in these situations?.SD: How you respond to a particular situation is partly that situation and is partly what else has been going on, right, like so why are you predisposed to respond in a particular way – some of which is your psychological makeup but some of it is other things in the past that have that have influenced your response to crises. And so there can't be kind of a one-size-fits-all response, if you really want to address people. And I think one thing that governments maybe don't have enough of an emphasis on, and that they could have more of an emphasis on, is just finding avenues or having systems that allow people to actually express some of their fears, apprehensions, whatever the case may be. And have someone to talk to. So, I mean, it's this sort of the basic idea of like, you know, helplines or counselling and things like that, which don't necessarily, they don't necessarily resolve the problem, but they may at the very least make people realize, you know what else is going on, and give them some context in which to place what's happening..So it doesn't surprise me that you would get like very different responses, you see different responses you know even within the same family. By asking people to respond in a sort of a particular way, that is going to have different effects on different kinds of people. For instance, you're asking people to stay at home, but what about what about you know domestic violence? Or intimate partner violence? What about people who it's dangerous for them to actually stay at home, with you know somebody who is potentially abusive?.So I think the default for the police in a lot of the developing world is, you know, you see a street vendor, you sort of beat them up. And you're seeing a lot of that and I think at some level, the government needs to realize that that's their default, so unless we give them a new default, we tell them what to do instead explicitly, they will just do what they are used to doing..At the end of the day this all ends up sounding like the government needs to have invested much more in, you know, health and counselling and resources and things that people can turn to, and if they haven't done that, we're sort of stuck with it. But it sort of highlights the fact that these things are important, because any measures that you take are going to have different effects on different kinds of people, and just you know it's their psychological make-up, but it's just then living circumstances, right?.In the US, for example, all the universities said, you know, students have to go home, right? And then, it's only then that they started thinking, well some students don't have a home to go to, or you know, there are international students who can't go back to their home countries because flights have stopped, right? So we've got to, then the university has to come up with a specific response for that set of people because they can't just you know leave them on their own, what are they going to do, right? And so I think it's the same thing on a large level..I do think this again speaks the fact that one other thing that it would be good for governments to have done. I think decentralization, more devolved decision-making, more local you know political power, and ability to make decisions all of these things stands you in good stead in a crisis like this, because it's so graduated, it has to be so graduated, and so differentiated, that you cannot expect, you know, a top-down approach to work here. You've got to outline the principles, perhaps from the top, but then, each local authority and within that different, you know, divisions, have to figure out how this applies to your case and and what to do..And I think you know again, you know, go back to the migrant workers. People blame them, right? They say why can't they just stay where they are? And I think that's not the right question. The right question is how have we not made it possible for them to actually stay where they are?.And actually one thing I wanted to mention is something that sort of is so striking in the Southasian context and actually the developing country context, is the fact that if you don't clearly tell arms of the state what they're going to have to do, they will sort of essentially revert to their default. So I think the default for the police in a lot of the developing world is, you know, you see a street vendor, you sort of beat them up. And you're seeing a lot of that and I think at some level, the government needs to realize that that's their default, so unless we give them a new default, we tell them what to do instead explicitly, they will just do what they are used to doing. Without thinking about how it affects the situation, in this particular setting, and so this sort of thinking, I think, has been quite missing from the response and it's unfortunate..AA: After policies are implemented, or designed and implemented, communicated to individuals, there's panic buying, there's hoarding, and then, there is this phenomenon of policymakers blaming individuals, saying that they don't have enough self-control. And there is this movement between the responsibility of a policymaker and the responsibility of an individual, and, you know, as you mentioned, the responsibility of those intermediaries in government, as well. Because COVID really requires a whole of society approach, nobody really knows who to blame, and nobody really knows whose responsibility managing the crisis is. How do you think about this, this issue?.SD: You know I think it's a it's a particularly – so okay, if you ask me this about any normal situation I would tell you that, one of the things that we learn from behavioural science is that, we tend to attribute people's actions to sort of themselves and their psychological makeup, and something innate to people. But in fact what the research shows us, over and over again, is that the majority of people's behaviours is actually context driven. So, in fact it's almost always the wrong response to say, this person did this thing because they are like that. It's in fact much more likely that whoever did whatever they did, they did it, because the circumstances around, the contexts around them, was such that it created the need to do that. I think what makes it particularly and I would stand by that, in this situation as well, and so you know, I think it's sort of pointless to say, it's pointless to say why didn't this person follow the rules?.I think you have to ask, how do we design the rules in a way that makes it possible for people to follow them, right? I understand the government's dilemma a little bit more in this case, because this is a situation where we simply kind of don't, we don't know the full consequences of our own actions, no? Because any individual saying, you know, I'm fine I can go out, I can do X or Y, they don't actually know whether, for instance, they might be positive, if they might be spreading it to a lot of other people, so it's a very complicated situation..Having said that, I think for the most part blaming individuals for not following things or not doing certain things, sort of misses the point, because I think no one is really, or I think a vanishingly small fraction of people are doing this because they sort of you know they just don't want to listen to anyone or they want to go out and flout the rules. They're doing it out of necessity..And I think you know again, you know, go back to the migrant workers. People blame them, right? They say why can't they just stay where they are? And I think that's not the right question. The right question is how have we not made it possible for them to actually stay where they are? And that's something we have to acknowledge and once we acknowledge that, we're basically saying 'how dare they want to survive' which is not a valid question to be asking, right?.And so I think you will see this in a lot of situations, where people say well you know people are not following certain rules or they're they're going out when they're not supposed to be going out, or X or Y. Well then we've got to ask you know what compels them to do that and can we change that? Can we change the environment or the context around people and can we design policies and rules in a way that takes this into account? I mean that's the fundamental thing I think that all of these fields tell us is that you can't, policy cannot be designed in a vacuum, right?.It cannot be designed without reference to people's, you know, experience and so I always go back to the example of, you know, it's great we're telling people to wash their hands, but we've got to acknowledge that you know there's like some large percentage of people in the world who don't have running water, who don't have clean water to wash their hands, who don't have soap, right? So this is pointless advice. What are we doing to give them that, or to enable them to have that?.And for any of these kinds of crises the more equipped we are in terms of the infrastructure of a public response, a safety net for people, ways of distributing things in cash in kind, that can be ramped up, they help, right? So it's today it's COVID-19, tomorrow it might be a massive recession of some sort, you know, it might be some other health issue, it could be a natural calamity. If you have systems and you have a way, then you can sort of ramp up and ramp down as you move, but you've got to invest in those in that capacity, if not, you will just be stuck with you know scrambling to do what you can..AA: I'm sure our audiences will glean a lot from what you've said. Thank you for joining us today..SD: Thanks very much. It was a pleasure..***.More podcasts and interviews by us.