 Good afternoon, everybody. At least good afternoon in Europe. I know that there are participants from elsewhere, so I probably should say good day, everyone. Welcome to the fourth wider webinar on COVID-19 and development. The webinar series features a lineup of imminent researchers and development specialists presenting new research on the implications they foresee of COVID-19 for global development efforts and economic and social impacts for the global south. The idea of the webinar series is that by presenting the research findings on these urgent matters and by discussing them together, we're able to not only make the issues visible, but also to contribute to finding solutions to them. My name is Fin Taup and I'll be chairing today. And today we will discuss Africa's lockdown dilemma. And let me now briefly give you a short introduction to the speakers. First, there's Eva Maria Ekker, who is an applied economist holding a PhD from the University of Sussex. She is a wider research fellow based in Mozambique, where she works as a technical advisor for poverty assessment to the Ministry of Economy and Finance. Her research focuses on rural development, migration, climate change and labour markets. And she has researched and freely worked experience in Brazil, Ghana, Peru, South Africa and now Mozambique. Her colleague, Ricardo Santos, is a UNU-wide research fellow stationed in Maputo in Mozambique as technical advisor to the Center of Economics and Management Studies at the Faculty of Economics at Eduardo Montlani University. Ricardo holds a PhD in economics from the Institute of Development Studies at the University of Sussex, a master's degree in economics from the Nova University in Lisbon and an MA in development studies from the IDS. His doctoral research examined the post-conflict labour market and education in Timor List. Eva and Ricardo will be the ones presenting their research today. And then we have Tillman Brook, who is the founder and director of its National Security and Development Center, ISDC. And Tillman is also professor at the Natural Resources Institute, NRI, of the University of Greenwich. He's visiting professor at the London School of Economics and Political Science and team leader, Development Economics and Food Security at the IDS near Berlin. Tillman is also the co-founder and co-director of the Households in Conflict Network and the principal investigator of the Life in Kyrgyzstan study. Tillman's research interests focus on the economics of household behaviour and well-being in areas affected by violent conflict, fragility and humanitarian emergencies, including the measurement of violence and conflict in household surveys and the impact evaluation of programmes in conflict affected areas. So with this very good lineup of three experienced speakers, I'm very much looking forward to what they have to say. Now, before handing over to the speakers, let me stress that during the session, please use the Q&A tab. So this is down in the row at the bottom. You can see a Q&A tab and there please write your questions there and I will then ask the presenters and the discussion to address them after the presentations. Please do put clearly your name so that I can see who has asked the question and then I will read the question to the presenters and to the discussion asking for their reactions and comments. And I'm very much looking forward to this because the topic of today is a very hot one. It's a very difficult one and it's an intellectually stimulating one also. So with these words of introduction, welcome all of you and now over to the speakers. Hi everyone and thank you so much for your interest in our presentation. I'm Eva Maria Egger as Finn already introduced me and I'm going to start with a presentation before Ricardo takes over and we are presenting today a study that we have recently published a work in paper called Africa's Lockdown Dilemma, High Poverty and Low Trust. And this is a study we conducted jointly with our wider colleagues Patricia Justino, Sam Jones and Ivan Manik and who are all part of the Mozambique team of Univider. But before we dive into the presentation, let us first get an idea of what you, the participants of the seminar, think of this topic. So we prepared a small opinion poll for you to respond to. We have two questions that you will now have time to answer. The first one being, do you think that strict lockdowns are the appropriate response to COVID-19 in sub-Saharan Africa? Yes, no or localized yes. And the second question is, how high is the risk for social unrest in sub-Saharan Africa if strict lockdown was introduced? Do you think it's high, medium, low or there's no such risk? Please respond now, we'll leave you some time for this. I hope you all cast your vote and we will look at the results a bit later in the presentation. And now let's dive into the topic. So as was mentioned earlier, we are looking at the question around lockdown in lower income countries. Because as we know, there is quite a gloomy outlook of the economic impacts of the pandemic. And so we know that poor countries actually face a huge challenge by trying to minimize these economic impacts and at the same time trying to contain the virus and to reduce the burden of the pandemic in their countries. And as Mozambique country team, we were sitting earlier this year, still most of us in Maputo. And we saw how the pandemic was unfolding in European countries and North America and lockdowns being introduced. And first cases arrived in the African continent and we started wondering what advice can we give to policymakers is a strict lockdown the right response for a country like Mozambique or comparable countries. So obviously, there's a huge challenge that these lower income countries face on the one hand, and living conditions are relatively poorer. And so maybe living conditions don't allow people to actually stay at home for a certain amount of time and practice social distancing. Also another huge challenge is posed by high economic informality and lack of social protection or things such as unemployment insurance or the state, the fiscal capacity of the state to support those who are unemployed and do not have a regular income. At the same time, going beyond that, we thought about the fact that actually containing a virus and imposing a lockdown and stay at home measures poses a collective action problem. You can only contain the virus if everyone abides by the rules of distancing and staying at home. And this has been shown in the literature that this type of collective action problem can actually be addressed if there is a certain amount of trust in the government's effectiveness in addressing these issues but also in of trust in others. If you trust that your neighbor will also stay at home and keep distance, then this give incentivizes people to also buy to these type of measures. So we were wondering if a full lockdown is not feasible because of these rather poor living conditions as well as economic constraints to lockdown can hire trust maybe help to upset this lower level of preparedness. And if this is not the case, is there risk of social unrest. And we focus in this study on the Sub-Saharan African continent and by assessing a level of lockdown readiness for the countries and also looking at how this relates to trust and the risk of social unrest. And we use Afrobarometer data from 2019 for 30 countries, which I will quickly describe to you here. So the Afrobarometer is an initiative that conducts public attitudes and opinion surveys on topics like democracy, governance, and also economic and social factors. There's several waves already and the most recent data set available is actually from just last year. So we have a sample of almost 38,000 individuals in 30 countries in Sub-Saharan Africa. We eliminated North African countries to focus on common characteristics and a huge advantage of this data set is that it includes information about both the living conditions of individuals but also about their opinions and their level and indicators that will help us to measure trust as well as risk of social unrest. And we complement this data with some information also from the World Development Indicators. So what do we mean when we talk about preparedness for lockdown or as we call it so-called lockdown readiness? So we have, we post the definition that lockdown readiness is the ability of a household or family to stay at home and avoid public spaces without irreversible damage to the health and welfare. As you can imagine, in many contexts staying at home means that you need, for example, safe drinking water or sanitation system at home because otherwise you will need to leave your home several times per day and per week to actually, for example, go to a communal tap. At the same time, however, we also consider the fact that in low income country context, people require either irregular income and if they cannot work from home, which the majority of people in these contexts cannot, they would rely on savings. And so we try to find variables that capture this type of economic security to some degree. So we construct a very simple measure composed of five variables and that are actually easily available in most common household surveys or also in census data. And as we did, for example, for Mozambique already. And these components specifically are safe drinking water access in the home, basic sanitation and source of reliable energy in the home. And this is complemented by a measure of whether the household has access to information or communication via a mobile phone or fixed telephone, as well as a measure to capture whether the individual has an employment that provides a regular frequent cash income, or to say differently, the individual does not frequently go without cash implying that they have a lack of savings or don't have regular incomes, which would force them to leave the house to buy the food that their family needs. And so we also differentiate here between partially ready households and fully ready households. Partially ready are those who actually have access to basic services in their home. So with additional support informed, for example, of cash transfers or food transfers. These families could probably stay at home for a certain amount of time, because they have access to safe drinking water sanitation in their homes. And then the full readiness would be if they also are prepared in economic terms and access to information terms. So let's let us look at the results. What we find in the data is unfortunately, but maybe not unexpectedly a relatively gloomy picture, only 6.8% of all the households can be considered fully ready. And however, if you consider urban and rural areas and separately around 12% already in urban areas, whereas only 2.5% in rural areas can be considered ready. And this on the one hand, of course, this means that we that in rural areas you find very poor access to basic services or stable employment. At the same time, rural areas do not face such a high risk to the virus spreading because of much lower population density. So we will now focus a bit more on urban area results. And because of this context. What is not shown here, but interesting to mention is that actually access to information, for example through phones is relatively high. But the really low and the big indicate the indicator really driving this readiness down is access to a stable source of income, only on average around only 14% report to have the stable source of income. So if we consider those partially ready, we see that relatively more households have access to basic services, however, with a huge variation again across countries so in Liberia, only around 13% of families can be considered partially ready, whereas in Argentina, Senegal, this goes up above 80%. And then, of course, you might think that our readiness measure is maybe just capturing what also income levels are capturing. And to some degree, this is of course true that these are closely related. So here what we do is we plot real income per capita against the level of readiness at the country level. And what we find is a log linear relationship and what does this mean is that if you would approximately if you would double real incomes. We associate it with an increase in just 5 percentage points in the share of ready population. So, actually income levels alone do not determine how ready a country is. And now I will hand over to my colleague Eric Cado on the second part of the presentation. Thank you Eva. And so as we as Eva said, if we if we looked into into readiness and and the picture isn't necessarily very positive. We would try we try to look into to trust and see at what level and if it would be possible for trust to be an offset of of this lower readiness. And to do that, we looked into two dimensions of trust. And one dimension that looked into the capacity of collective action within the communities. We looked into dimensions of trust we we looked into what is trust towards institutions trust towards state and its representatives. And it's basically we did it by constructing a variable latent continuous variable that is standardized. And it's basically is constructed with answers to the to the questions on trust, the people have on the president, the parliament, the police force and traditional leaders. We also looked into the horizontal trust so trust amongst peers or within the community, and to pick that up. There's no direct question that picks that per se but a very good way to pick it up is how much do you trust the person, what close to you the vendor close to you where you buy your groceries. And on Afrobarometer there's a question that asks basically when a vendor sells you grains, how sure are you that you get the correct amount of change. And if if the answer is positive if the person is sure then it shows that there is a level of trust at the community level. And a third dimension that isn't necessarily a dimension of trust but does tell something about how able is the community are people to organize themselves for community for collective action, including possibly protest. So what we do see we do basically construct a variable that is again, latent continuous variable and based on answers to to to whether the person, the people are members of groups or participate in meetings. So these variables with these indicators of trust and capacity of collective action is that we relate them now we're going to look how do they relate with lockdown readiness and then how together lockdown readiness and trust may relate with with some activities that are related to protest. The first exercise we did was we we constructed a log linear model and basically what we did is it this is not very common, necessarily in economics more common in other social sciences. So we run a regression on basically would be a multi way contingency table and what we tried to do and we will try to pick up is how likely is someone within the the sample that we have to have a particular characteristic that can be picked up and that that we want to pick up. And this is given by odds ratios. So to read the table, we have to do something which is sometimes not very usual, which is, so it's not whether the signal is positive or negative it's if the if the coefficient that we have there the value that we have is above one or one that actually it's above one that means that there is a higher probability in this case of finding someone with that particular characteristic. If it's below one. It's a lower characteristics, let me illustrate for instance. So if we're looking to fully ready in the first column where we have a baseline in this in this particular in this particular regression that we did we basically established a baseline and we assumed what if there's no relationship between someone being fully ready and have of showing inter institutional trust or having showing capacity to associativeism, what if there isn't all of these are independent. So there are no inter relationships between these characteristics. So if you look into the first one and we see that there's a very low probability of finding someone that is fully ready and we already had seen that before, but this basically confirms so 0.002 is significant and very, very low. And we also also see that basically all of them are below one so basically it's more likely to find to find someone that's not fully ready someone that actually doesn't show institutional trust that actually doesn't isn't very prone to associativeism and looking only to those characteristics that show to be significant. Now more interesting is actually to look into the next column so column two, three and four. And what we see actually if there is a higher probability of finding someone that is both fully ready and trusts the state and also significantly higher probability of finding someone that is fully ready and trusts their peers. And when we look to when we look into column three what we see is that institutional trust and the relationship with readiness mitigates a bit. It becomes non significant, but the readiness the community trust and actually keeps being significantly correlated with being fully ready. And amongst the higher income group there's no no such signal so what is important to what we can see and find here is as an important. The reference point is that if anything, there is a positive relationship between readiness and trust and so the offset effect isn't there. And actually, then it's actually more felt amongst the poorest in the population this this eventual positive relationship. So next slide man. Thank you. And here what we look that is what's the likelihood of social unrest, given levels of trust and lockdown readiness. And again, I think the first signals if we look, first of all, these are also odds ratios so they should be interpreted as as before. And, and again the signal might not be the one that we would hope for, but so it's more of a high opener. First of all, I looking into the urban areas to it's important to notice that people in urban areas are more likely to participate in protest than people in rural areas and also are less likely to like or to accept curfew. Then what we also see is that institutional trust and community trust correlate negatively with participation in protest so the more people trust the state and more people trust themselves when it's significant, it shows that they would be less likely to But the capacity to mobilize. I think it would be expected actually indicates a more like a higher likelihood of protesting. But if we look into readiness, particularly in urban areas we see the more, the more positive the situation is for families so the higher the number of ready dimension so dimensions in which household is ready. The less likely is that people participate in protest. So again, readiness and trust go together towards more peaceful setting and lack of readiness and lack of trust will point out to worse, more higher likelihood of protest. And if we see how people would would agree with with the government and perceive the government as working towards reducing inequalities. What you see again is higher trust and higher readiness would are significantly correlated with more trust that the government is actually pushing for lower inequality. And finally, the point that I made before higher institutional trust and and more more and being more ready for lockdown actually increases the acceptance of a curfew by by the population. So all in all, what this actually tells us is that it's in with high probability, we wouldn't we shouldn't expect trust to be offsetting at least in this context of Sub-Saharan African countries trust to be offsetting lack of readiness and as we as ever mentioned before, we were not seeing a very positive picture in terms of readiness. So what, what should we conclude is locked down an option then for countries in Sub-Saharan Africa. And what does trust play what role does trust play in this fight against COVID-19. What are the risks of social unrest. So before trying to look into this, let's see what you've thought of what you answered. So, these are the results that you have and so I in general I think the majority of you were already a little bit septic, skeptical about the effectiveness of strict lockdowns, but still the majority would expect would propose that localized lockdowns would be appropriate. So I'm not going to try to delve into what each one of us thought with localize meant, but there was an expectation that it would in certain situations work. But I think even more in more significant is that 91% of those that responded think that there is a relevant risk of social unrest. But the majority actually thinks there's a high risk of social unrest. I think if anything, our results are confirming your expectations. So, what can we conclude then. So the findings do suggest that especially in poor countries, people are not only least prepared for lockdown. Actually, their trust in government and also in the community level is lower and the risk so therefore the risk for social unrest is higher. And that is also particularly noticeable in urban areas. So what can be done about it. Lockdown is one, one measure to push for social distancing. Of course there are certain measures policy that have been put in place and we don't. We are not testing we didn't test any any policy measures so we will not be advocating for one over the other. We're just trying to figure out whether there are possible measures that one would would consider and of course we have the use of masks and community masks is now being a general generally used throughout and also we don't need to necessarily lock down people into their houses. But some measures can be made of not of not having opened certain certain places or not organizing events where where concentration of people is more likely. But what we could do could look into is the world of social protection policies and this could be important into creating conditions for families not to suffer so much with with the lockdown and being able to actually endure it a little bit more and reduce it and that could be basically via cash or food transfers and in other situations this has been explored by by and and showed to be effective by Titus and pexson, the one and bank and justin and martirana. If anything, it's it's clear that this is a huge challenge and one where where governance will be tested. If any silver lining can be found is that if governments actually find and prove themselves and effective in handling this crisis public trust in them with actually can actually increase. So, hopefully that challenge can be met. And with this knowledge that lower readiness and lower trust is present. We can find a way. Thank you. Thank you very much, Eva and Ricardo. And we will now turn over to the discussant. Tillman book. Tillman, the floor is yours. Thank you very much. Finn and Eva and Ricardo for both organizing this interesting seminar and for your very interesting work. Before I start with my task as a discussant and earnest. I would like to have a brief advertisement. And because with Patricia justina who is one of the co-authors of this paper and uncle her flat the University of Constance. We have started in March, a survey project called life with corona, which addresses very similar topics to the ones that we're discussing here in the seminar today. And if you would like to fill in the survey which is a global survey and 25 languages, please feel free to go to life with corona.org and take the survey there. There's one wider working paper out already where we look at trust and I will cite some data from our paper and from our survey. As I discussed your paper, it's very closely related it's like a twin project you could say I think. So that's the end of the advertisement. And now coming to your very interesting paper. I think it's really a thought provoking and it's really interesting and I'm glad you made it because I think we need a better understanding of how not just the pandemic. The pandemic is impacting all of us in terms of the health impact, but also how the fight against the pandemic and is impacting us and the first measure of choice until we get good drugs and good immunization in the fight against corona virus and covert 19 are behavioral strategies so the Americans are trying to encourage us to do, you know, sort of less of this and face scratching and more of that, you know, safe coughing, and then more of this sort of social distancing. And of course in the extreme, not at the individual level but at the collective level lockdown although a challenge in the moment the idea that this is an extreme measure, and at least everywhere at all times so we have these new social phenomena or policies or strategies and we need to understand them how they work, but also what impact they have in turn because maybe for some people the impact of the fight against covert 19 is bigger than covert 19 itself, especially if you're not affected with the virus itself. So it's asking very important conceptual questions and it's asking important empirical questions and it's using which is also to be commended existing data with all the limitations that the audience has already noticed in the very interesting q&a. So a couple of things that you are effectively doing, although these are my words and not yours, you know, how can we deconstruct lockdown what does lockdown mean is there any ways of measuring it or approximating it. And I think it's great that you have this sort of readiness indicator index yeah that's that's a great concept and very useful and very practical. So when does it work and for whom. And what are the implications of this yeah all the way to social unrest which sounds very scary at first sight yeah you're using it a bit as a sort of like, that's something you want to avoid yeah, which in times of not having a pandemic. You know, if they're not violent, I mean protest are healthy right so they tell the government what people think so you might want to step back a little bit from just looking at it as a negative thing, although of course, you know protesting safely in times of corona is a tricky organizational thing. So let me just comment briefly on a couple of concepts and a couple of empirical things and I'll try to raise through as things only 20 slides or so it's not very much but I'll try to be quick, so that I don't take more than my 10 minutes. First the the concept of the lockdown the the purpose is in my view, not to stop the pandemic and I think you have to maybe spend a little more time on, you know why people do or why governments and post lockdowns yeah, I think all it can do is to buy time, and the first hit by the pandemic needed to lock down first some did it some didn't you know differential impacts there, but the ones who did it early, and they benefited most, and they bought time to come up with other strategies, social distancing, not touching your face, coughing and all that yeah. What this means is that lockdown is only part of a portfolio of measures, it's not the, you know the silver bullets that we would like it to be, because when the lockdown finishes and it logically has to finish at some point. The virus is still there nobody can lock down for 18 months until some scientist comes up with a with a vaccine right so so naturally it has to have a an end data sort of a half time yeah where you where you have to stop relying on it. So that's that's an important thing to bear in mind. Now what are the characteristics of a lockdown lockdown is not total, unlike the name may suggest because we really locked down, I mean you couldn't do anything anymore. You know, unless you have three months worth of food in your basement or your house on your hut or in your tent or wherever you live. You know it's not going to work and you're not going to have any nurses and hospitals or doctors you're not going to have any train drivers not going to have any. You know the hospital or the doctor and so yeah chances are you have some people who work. Yeah, and, and so that it's a, it's a thing of degree, right it must necessarily be a thing of degree, unless we're like a nuclear war and everybody dies who leaves their bunker. Yeah, so. So that's one thing. The other thing is geographically it could be a national thing, but it could also be a regional thing. It could be a local thing in fact an individual lockdown and things usually called a quarantine. You know it's sort of the spatial thing and as we process or progress I don't know if that's the right word if we proceed maybe rather with the pandemic, we will find we've bought the time at the beginning and some countries. We know what else can be done people got at the point basically so we can use more you know disinfecting all the other measures and maybe use less lockdowns but we will need them if they're outburst locally in order to step them from becoming national outburst again so I think that's something important. And now the third thing I noticed is that your lockdown index is quite sticky because the type of variables you choose don't change quickly right I mean, you know these are big topics employment wash electricity these are like decades long development agendas yeah and, and so basically the policy implication cannot be to try to improve preparedness, because these lockdowns have to be imposed short notice we can't prepare for them really. But at least now it's too late if you haven't stockpiled on the food you know it's too late by the times lockdown starts. And so I think it's more about helping to understand coping which is of course part of what you look at, but I just want to sort of contextualize it a bit. Okay, so, yeah, and then some people ask already in the discussion, you know GDP might be one thing I think if you looked at other concepts like poverty or inequality. I'll come back to this comment later if you look at the deviations from your fitted line in the cross country. Regressions yeah, maybe it's be interesting to see you know what explains that deviation, you know there's some outliers some countries who do really well for low GDP they're actually very well prepared and others are badly prepared and maybe it's like population density maybe it's poverty rates for a given level of GDP etc so you know be interesting to to play more with that. Yeah. And I already said that lockdown is a sticky concept trust on the other hand is not very sticky. And it may be very asymmetric it's it's it's very hard to win somebody's trust and very easily lost. Yeah. And I think that's something that we're seeing now there are some countries where the governments where we trust it in principle but trust is slipping. It's a vertical trust is vertical relationships are tripping so it's much more volatile than you locked on preparedness and so you're having a sticky concept and a lot volatile concept and you need to think maybe a little bit about what that means for empirical analysis I think the other things the polarization again as we move out of lockdown and somebody commented on that that that's an important topic I agree with that comment and you know it seems to be polarization some people true believers in lockdown and other people you know don't really care they go out partying and so on yeah and so so there seems to be something happening you know something pulling us apart yeah and and it's not so much the average level of trust that's of interest but it's the polarization into trusting people and non trusting people if you like yeah so I think that's interesting. Okay I've said this thing about the going in versus coming out. Yeah, and very briefly two pieces of data from our survey. And here this plots for three countries were early on in the survey we had a lot of responses I'm sorry there's not developing countries but I think in principle these are universal issues that we're looking at, and the share of respondents who support measures by the government, and that is dropping in several countries that we plot here over time. These are different time periods I'm sorry yeah sorry it says there. So this is like early April and late April and early May so you know blue and orange, and these are quite significant changes, yeah, in the response at sorry in the support for government measures yeah so I hopefully, you know that shows how dynamic this situation can be yeah nobody's going to get an extra toilet or more drinking water mobile phone contract during that period. So if you show very important trust. So if you show very low trust or low trust, then you have very little relatively little support in the government measures still actually more than 60% but still relatively low. If you show very high trust, then you are much more supportive of government measures yeah so trust support for policies policies etc. You know they really are very different. Yeah, so these are like clusters of data. Okay, a couple of quick measurement issues. And the density of the living space, I think must also be surely very important I wonder if something should be in the surveys there. So if you have large families, you know you, you can easily reinfect within the family, even if you're on lockdown, you know one bad cheap one, one sort of person, you know who gets infected maybe the person who has to go to work can infect the rest of the household, and also just simply the sanity of it yeah if you have low, no sorry high density. Many people per living space it's how much harder to cope with lockdown and if you have a large space in which a few people can can sort of disappear into yeah. And childcare homeschooling parental work I think those are all also important variables for making, you know it seems that a lot of families people of middle age, you know parents basically, and find it much harder to cope, especially if the kids can't go to school. Yeah, so homeschooling and all that. And so I think those are also other variables I think I'm a bit worried that if the index is too narrow I realize you have to do what you can get but these are other issues we have to make sure that we're not omitting them. Then you about measurements you group that into, you know, few factors fulfilled and many factors fulfilled and I wondered if you wanted to use the continuous because you otherwise just measuring the difference between three or less versus four or more and I think that you know you're losing some data basically so I thought that was a pity. And I realized afrobarometer won't be able to answer that for you but the lockdown readiness of a time again going in and out we have already discussed that things the dynamics matter maybe there's other ways of how you can exploit dynamics. Briefly on the interpretation as somebody also said in the comments we know that an urban areas protest are more likely, and that aligns with your findings which is good in principle. So whether the, you know, maybe you can explore the outliers a bit more and you know maybe you can tell a story even with country, like with individual data from certain countries, you know, study what happens there why are some countries better prepared than others and what impact does that have, maybe group by that. Yeah. And, and then again somebody in the comments said that so my job has been half done if you read the q amp a want to trust variables really mean they are very specific representations or proxies of trust. They're not really trust their sort of, you know, proxies and so you have to be a little careful. If you if you think the trader is a cheater then that's not the same as not trusting anybody. Yeah. And, and I'm a little worried about omitted determinants of unrest, maybe there's something missing like the blue holding sides together and very briefly. This is basically your theory of change from lockdown readiness to trust to unrest. That's plausible, I think that it's good and I'm glad you're doing it. But, you know, we have to question a little bit the connections yeah the question marks I get, you know, could it go the other way around so let's break this down a little bit we have from lockdown readiness to trust Okay, that makes sense. Yeah, if I really difficult and the government imposes something that maybe I don't like the government because they're disregarding my difficult work, my difficult living conditions. But you could also have from not being ready for lockdown because I have no job directly turn rest there's a huge literature on that. Yeah, so you don't need to go through trust so at least you have a sort of direct connection from lockdown readiness for example unemployment to unrest. Yeah, I think that's something you may want to account for in your modeling. And but maybe more importantly, you know, you could even have it the other way around, maybe an area where the population is very like this, you know, and is going to have less jobs and therefore less jobs and poor living conditions. I'm not going to go as an NGO go into an area where people constantly throw stones. Yeah, so I'm not going to work there. Yeah, so, so you know maybe there's even a reverse causality and you want to be able to resolve this easily I think with a cross sectional but it's just something maybe that needs to be accounted for very briefly last slide I think is locked down an option. And does medicine taste bitter. I think it does actually. And so the what I think on the one hand if you knew the benefits of lockdown you could really, you know, discuss the options and the sort of cost benefit analysis more. Or if you said what's the combination of measures that is feasible. Yeah, maybe there are other measures that are more feasible if lockdown is less feasible. Yeah, and so, like if you looked at sort of what do you need for information campaigns well you also need the communication for example so the communication is key, because if people don't have phones or radio you can't do communication campaigns or at least it's more costly. Okay I've basically said that. And final point I agree that cash transfers can help poor people in the pandemic. I'm not sure your paper causally proves that but I think what I'd be care beware is that I'm not sure you should sort of sell your paper with this sort of you know let's reduce the chance of unrest because I think the production function of unrest is very complex and as I said right at the beginning, you know a little bit of unrest if it's not too violent actually not too bad at least sort of protesting you know can be quite healthy. So, to sort of use that as a, you know, do what we say or else you get unrest in your country. You know I think it's a very strong and maybe overly sold policy prescription or motivation. Yeah, so thank you for your attention and look forward to the debate with the live the audience and our great presenters thank you very much. Thank you very much for this very succinct discussant comments. I think I want to turn to even Ricardo just to begin with just whether they have some immediate reactions and comments responses to your to your observations. Meanwhile, I will be looking at the questions that have come in and start preparing for those but Eva Ricardo, you have any reactions to gentlemen. Yeah, maybe I quickly start I mean first of all thank you very much Tim and I think all of your comments are very, very thoughtful and really useful insights. And I think I probably will also answer indirectly to some of the comments that I saw in the Q&A that you're absolutely right that our kind of chain of the argument is based a lot on the fact that what we measure and the way we measure it is assuming a certain initial condition situation right to say okay let's say before the pandemic arrives. What are the conditions under which people are living and can be from these conditions then expect the population to be able to deal with a strict lockdown so I think it's very right to to to mention that we have to be very explicit about the assumption that we are making here and that we are not that indeed our type of lockdown readiness is not at all claiming to capture everything that could be important and is important for living through a lockdown and it's also not able to actually or it's not we are not intending to capture the dynamics over time with this type of lockdown readiness because I think that exactly would then require much more refined approach to be able to really look at all the different aspects that we now observe this life under different scenarios and measures that is requiring and maybe another thing to mention that also might respond to some of the questions is that we are explicitly not looking at the health sector as such right so we are not including in the preparedness the idea of is the health sector of a country actually able to do contact tracing do testing adequately to identify whether there are suspicious cases and so on so forth and to deal with the pandemic so that's a research that we really trust the health economists and health experts to do so we are really approaching it more from the social economics side and yeah that would be my response. Thank you. Thank you. Let me just echo first the things that from that ever said to tell me about the questions they are really good they really help us think and to to structure what what are our ideas. I think I was going to take I'm going to take what some points and one that I find very interesting to mention and we actually value very much what we what you said and agreed that this is not about GDP per se and GDP actually doesn't show up as one of the variables necessarily that look into they don't it doesn't it doesn't come into readiness per se we basically what we found is that there is a relation between being more ready rest ready and the indicator of GDP per capita and and finding that first of all is log log linear which means that basically increases in GDP become less and less effective that's what characteristics of the logarithm and second that when you mentioned that it actually related with with a lot with poverty. Yes, very much so it actually relates a lot with multi dimensional poverty. And we would be able to find those indicators of of lockdown readiness to to be akin to indicators that are used when when looking into multi dimensional poverty. So what what I think a very important point that that I think we we brought, but maybe not very explicitly is that the of the initial conditions that ever was was mentioning and, and this is true on both on both indicators of of lockdown readiness that we that we looked at and were not able to look into with with the data we have, but also on this on the kind of like the baseline of trust, we could find at this moment when when we look into the information and making use of the fact that the after barometer surveys that we we used our very, our contemporaries are almost contemporaneous to to what we're finding what we're looking at right now. So they kind of give us a baseline of what how would be, how would it be possible and how what would, could we eventually expect to be happening in these countries. If a strict lockdown was to be put in place. One of the things that I think might have motivated us to go this into an underlying idea that we were looking into a strict like lockdown is that what we found in countries like Mozambique where we work and in other countries is that they moved to a much higher form of lockdown. When from the get go in very low cases a very low baseline of cases, and they already moved in immediately moved to a to a higher level of lockdown, then we found in countries like in in Europe, or in other countries, so which would make us kind of a higher risk of moving into an even more extreme form of lockdown if if numbers increase. So it would be so we found that important to try to measure the bed the baseline on whether the families whether the households would be ready to endure. And they were asked to be in a in a stricter lockdown and, and whether they would trust their, their governments in those policies, and eventually react against. I think one of the points that maybe on the theory of change we could look into, but basically, we don't think maybe we should be more explicit but it's not necessarily the case that lockdown readiness. There's that there is a causal linkage between lockdown readiness and trust they actually explicit correlation that we look there we don't those. What we what we look into is whether there's a likelihood of finding both households that both have higher lockdown readiness or higher capacity to endure the lockdown and higher levels of trust towards towards the community and towards the government bearing in mind the notes the points that you raised about how how strong are the, or the proxies in actually picking up and and and saying this is trust in the community or this is trust in vertical trust. So it's a good point on social protection very good point that not to sell it as an arrest mitigator. We actually look into more as a mitigator of the consequences of lack of rocked on radiance. And it's more about trying to mitigate the, the effort the stress over families, and given that they might not be. They might suffer from lack of capacities on those characteristics. I'm going to shut down and give space for questions. Okay, thank you very much Ricardo. So we will now turn to the questions that have come in. And basically, I will have to be quite choosy. And then some of the others, you would have to get responses to in writing. But I mean there is a I think personal question from Simone was asking, is employment the most reliable source of income in times of cobit 19. Our sources, e.g. pensions government transfers, necessarily less stable. So so so that sort of relates to the to the index. And I mean that there is a really good question, not too specifically to what Simone is asking, but also to the index. And that is Kabuga who's asking, please comment on the impact of food supply chains and lockdown readiness, noting that in countries of higher compensation curfews have immediate negative effect on households and may result in unrest. Could you maybe kind of elaborate a bit on what do you see as a sort of missing. And are your assumptions necessarily right. I mean, the regular indexes as I saw it doesn't have a food availability in there. And so I mean, you might want to just comment on these two first questions. So I quickly start and and this is, of course, are all very good points and we have looked, I would want to blame the data right as usual it's like it's all about data availability but it's not for the only reason right so we have we also looked at a question that was asking have you gone without food that is I think there's some related questions in the afro barometer. And the results are not that different so the overall picture remains very similar. And one idea why we kept it fairly simple was that actually this way we can also look at okay this differentiation between okay basic services for staying at home, and then the other dimension of readiness and not become too complicated and then having the challenge of how to actually aggregate it. And maybe regarding the the same is also true for for whether there's. Yeah, the employment is the only relevant source of income, and which we weren't able to look at whether individuals receive other sources of income such as government benefits but also for example remittances is a big topic in these times right now I expected to dry out. So I think that it has a lot to do with on the one hand data availability and at the same time, trying to keep things relatively simple to convey an idea, instead of trying to find the perfect exact measure. Okay, Ricardo you have anything to add. Nothing to add. Okay. And then there are a couple of questions on sort of relating lockdown readiness to GDP. One person's ask or states relating lockdown readiness to GDP is very misleading trusting capacity to do mass testing is even lower in countries like Nigeria that in countries with a lower GDP. But then, on the other hand, there is senior NAMO, who says, I am also concerned about lumping all these developing countries together. The results are likely to differ significantly. If the countries are grouped by income levels. And then there is the weighting by country population. So we have here sort of two quite different perspectives on, on relating to the red and the intake and GDP, I don't know, do you have any reactions to that. I think I've tried to stress that so the relation with GDP is one that became apparent, but it's really not one that we that we have on the on ingrained in the readiness. So, basically, the readiness, the lockdown readiness index that we produce it's generated from answers on on characteristics on the household and personal characteristics. And I think it should be very important to stress the point that Tillman made, there is a relationship with outliers and the outliers might tell very interesting stories that we should look at. I think I wouldn't make much of a conversation about about the GDP there other than there is an apparent relation and that this is not a direct relations actually mitigates with with with higher levels of GDP. I also like to note the point that as ever said, it's not about the readiness of the health system so it's not about the readiness of doing contact tracing to do contact tracing which is the readiness capacity of the health system that we looked at so that that would be something very interesting to look at from, but to be in a way a complimentary research question to our one we found what we looked at was at the household level. So readiness, how were people ready to endure lockdown, and even not how whether other measures would be more or less effective or are being put in place more effectively or not depending on GDP. That's not what we looked at necessarily. I don't know, ever maybe you pick up the second complete what I said and pick up the second point. Yeah, I don't know whether we have time it's. Please repeat even do we still have time for other questions or not. Well, there will be one more question coming for sure. I think regard has to address. So we might want to move on. So let me move on to the next question that is there which I find interesting, which is a question from Nina 10 year. What is what is asked is, did you look at government systems or the level of democracy in in sub-Saharan Africa as part of your research. Authoritarian or military regimes versus more democratic governance models and correlation with readiness for lockdown. So, this is a question I think you should be able to address. I think this is a very good question and admittedly we haven't yet looked at this in detail. And but I think it relates a lot also to Tillman's comments about the whole how what is exactly our chain of reasoning from the lockdown readiness and trust to the social unrest risk right so in participation and not in every context the bad thing. And that also relates to the type of regime as well as the question around inequalities and the perception of inequality that people might have. And they related to this the levels of trust in the government and how the government copes might actually differ by the political system. And they have there has been this the wider seminar on exactly posing that question whether autocracies might be better than democracies and handling the pandemic, because of certain, you know, power in imposing strict measures. But this is something we can definitely explore more with our data set and in this context. Okay, thank you very much. Now, there is a question from Shores, which I think all three of you, Tillman, Eva Ricardo might want to chip in on here at the end. I mean, Shores starts by saying first grateful for the good presentation. And then he says, African Mozambique in particular, I'm moving towards the most critical phase of the pandemic. Now knowing about their unpreparedness for severe measures to combat the pandemic. What are the possible ways out. And in other words, what to do. And I was wondering whether the three of you could briefly reflect on that. It does relate to also, for example, a question from the one that does who says, whether government preparedness and management influences readiness question mark, public health and institutions are important for readiness. What is the role of public policy. I was wondering whether this was a way to come with your final remarks and observations. And let me begin here now with Tillman. Thank you. So on the one hand, none of us are doctors, I think, yeah, at least not medical doctors. On the other hand, I think we exactly need to overcome that sort of silo thinking so we now, you know, as a result of this study, we probably had an intuition that was true before anyway. You know, we know that a strong wide reaching lockdown is practically not feasible. Yeah, and we'll lead to either people breaking the rules and therefore creating other secondary problems as a result of rule breaking. Or just people getting ill anyway, and it might even increase then the inequality, you know, based on some desperate people have to break the rules and the better off can stick to them. So I think we need both more, you know, if we need lockdowns more localized more intelligent to account for the circumstances of the locality. I mean, in a way, many African countries are still quite lucky for having relatively low infection rates compared to some other countries. So maybe a more localized approach as possible. Again, we need to buy time and we need to combine elements of a strategy, I think, and including a lot of public awareness and behavioral change strategies. Yeah, and so I think there's a lot of effort has to be put into that. And because every bit we can do to reduce the spread is worthwhile. Yeah, so there's our this famous our number. Yeah, so we can bring that from three to two to one, you know, we'll save lives. And so, I think that should probably be the starting point from my perspective as a humble economist. Thank you. Thank you very much, Tim. And there are a couple of specific questions to you also, which I'm sure you'll be able to address. One is asking whether they can answer your questionnaire twice, for example. So you will be able to see these questions afterwards. Yeah, I would totally agree with Tim and maybe to add is also that for the global community that this really is asking for support because there is, we all know that there is for poorer countries there is a risk a constraint in spending money you can't just go and now support everyone in order to have them stay at home. So that's not as easily done. And as said, and one thing I came across a lot reading about the covert challenge and poorer country context is that we cannot forget that there are many other challenges at the same time so it cannot be that now we focus everything on covert and forget address about addressing other challenges that still persist, but actually maybe to be more comprehensive in the approach we take that then will improve and the covert situation but at the same time still pursue the agenda of improving overall living standards. Okay. Thank you, Ricardo. And not not adding much more to what was already said in which which I agree I think what we find here is that it's very clear that the, the quick reactions that we had in northern in the global north, don't necessarily work in the global south. And that's a that's a very important warning that we need to need to think about which means the lessons the solutions that we were ready to put in place and probably were put in place more quickly in in the global south and more stringently in being more more stringent in the lockdown measures in the global south than than when we started in the global north. I don't seem to be sufficient and don't seem to be able to to give the answer that we need so we need to think about what are the answers and and this as Finn as Tillman said, I think, and I agree completely. It's something that needs to be done together with with the public health experts, economic economists and what we can give into this to the solutions and try to figure out what what we can do. But we're still trying to grasp what what we can do. Okay, thank you very much to our discussion Tim and Brooke. Thank you very much to the presenters Eva and Ricardo. Thank you very much to the organizers. And I mean, there's one sort of question, which was also put up which which I'd like to sort of addressing in concluding. And this is not to from Kenya who has finally what can think tanks do so that their voices heard by world leaders, compared to leaders listening to rumors. And I think one thing that think tanks can do is exactly what we've been doing this afternoon. So I encourage all of you to look out for the webinar series. I encourage every one of you to engage with the material available on the UNU wider website, as well as other websites. Tillman also listed his go there, get informed, a contribute, engage, and we have seen this afternoon that it isn't part it is possible in spite of challenging circumstances. I would like to say thank you very much to the more than 100 participants who have been in this part of the seminar and thank you to the 77 who are still with us. May you all have a very good rest of the day and see you in the next webinar. For some reason, I cannot open here the chat opened. And gentlemen is saying thank you to all for a great seminar and visit us at the website he has. Yes, so that's another added response to the question that came up. I apologize for those where I could not include the question I had to make quick assessments, either that they were already a holy or in part responded to, but the presenters will go through the questions and will then address those that they can after the webinar. So with this, thank you very much to everybody and have a good day.