1 00:00:02,410 --> 00:00:06,640 So can you start by saying your name and your current affiliation? 2 00:00:07,270 --> 00:00:12,190 Sure. My name is Petherick and I'm a department, a lecturer at the school. 3 00:00:13,090 --> 00:00:17,260 Thanks very much. And without telling me your entire life story. 4 00:00:17,260 --> 00:00:19,120 But just to give a little bit of your background. 5 00:00:19,810 --> 00:00:29,650 Can you tell me just roughly from how you first got interested in sciences, which I understand you studied first to how you got to where you are now? 6 00:00:31,600 --> 00:00:38,380 Yeah. Curious path. So my undergrad degree was Natural Sciences at Cambridge. 7 00:00:38,980 --> 00:00:44,140 And I would never have imagined at that stage of life teaching politics. 8 00:00:45,580 --> 00:00:48,520 Certainly. Certainly not in a public policy school. 9 00:00:49,270 --> 00:00:56,920 And I at Cambridge, you start off doing all of the different natural sciences and you sort of whistle them down. 10 00:00:57,610 --> 00:01:07,269 In my final year, I did population modelling and evolutionary genetics, which sounds completely disconnected to my job now, 11 00:01:07,270 --> 00:01:13,060 except when it came to COVID and then actually being able to understand that world, 12 00:01:13,090 --> 00:01:18,580 even though it's obviously moved on and I need to do it, undergrad level has been very, very helpful. 13 00:01:19,510 --> 00:01:26,049 And then from undergrad, I did actually mean to do a Ph.D. in these subjects, but I became a journalist. 14 00:01:26,050 --> 00:01:31,870 I was offered a job first in San Francisco, getting meant to do the Ph.D. in evolutionary genetics, 15 00:01:31,870 --> 00:01:35,650 but then worked for the economist in the nature covering science. 16 00:01:36,310 --> 00:01:42,040 Then out of the blue was asked to be a correspondent for The Economist in South America. 17 00:01:42,760 --> 00:01:49,360 And I thought, Why not? I'm at the time I was about 26, 27 and I mortgage no kids. 18 00:01:50,020 --> 00:01:56,979 Let's do it. And then I, I became dissatisfied with what I guess political correspondence, 19 00:01:56,980 --> 00:02:05,740 considered evidence with my natural science brain and applied to come here for graduate school, basically in the politics department. 20 00:02:07,130 --> 00:02:11,330 So not even comfortable about Nick at that stage. You didn't do the master's in public policy. 21 00:02:11,330 --> 00:02:16,100 You did. You did. I did. The Mphil in comparative government, which is a two year. 22 00:02:16,100 --> 00:02:23,630 Yes, nicely data driven course. And then I did the speech at the Dphil and kept the same supervisor. 23 00:02:23,690 --> 00:02:30,190 He's been a huge influence on me and his son, Tim Power, who is now head of social sciences. 24 00:02:30,200 --> 00:02:36,160 Back then he was just doing Brazilian politics. So hadn't. 25 00:02:38,420 --> 00:02:41,900 Yes. So you've moved between research and journalism. 26 00:02:42,260 --> 00:02:46,250 How do you think that those two professions complement one another? 27 00:02:47,970 --> 00:02:51,000 But it's an interesting question. So. 28 00:02:52,100 --> 00:03:01,910 I mean, it depends how you do research. The research I do today feels very much like a more thorough job of the journalism I used to do. 29 00:03:02,930 --> 00:03:08,990 But if you work for me particularly, you cover science and you cover it for someone like nature, 30 00:03:09,470 --> 00:03:16,220 it's very much a case of reporting on truth seeking, right? 31 00:03:17,120 --> 00:03:22,100 As though it's a first cut of of what we know of the present day. 32 00:03:22,700 --> 00:03:30,420 And I feel like particularly The Cove, it work or in general, the kind of work that often gets done in a school of public policy, 33 00:03:30,440 --> 00:03:35,720 school of government is trying to get at that just you have months and big datasets 34 00:03:35,790 --> 00:03:40,940 to answer the questions instead of a few days and the right people to call. 35 00:03:42,480 --> 00:03:47,490 That's is. So what kinds of questions really drive you? 36 00:03:50,100 --> 00:03:56,819 Gosh. Wow, what a big one. All kinds of questions, honestly. 37 00:03:56,820 --> 00:04:02,760 All kinds of questions. When I was an undergrad, I used to go to lectures from all subjects. 38 00:04:04,470 --> 00:04:08,310 Really? Really. I would go to anthropology, lectures, everything. 39 00:04:09,550 --> 00:04:14,950 But I think it was the way of answering questions. 40 00:04:15,670 --> 00:04:21,400 I felt like science taught you the best way to answer questions, at least that we figured out so far. 41 00:04:22,360 --> 00:04:29,980 And one year of my undergrad, I did History in Philosophy of Science, which is affectionately known as his [INAUDIBLE]. 42 00:04:30,310 --> 00:04:40,990 And then people who did that course and I think I just it's hard to I am very probably curious about the world. 43 00:04:41,680 --> 00:04:54,160 I'm fascinated by. Why people are the way they are, why decisions get made, even though collectively they can seem bonkers. 44 00:04:54,810 --> 00:05:02,080 Um. Why even, you know, animals evolve the way they do and. 45 00:05:03,990 --> 00:05:08,780 Yeah. I hope I just keep trying to answer questions that I find interesting. 46 00:05:08,820 --> 00:05:14,130 It's there's very few questions I find uninteresting about the world at large. 47 00:05:15,750 --> 00:05:25,360 That sounds like a great start period. So let's just move on to be more specific. 48 00:05:25,690 --> 00:05:33,580 Can you remember where you were, what you were doing when you first heard that there was something happening in China that might be worrying? 49 00:05:37,400 --> 00:05:45,160 I think like most people, it was probably around. Early January, late December, around New Year's Eve. 50 00:05:45,160 --> 00:05:50,680 And again, it didn't seem like it was necessarily going to be a big deal at that time. 51 00:05:52,330 --> 00:05:59,800 And you sort of keep your eye on it. So when I was in South America, I had to report on swine flu. 52 00:06:00,340 --> 00:06:03,790 And most people I spoke to at the time, experts said this is going to be huge. 53 00:06:04,660 --> 00:06:10,000 Mexico's economy is going to be completely destroyed. And, of course, it wasn't nearly as big as we thought. 54 00:06:10,030 --> 00:06:21,579 So I think I even actually went to Brazil around Valentine's Day for a week, and I was beginning to get nervous. 55 00:06:21,580 --> 00:06:26,780 That week I was in Brazil. That was when things started to be happening in Italy, in Iran. 56 00:06:26,800 --> 00:06:31,550 Just the beginning. And when I got back, I thought, Phew, I better not travel again. 57 00:06:31,580 --> 00:06:38,120 But still, even the last week of term Hillary, Tim, early March, 58 00:06:38,930 --> 00:06:49,310 I remember it felt day by day like the order of magnitude of importance of this thing was just exploding quite literally. 59 00:06:49,760 --> 00:06:55,489 I think, you know, I had all these other plans of what I was going to do and teaching and all the rest of it. 60 00:06:55,490 --> 00:07:01,190 And it just for the following term, I was going to go to a Jamie Cullum concert of sorts, 61 00:07:01,190 --> 00:07:08,239 and you just realise each day your view of what was possible shrank by almost 50% per day. 62 00:07:08,240 --> 00:07:11,240 That's how it felt. Yeah. 63 00:07:11,240 --> 00:07:16,340 So it was when it hit to you, it was that patch where you realised it was going to be global. 64 00:07:16,370 --> 00:07:22,520 Mm hmm. So at what point did you and your colleagues decide that this was something that you, 65 00:07:22,860 --> 00:07:30,890 your your research team here and your colleagues here could actually do something about and pivot your research in that direction. 66 00:07:31,730 --> 00:07:39,810 So the project is really originally all down to Tom Hale, and Tom and I teach in Hillary time. 67 00:07:39,830 --> 00:07:44,630 We teach the politics course that all the master students you have to take. 68 00:07:45,290 --> 00:07:51,350 He does the international relations bit, which is at the end and I do the comparing inside countries back to the beginning. 69 00:07:52,700 --> 00:08:05,599 And we were we run seminars every Friday and that last Friday of the issue we were supposed to be discussing was the Greek sovereign default. 70 00:08:05,600 --> 00:08:15,140 And it just seemed so out of place. And it's so obviously we you just couldn't keep the conversation on it. 71 00:08:15,980 --> 00:08:18,610 Equally, we have incredible students here. 72 00:08:18,620 --> 00:08:26,780 So you have our typical student who's been out in the world for about two decades doing things in terms of public public policy. 73 00:08:27,590 --> 00:08:36,560 And so I remember one of my students in that seminar that last Friday of ten, um, I was a bit annoyed about because he wrote to me saying, 74 00:08:36,640 --> 00:08:45,590 I'm going to leave halfway through because I've got to go and test something about kind of online teaching or online meetings. 75 00:08:46,310 --> 00:08:51,110 And I thought to myself it had to be doing my seminar clearly. 76 00:08:51,440 --> 00:09:02,930 And then it turned out it was for the World Health Organisation, not some cute project for teaching here and he was a consultant for the W.H.O. 77 00:09:03,290 --> 00:09:08,360 And so it was really in that moment of confusion about. 78 00:09:09,390 --> 00:09:15,050 Um, I guess what was going on with the exams that were coming up for students? 79 00:09:15,060 --> 00:09:21,120 Well, you know, the feeling that we couldn't really stick with the agenda for what we were teaching in the course. 80 00:09:21,780 --> 00:09:27,570 The the idea, I think, crystallised that we don't really know what governments are doing and nobody does. 81 00:09:28,200 --> 00:09:35,189 And what is quite unique about this community and the school is our students. 82 00:09:35,190 --> 00:09:41,190 We have about 100 and somewhere between 220, 150 master's students each year. 83 00:09:41,730 --> 00:09:44,910 And they tend to come from 50, 60, 70 countries. 84 00:09:45,910 --> 00:09:54,520 And if you are trying to have a community that can understand intelligently public policy announcements. 85 00:09:54,670 --> 00:09:59,770 But in any of the world's official languages, this is probably one of the best in the world to start with. 86 00:10:01,540 --> 00:10:04,929 And so, yeah, that was Friday seminars. 87 00:10:04,930 --> 00:10:15,290 And then I got a call from Tom on Monday and yeah, we met on a video call, which I don't think we've ever done before at that point. 88 00:10:15,310 --> 00:10:20,770 And yeah, the idea was born, he'd been cooking up the, the bare bones of it over the weekend. 89 00:10:21,220 --> 00:10:28,090 So what was the idea? Tell me about it. The idea was just a very simply start recording what governments were doing. 90 00:10:29,050 --> 00:10:33,220 And at that time there wasn't much in place. 91 00:10:33,220 --> 00:10:39,850 There was the Italian lockdown. That was the way you had lockdown and there was stuff going on in Iran. 92 00:10:42,040 --> 00:10:47,320 But it was at that particular juncture where there was a sense that it was going to be global. 93 00:10:48,670 --> 00:10:54,970 Now, the challenge was if you were recording public policies, how do you do it before you really know what they're going to be? 94 00:10:55,660 --> 00:11:08,470 And so our original coding scheme simply said, well, let's come up with the main areas of closure policy, so schools, transport and so on. 95 00:11:09,760 --> 00:11:14,900 And let's have the simplest possible ordinal scale to measure strength of policy. 96 00:11:14,920 --> 00:11:18,790 So nothing recommended closure required closure. 97 00:11:19,540 --> 00:11:22,990 And we had an indicator for some sort of public health campaign. 98 00:11:23,590 --> 00:11:29,590 And that was how the track is started, recorded on Excel spreadsheets, which seems a very long time ago now. 99 00:11:30,790 --> 00:11:38,500 And then over time we have expanded that. So we now got more than 20 indicators and far more nuanced ways of coding the strength of policy. 100 00:11:39,550 --> 00:11:48,100 So let's give it its name. It's called the Oxford COVID 19 government government response track and response track of this. 101 00:11:48,540 --> 00:11:52,050 It's very hopeful that after the Oxford bit, the rest is an alphabet code. 102 00:11:54,970 --> 00:12:04,260 And you talked about the involvement of the students. So how were you gathering the data to feed into the tracker? 103 00:12:04,870 --> 00:12:13,510 So the extraordinary thing about this project is that it is entire, entirely volunteer based has been from the beginning. 104 00:12:14,860 --> 00:12:22,570 Now we have a core research team involved in the actual organisation of it, who we pay. 105 00:12:23,140 --> 00:12:27,400 But we've had over a thousand people come through as volunteers. 106 00:12:28,370 --> 00:12:33,480 At this point. And so how do we get them together? 107 00:12:33,500 --> 00:12:41,270 So at the beginning it was willing students and I am kind of in awe of them 108 00:12:41,270 --> 00:12:44,900 because they had exams a few weeks away and they didn't know what was going on. 109 00:12:45,470 --> 00:12:54,590 And the idea of giving of hours of your day to a project that you didn't know if it was going to be big or small or whatever at that moments, 110 00:12:55,340 --> 00:13:01,940 you know, hats off to them. That's really the vocation of public policy as opposed to just the outcome of your Oxford degree. 111 00:13:03,110 --> 00:13:06,799 And since then, it's expanded to different communities. 112 00:13:06,800 --> 00:13:09,920 So we have a. 113 00:13:11,310 --> 00:13:15,330 Groups. So other universities in other countries have got involved. 114 00:13:17,040 --> 00:13:24,000 We had to add we have a collaboration with Microsoft, say some of their employees can get time off to do volunteer work. 115 00:13:25,410 --> 00:13:29,879 Yeah, we sort of it seems to find its way into new contacts. 116 00:13:29,880 --> 00:13:34,530 And then these people create a new community public that wants to get involved. 117 00:13:35,040 --> 00:13:40,680 Lots of medical professionals actually do it. Mm hmm. So what does each participant have to actually do? 118 00:13:40,920 --> 00:13:42,360 Right. So good question. 119 00:13:42,780 --> 00:13:55,200 And so we ask about four, five, 6 hours a week and they will be allocated, uh, perhaps a country to coach for that week or a few countries. 120 00:13:55,200 --> 00:14:02,970 And that means that they have to get to grips with the coding manual and then for each of the policies. 121 00:14:03,810 --> 00:14:10,290 So for how the strength of school clashes, for if there's different vaccine policies. 122 00:14:10,290 --> 00:14:17,610 Now we code, we look at test and trace all sorts of things. They have to look out for what official government policy is. 123 00:14:18,540 --> 00:14:25,020 And so they have to check government websites and all sorts of official forms, sometimes quality media as well, 124 00:14:25,440 --> 00:14:29,880 to sort of check whether things are really happening on the ground for some of the indicators. 125 00:14:30,480 --> 00:14:34,170 And then they input this into an online database. 126 00:14:36,090 --> 00:14:42,600 And that's that. What is really hard to explain to a lot of people is when someone I'm thinking of my mum, 127 00:14:42,600 --> 00:14:48,179 when they say that when someone's in fact coding for the week, it's life, you know, 128 00:14:48,180 --> 00:14:50,610 from their laptop in Peru, 129 00:14:51,090 --> 00:15:00,600 it then automatically via an API ends up on the Financial Times websites feed through to all sorts of epidemiological modelling around the world. 130 00:15:01,320 --> 00:15:09,629 We do have a sort of a check of everyone's coding, but that takes about a week or two to do so. 131 00:15:09,630 --> 00:15:15,870 The last little bit of data is slightly more kind of a rough cut than everything, a bit older, but that's how it works. 132 00:15:16,950 --> 00:15:26,580 So when it comes to looking at the effectiveness or comparing the effectiveness of policy of government policy responses in different countries, 133 00:15:26,940 --> 00:15:32,729 that that's that's a research job for which your resource is is actually acting. 134 00:15:32,730 --> 00:15:38,130 Is the the database, is that research being done all over the world? 135 00:15:38,150 --> 00:15:43,650 Or are you do you do you hope that that a basket of information here? 136 00:15:44,610 --> 00:15:50,040 So the idea the philosophy has been we hold nothing. Everything is available instantly. 137 00:15:50,400 --> 00:15:55,860 It's available free of charge and we make it as easy as possible for people to use. 138 00:15:56,760 --> 00:16:00,419 Um, we do do a bit of research ourselves, 139 00:16:00,420 --> 00:16:08,999 but it's primarily a data project and this certainly it would make no sense for us to try and hold the data and we wouldn't have the skills, 140 00:16:09,000 --> 00:16:13,440 the capacity to do different types of modelling that would be very necessary for this. 141 00:16:13,800 --> 00:16:23,400 We have ourselves had a few papers about what works and thinking about a different way so that the first way we try to 142 00:16:23,400 --> 00:16:32,070 think about it is thinking about how different combinations of policies and correlates with how deaths and cases vary. 143 00:16:32,820 --> 00:16:38,250 We did some very quite simple modelling compared to the kind of modelling that you would. 144 00:16:38,940 --> 00:16:46,679 The people I used to study with ages ago would do where they have much more nuanced idea of what your counterfactual is. 145 00:16:46,680 --> 00:16:52,890 And that's the natural curve of the disease is the typical statistical workhorses of the social sciences. 146 00:16:53,130 --> 00:16:54,690 It's quite hard to get to that. 147 00:16:55,440 --> 00:17:07,470 But where we've I think really made progress is combining our data with behavioural data from mobile phone mobility and from survey data, 148 00:17:08,130 --> 00:17:16,020 and then we can look quite closely at how different policies affect behaviour, either self-reported in surveys or objectively, 149 00:17:16,020 --> 00:17:20,729 but which people tend to have smart things about how people are moving around. 150 00:17:20,730 --> 00:17:25,260 And that's yielded some really interesting insights. Tell me a bit more about that. 151 00:17:25,260 --> 00:17:28,410 So what did the results, which were the most interesting results being? 152 00:17:28,650 --> 00:17:40,740 So a few. So one of the big policy questions early on was whether something called behavioural fatigue or pandemic fatigue existed. 153 00:17:41,940 --> 00:17:49,510 And this was very contentious because at least Sweden and at one point the UK suggested, well, 154 00:17:49,560 --> 00:17:54,810 he just had this sweet spot if you know when the policies are going to be really effective because people give up on them, 155 00:17:54,810 --> 00:17:57,510 they get sick of them, then we need to, you know, 156 00:17:57,510 --> 00:18:04,230 essentially wait and bring in the herd closure policies at just the right moments rather than put them in place now. 157 00:18:04,590 --> 00:18:07,770 And that was a big issue in March 2020. 158 00:18:09,720 --> 00:18:12,830 So we. This particular question. 159 00:18:12,830 --> 00:18:16,040 So how long are able people able to stick to the policies? 160 00:18:16,910 --> 00:18:23,570 And what we broadly found for the physical distancing policies so the stay at home and so on, 161 00:18:24,770 --> 00:18:34,630 was that people didn't manage to adhere to these protective behaviours nearly as much as they did in the first month over time, 162 00:18:35,390 --> 00:18:42,560 but it never fully decreased. So there was a decrease in then almost hitting a threshold, that kind of internal compromise, 163 00:18:42,560 --> 00:18:46,760 I guess, for how much you can handle and hold on to in the long run. 164 00:18:48,770 --> 00:18:57,230 And this was this is remarkably consistent across countries and also across different societal groups. 165 00:18:57,890 --> 00:19:04,130 So, for example, it might be that different ages or certainly different genders have different initial compliance behaviours, 166 00:19:04,520 --> 00:19:11,120 but the patterns of reduction over time just completely echo each other for different groups. 167 00:19:13,130 --> 00:19:20,330 Which is fascinating because if you see these same patterns in rich and poor countries and amongst employed people and unemployed people, 168 00:19:20,750 --> 00:19:25,940 it suggests that the logic isn't fundamentally about whether you can afford to stick to the policies, 169 00:19:26,900 --> 00:19:32,300 and if you then compare the patterns we see for physical distancing, 170 00:19:32,660 --> 00:19:40,040 which is a very individually costly kind of behaviour compared to something like mask wearing, 171 00:19:40,340 --> 00:19:44,960 which is very cheap behaviour, you see radically different patterns. 172 00:19:45,590 --> 00:19:50,090 So physical distancing has this descent to a threshold mask. 173 00:19:50,090 --> 00:19:55,700 Wearing is just just rises and rises and rises controlling for the strength of policy. 174 00:19:55,700 --> 00:20:00,460 So people adopt mask wearing over and above what governments tell them to. 175 00:20:01,730 --> 00:20:11,719 And then if you think about why that might be, well, mask wearing is habituated, kind of like wearing a seat belt or helmet. 176 00:20:11,720 --> 00:20:14,900 Like if you try and drive without wearing a seatbelt, you feel kind of naked. 177 00:20:16,390 --> 00:20:23,750 Where is staying? At home day after day after day is more like continuing to do press ups over time. 178 00:20:23,750 --> 00:20:31,520 Like each one gets harder, it's cost accumulating. And so that in itself is very fascinating. 179 00:20:31,520 --> 00:20:36,350 It also suggests that, you know, risk perception isn't a fundamental driver either, 180 00:20:37,190 --> 00:20:42,710 because if it was risk that was bothering you, makes you do everything as much as possible that could protect you. 181 00:20:43,730 --> 00:20:49,340 Right. So in some ways, to me that project sort of showed that, you know, 182 00:20:49,340 --> 00:20:56,570 the behavioural economist behavioural sciences have probably got the best insights as to what's going on, at least for that kind of question. 183 00:20:57,320 --> 00:21:02,600 The other thing that we found that was fascinating with that project was comparing 184 00:21:02,990 --> 00:21:11,780 countries with different levels of trust in institutions and in strangers in society. 185 00:21:12,470 --> 00:21:19,550 And certainly very early on in the pandemic, the assumption was that the more that people trusted their governments, 186 00:21:19,880 --> 00:21:21,350 the more they would do what they were told. 187 00:21:22,810 --> 00:21:29,500 And yet we saw countries with high levels of trust in governments have really high debts and low debts just all over the map. 188 00:21:30,850 --> 00:21:40,240 And what we saw was that when we compared the adherence patterns to protective behaviours, that high and low trust countries behaved exactly the same. 189 00:21:41,050 --> 00:21:48,490 But when we looked at whether people trusted strangers in their society, interpersonal trust, we saw divergent patterns. 190 00:21:49,030 --> 00:21:58,180 So if you trust strangers a lot, you're far more likely to stick to the physical distancing protective behaviours than if you don't trust them. 191 00:21:58,780 --> 00:22:03,550 And there are inter-country differences in this level of trust in both governments and strangers. 192 00:22:04,450 --> 00:22:08,410 See, we looked at above and below the median and yes, there's quite a big, big difference. 193 00:22:08,620 --> 00:22:14,440 Examples of countries. Yeah. So Germany would be a high trust country for inter-government and strangers? 194 00:22:15,970 --> 00:22:20,350 Yes. At least in the the data I have about Germany and its government. 195 00:22:20,650 --> 00:22:26,020 So one of the big questions that we don't know, of course, is how much levels of trust have changed during the pandemic. 196 00:22:27,100 --> 00:22:30,729 Actually, for Germany, there is now as an outlier, a country where there are good surveys. 197 00:22:30,730 --> 00:22:33,850 But for most countries, we don't know much about that. 198 00:22:34,360 --> 00:22:43,510 And I wouldn't be surprised if something this salience in people's lives sort of shifts trust far more quickly than social science is used to. 199 00:22:44,350 --> 00:22:47,740 Brazil would be a country where people don't trust strangers very much. 200 00:22:49,370 --> 00:22:59,680 And so the think the kind of the the political science logic behind all of this is actually collective action theory, just recast. 201 00:22:59,690 --> 00:23:05,149 So if you're trying to create a public good, so public goods like climate change, reduction, 202 00:23:05,150 --> 00:23:09,530 all that stuff, in this case, the public good would be reducing the infection rates. 203 00:23:10,630 --> 00:23:13,060 And you have individual costs of staying at home. 204 00:23:14,080 --> 00:23:19,540 The only reason you would want to pay that individual cost is if you trust that everyone else is doing the same. 205 00:23:20,050 --> 00:23:24,610 If you don't trust that, then why would you pay that high individual cost? 206 00:23:24,670 --> 00:23:31,320 You're not going to create the public good. And that's and that is what was found. 207 00:23:31,440 --> 00:23:42,530 That is what was found. Yes. Yeah. Yes. 208 00:23:42,530 --> 00:23:50,299 We might go into some some more findings, but I didn't ask before about whether you needed to raise funding to support the work you mentioned, 209 00:23:50,300 --> 00:23:53,990 you know, have some paid staff and how how easy was it to. 210 00:23:55,000 --> 00:23:59,620 Find a funding to do this work. The beginning. 211 00:23:59,770 --> 00:24:04,840 I mean, all of the institutions that create grants weren't ready. 212 00:24:05,170 --> 00:24:14,910 I mean, no one was ready at the beginning. And so the first few months, I mean, we worked so hard, we were doing everything, the routine. 213 00:24:15,490 --> 00:24:22,600 You know, I remember sort of 2 a.m. on a Saturday night trying to code Middle East countries with Google Translate open. 214 00:24:25,130 --> 00:24:28,520 And then I think it was about three, three months. 215 00:24:28,520 --> 00:24:31,759 And Rush gave us some money for this project. 216 00:24:31,760 --> 00:24:38,420 We heard about it not that much, but when you're desperate for a few research assistance just so you can sleep, 217 00:24:39,170 --> 00:24:41,900 that's pharmaceutical company, then pharmaceutical company. 218 00:24:42,410 --> 00:24:50,030 And then since then we've had a number of funders and you know, the main thing we need to pay for is just the research assistants to, 219 00:24:50,420 --> 00:24:54,230 you know, manage the contributors and the emails, check the data, all that stuff. 220 00:24:54,950 --> 00:24:58,250 And so actually our main funder has been the Blavatnik Foundation, 221 00:24:59,390 --> 00:25:04,850 but we've applied for various funds since then and been quite, you know, been quite successful. 222 00:25:05,450 --> 00:25:16,160 So UCL, we have a, an agreement with the Cabinet Office and Tom and I every Tuesday during COVID have joined a pool of academic advisers to them. 223 00:25:16,880 --> 00:25:20,360 Oh, you've printed other of my questions. Oh yes it was. 224 00:25:20,740 --> 00:25:24,880 Thanks. It was quite. 225 00:25:26,300 --> 00:25:37,050 And. So, yes, I mean, clearly, this was something that was worked very collectively on this with a very wide range of people. 226 00:25:37,470 --> 00:25:40,620 Did that feel different from how research is normally done? 227 00:25:41,490 --> 00:25:48,000 No, I'm sure it's something that, you know, research can be a very competitive arena to be. 228 00:25:48,270 --> 00:25:52,140 You know, it's extraordinary, particularly, I think. 229 00:25:53,190 --> 00:26:01,860 So I now have tenure. But when it started, I didn't. And a few people told me it was bonkers because obviously you can't keep up all your 230 00:26:01,860 --> 00:26:07,049 current projects and you the main emphasis was day to day to day to the other. 231 00:26:07,050 --> 00:26:13,140 People need a in an emergency rather than your own, you know, publications or whatever. 232 00:26:18,270 --> 00:26:21,750 Yes. So, I mean, where to start with how it's different? 233 00:26:22,860 --> 00:26:27,870 First of all, just the the collective energy. 234 00:26:28,950 --> 00:26:31,079 So it's it's volunteer driven. 235 00:26:31,080 --> 00:26:47,830 So just harnessing and frankly, enjoying that community spirit is how the project has carried on to this day almost, you know, a year and a half in. 236 00:26:49,610 --> 00:26:58,350 And that's just a world away from, you know, impact factors and how do you compare to other people and and all of that stuff. 237 00:26:59,070 --> 00:27:06,330 I've learnt so much about, I guess community science and frankly about leadership as well. 238 00:27:06,330 --> 00:27:17,310 We have a very sort of open hearted way of running the tracker, very much a sort of collective feel of what skills do you want to learn? 239 00:27:17,550 --> 00:27:21,450 How can we help you do that? It is exhausting. 240 00:27:21,450 --> 00:27:27,540 It's more time often to do it that way than just to sit in the corner and run your models and write papers. 241 00:27:27,540 --> 00:27:33,780 But it's taught me so much and none of it would have been possible if we'd been more selfish about it. 242 00:27:35,100 --> 00:27:42,470 That's really interesting. Yes. 243 00:27:47,540 --> 00:27:56,030 Yeah. I've asked you about how individually the volunteers do that data gathering and you put it together that you've, 244 00:27:56,080 --> 00:28:02,620 um, and I noticed from one of your papers that you've used data from our world in data and they've used your data. 245 00:28:02,630 --> 00:28:12,200 So are there other, other collective groups out there who are now collaborating really across the whole sort of data driven landscape? 246 00:28:13,280 --> 00:28:17,090 Yeah, so I have I mean, it's mixed, if I'm honest. 247 00:28:17,390 --> 00:28:26,470 It's been very much, um, kind of, um, I would say mushrooming of data sources at each stage, right? 248 00:28:26,480 --> 00:28:33,620 Of, um, you know, in John Hopkins, at least probably in the beginning and it's been a bit of a trying to put together a 249 00:28:33,620 --> 00:28:37,770 patchwork as these kind of different projects have got off the ground to check the data, 250 00:28:37,820 --> 00:28:42,140 can talk to it to each other's different datasets. 251 00:28:43,190 --> 00:28:46,879 And, you know, sometimes it's worked. 252 00:28:46,880 --> 00:28:54,800 Everyone has been insanely busy. That's the thing about the people running these data projects that if there's one barrier to collaboration, 253 00:28:54,800 --> 00:28:58,310 it's generally not the philosophy of doing emergency work. 254 00:28:58,910 --> 00:29:05,389 It's just the you know, we have four minute meetings sometimes that kind of you're in that world. 255 00:29:05,390 --> 00:29:09,620 It's unrelenting. Um, but things have settled a bit now, 256 00:29:09,620 --> 00:29:16,399 and certainly I've started working quite closely with someone in payroll quality 257 00:29:16,400 --> 00:29:22,670 who's been running surveys in lots of countries every two weeks to try and align 258 00:29:22,700 --> 00:29:26,120 or create synergies between the questions that are asked that you can only get 259 00:29:26,120 --> 00:29:30,860 at and survey data that are relevant to interpreting how policies are working. 260 00:29:32,530 --> 00:29:37,390 Yeah. Because that's the big question really, isn't it? It's either you're collecting data about what the policies are, 261 00:29:37,780 --> 00:29:43,510 but you really need to look at the effectiveness of those policies, which talk about, you know, how people actually behave. 262 00:29:44,020 --> 00:29:47,379 And that's that side of it. You're dependent on other people to collect. 263 00:29:47,380 --> 00:29:51,160 Is that right? Yes. Yeah. 264 00:29:51,160 --> 00:29:54,300 I mean, I. Yes. 265 00:29:54,320 --> 00:29:57,800 In short, yeah. 266 00:30:00,080 --> 00:30:03,080 So I mean, just to go back to sort of the big question, I mean, 267 00:30:03,080 --> 00:30:14,950 how how easy is it to make comparisons between countries or between substate entities made up of a similar enough view to make? 268 00:30:14,960 --> 00:30:21,860 Perhaps if you seem to I mean, do do the comparisons stand up one against the other? 269 00:30:22,490 --> 00:30:27,830 And you talked about using quite a simple scale to presumably try and get around that problem. 270 00:30:28,040 --> 00:30:33,349 Yes, it is. The answer probably depends on how what you what you're trying to do with your comparison. 271 00:30:33,350 --> 00:30:40,219 Right. If you're trying to answer the big questions about what works using policy data, 272 00:30:40,220 --> 00:30:48,020 then that's quite a bit harder than, if you like, simply comparing the strength of policy in different countries. 273 00:30:49,790 --> 00:30:58,250 And we have an interpretation guide for every possible ordinal jump up for every possible indicator. 274 00:30:58,730 --> 00:31:08,630 That is pretty exhaustive. You know, if, for example, schools are only opening for exams but aren't opening for any classes, 275 00:31:08,960 --> 00:31:16,790 whether how you code that in our system will be in the interpretation guides if buses are being extra cleaned. 276 00:31:16,790 --> 00:31:20,210 But that isn't affecting the bus timetable, it's in the interpretation guide. 277 00:31:20,720 --> 00:31:31,310 And so that kind of thing. I think we I mean, we have meetings every week to reflect on whether we're consistent enough, 278 00:31:31,310 --> 00:31:35,600 whether we're detailed enough internally with our code book. 279 00:31:35,600 --> 00:31:37,940 And then we have calls every week, 280 00:31:38,420 --> 00:31:46,850 multiple calls with different coding teams where we answer questions from all of the coders about issues and questions they have about what policies, 281 00:31:46,850 --> 00:31:54,620 how to code particular policies. And then we have WhatsApp groups which acts like help desks, which the research assistants man. 282 00:31:56,150 --> 00:32:04,310 And actually one thing to point out about whether you can truly compare, I think one sort of silence, 283 00:32:05,060 --> 00:32:14,209 huge benefit of this project that is fairly unrecognised is that because we have 284 00:32:14,210 --> 00:32:18,530 this objective code book to compare the strength of policies that I think, 285 00:32:18,590 --> 00:32:24,770 you know, stands up to a lot of battering and questioning, which it gets every single day, pretty much. 286 00:32:25,670 --> 00:32:35,060 It's really I think it's a political scientist move the conversation on in an age where there's so much questioning of the facts. 287 00:32:36,000 --> 00:32:40,560 It's the question of, well, we're doing this and you're not. 288 00:32:40,710 --> 00:32:45,410 Is that really true? Our policies are stronger than yours, even though they're quite different. 289 00:32:45,420 --> 00:32:48,570 And it's hard to seemingly compare just anecdotally. 290 00:32:49,620 --> 00:32:58,050 That's just not been a noisy part of the conversation until we've just moved the conversation on to what is working and what's not. 291 00:32:59,700 --> 00:33:06,990 What about if you've got a variable like how authoritarian a country is or even how 292 00:33:07,410 --> 00:33:14,260 well they score on your on on your stringency and other levels of what's being done? 293 00:33:15,090 --> 00:33:18,720 But then you've got different levels of social inequality in different countries. 294 00:33:19,110 --> 00:33:24,120 Does that have an impact on the results? Out of the economic inequality. 295 00:33:25,170 --> 00:33:29,580 So what do these things do? Affect results in different ways. 296 00:33:29,610 --> 00:33:40,020 So I wrote a paper about different forms of authoritarianism with some other academics here in the school to say Joe Wolf and my tutor, 297 00:33:40,610 --> 00:33:50,670 and sort of thinking about how leaders with an authoritarian style might respond to this kind of opportunity or moment in time. 298 00:33:52,860 --> 00:33:58,439 And what we argue is that depending on your type of authoritarian, you go one of two directions. 299 00:33:58,440 --> 00:34:04,260 If you seise the moment and you get it right, I can put and put in place the really strong policies and then hold them there. 300 00:34:05,010 --> 00:34:12,120 I say the big issue is whether there are sunset clauses or if you're the more science denying. 301 00:34:12,120 --> 00:34:13,730 I prefer small governments. 302 00:34:14,760 --> 00:34:22,830 You're in a bit of a corner when a pandemic comes along because you can't necessarily say, wait, time for big government, let's learn from science. 303 00:34:23,370 --> 00:34:31,680 And then you were in the Bolsonaro Trump camp of policies are very, very weak compared to other countries, at least at the federal level. 304 00:34:33,960 --> 00:34:40,260 And then for different questions about what works on the ground, how people are responding in terms of their behaviours, 305 00:34:40,260 --> 00:34:44,450 then things like questions of inequality and so on become far more prominent. 306 00:34:44,760 --> 00:34:51,980 Right. But I guess the all these different variables matter in quite complex ways depending on what your question is. 307 00:34:54,010 --> 00:34:57,219 And and I thought it was interesting that they. 308 00:34:57,220 --> 00:35:05,560 Were you surprised to discover that the the kind of wealth in the sense of the you know, 309 00:35:05,560 --> 00:35:14,290 the economic ranking of countries didn't necessarily relate to how effective they were at implementing their COVID presenting prevention policy. 310 00:35:14,560 --> 00:35:20,380 I think we all were. I mean, particularly not just sort of just about wealth. 311 00:35:21,580 --> 00:35:27,910 I think when you look at reports that were written before this pandemic about countries preparedness to handle a pandemic, 312 00:35:28,450 --> 00:35:32,389 all of these indices, which is way off. Absolutely way off. 313 00:35:32,390 --> 00:35:37,030 So the U.S. and the U.K. did very well on those. Yes. Yeah, we had. 314 00:35:37,090 --> 00:35:40,870 But then the performance didn't match up to that. Exactly. 315 00:35:40,870 --> 00:35:45,850 I think leadership going early. 316 00:35:46,450 --> 00:35:50,920 Coordinated policies have been very, very important. 317 00:35:52,990 --> 00:35:56,320 More so than how much you spend on your health system, necessarily. 318 00:35:56,380 --> 00:36:03,820 Mm hmm. Can you give an example of a country that would be thought of as a low to middle income country, but actually which did pretty well. 319 00:36:05,830 --> 00:36:10,180 So I guess Vietnam for a long period of time did very well. 320 00:36:13,220 --> 00:36:18,740 It's I would say a lot of. It's I don't want things changing covered so quickly. 321 00:36:19,110 --> 00:36:20,600 You ask me this question. 322 00:36:20,960 --> 00:36:28,970 I didn't know even a month ago I would have said southern African countries, some of them South Africa certainly had some big waves, 323 00:36:28,970 --> 00:36:33,140 but some of them have done very well, although, you know, the data is really, really patchy. 324 00:36:33,650 --> 00:36:41,150 And in some ways, Omicron has shone a light on just how patchy the data is because. 325 00:36:42,190 --> 00:36:48,299 Sort of speaking though in his rookie he genetics play rather than my own work but I saw 326 00:36:48,300 --> 00:36:54,300 a sort of a genetic tree a family tree of all the different variants and the deltas, 327 00:36:54,300 --> 00:37:01,380 the betas, the alphas, they all came off the same tree omega and it looks like it split in around February 2020. 328 00:37:02,470 --> 00:37:05,710 Which means one of various things. 329 00:37:07,730 --> 00:37:13,790 And so, you know, it could be in animals or that time or whatever. 330 00:37:13,790 --> 00:37:17,719 But one of the options, probably low probability. 331 00:37:17,720 --> 00:37:21,860 But who's to say is that there's other versions of the con patting out that 332 00:37:21,860 --> 00:37:26,860 family tree that we've never found that haven't blasted out like on the corners, 333 00:37:26,870 --> 00:37:30,050 but were there somewhere along the way and no one was looking. 334 00:37:31,470 --> 00:37:34,680 So it's hard to answer the question about foreign countries. 335 00:37:34,740 --> 00:37:42,450 I think one of your papers talked about how some countries had had four waves and some that were a couple who won the journey had one wave. 336 00:37:42,460 --> 00:37:46,470 But I don't know if that's still true is everybody had more than one wave now. 337 00:37:47,280 --> 00:37:51,420 Question Which country was that? Only had one wave. Um. 338 00:37:52,200 --> 00:37:58,799 I'd have to check. It's funny, like, even so, I've got a paper under review right now with The Lancet. 339 00:37:58,800 --> 00:37:59,879 And in that paper, 340 00:37:59,880 --> 00:38:08,210 we draw a quite a stark contrast between what we refer to as the two main strategies of mitigator countries and eliminate two countries. 341 00:38:08,220 --> 00:38:08,640 Oh yes. 342 00:38:09,090 --> 00:38:18,270 So mitigator countries being countries like the UK that have just tried to keep the number of cases kind of below what the health system could handle. 343 00:38:18,270 --> 00:38:22,730 Squash the sombrero. I really haven't had that here. 344 00:38:22,740 --> 00:38:28,140 That was that was that was the Boris. I think you flatten the curve, squash the sombrero. 345 00:38:28,150 --> 00:38:35,400 Yes. Oh. And then countries like Australia being eliminated as like they the moment is 346 00:38:35,400 --> 00:38:39,480 the slightest hint of a case or the possibility of community transmission. 347 00:38:39,750 --> 00:38:47,730 So make sure it doesn't go any further. We're not going to have this as a sort of a feature of of population biology. 348 00:38:49,740 --> 00:38:55,980 And, you know, what's happened in Australia and New Zealand in the period this paper's been under review, 349 00:38:56,340 --> 00:39:00,330 it looks like China's the only eliminator left strategically now. 350 00:39:01,020 --> 00:39:04,860 So such is the world of COVID research. 351 00:39:04,860 --> 00:39:12,230 It moves so quickly you just can't keep up. But the implication being that the elimination strategy doesn't work, you can't fight it that way. 352 00:39:12,780 --> 00:39:19,589 I don't want to agree with that because I think the elimination strategy worked 353 00:39:19,590 --> 00:39:26,780 for a very long time and was potentially very preferable to what we got um, 354 00:39:27,720 --> 00:39:33,480 on many grounds. And that particular paper is about mental health and we find it on the grounds of mental health apart from, 355 00:39:33,480 --> 00:39:38,190 and that's completely aside from the even more important issues of deaths. 356 00:39:38,190 --> 00:39:41,730 Right. And hospitalisations and long COVID and all the rest of it. 357 00:39:42,480 --> 00:39:55,070 Yeah, I, there's a lot to be said for eliminating. So you talked over you mentioned that you were having weekly conversations with. 358 00:39:56,630 --> 00:40:04,650 The Cabinet today was the Cabinet Office. That was what you said. And. Can you see evidence that your work has directly influenced policy? 359 00:40:07,420 --> 00:40:12,430 And I can see evidence that our work has really. 360 00:40:13,490 --> 00:40:22,040 Very deeply infiltrated the thinking of the Cabinet Office to the extent that they even talk in the language of indicators, 361 00:40:22,040 --> 00:40:25,740 which sounds a bit like a Chinese takeaway menu. 362 00:40:25,760 --> 00:40:33,049 So for example, if I say to one of our coders C7, they know it's about internal movement closures, 363 00:40:33,050 --> 00:40:41,420 right in the cabinet of his speak in this language as well. I can certainly say that their slides that they distribute in inform. 364 00:40:42,440 --> 00:40:47,270 Important people with reflect you know I work. 365 00:40:47,990 --> 00:40:55,309 I it's hard to draw a straight line between you know, our discussions on these calls with other academics of different stripes. 366 00:40:55,310 --> 00:41:01,160 So geneticists, epidemiologists and so on. And the policy outcomes coming from the government. 367 00:41:01,220 --> 00:41:08,460 Mm hmm. And is that something you expected to be happening at this stage in your career? 368 00:41:09,130 --> 00:41:19,800 I mean, how do you feel about that personally? It's the question of how do you feel about anything is not something you ever have time to drink. 369 00:41:21,100 --> 00:41:24,100 Said that to me. I said, we don't have time to think about it. 370 00:41:24,130 --> 00:41:26,830 It's announced that people. Yes, people have given me. 371 00:41:27,580 --> 00:41:36,850 Um, it's, it's strange to say, I think, you know, before that I would be quite in touch with how I felt about different things I was doing at work. 372 00:41:36,850 --> 00:41:39,950 So if I was giving a talk, you know, am I nervous? 373 00:41:39,970 --> 00:41:45,250 Like, if I prepared enough, everything goes out the window and you're moving at this pace. 374 00:41:45,910 --> 00:41:52,600 I've been used to now, you know, joining conferences and probably I didn't know how to this or not, but it all went fine. 375 00:41:53,560 --> 00:41:57,520 A number of times I've had to join calls where we're presenting in different 376 00:41:57,520 --> 00:42:01,209 venues and knowing that my colleagues are doing the 10 minutes before me and 377 00:42:01,210 --> 00:42:08,650 having to kind of write my bits while they're talking and then do my bit and jump off and do the next one or be in multiple meetings at the same time. 378 00:42:09,400 --> 00:42:13,500 It's been a, you know, a hamster wheel like nothing else. 379 00:42:14,590 --> 00:42:23,110 Hmm. That's that's the question I was getting to see what this might be with your journalist hat on rather than your public policy hat, but. 380 00:42:24,400 --> 00:42:32,860 I mean, one of the difficulties for policymakers has been that there hasn't always been a clear consensus coming from the biomedical community. 381 00:42:33,020 --> 00:42:38,979 Why do you think it's been difficult for biomedical scientists to come up with a clear, 382 00:42:38,980 --> 00:42:44,470 unified message of I suppose one obvious example is in the UK case this. 383 00:42:45,810 --> 00:42:50,160 Feeling in the first week or so that herd immunity was what they were going to go for. 384 00:42:50,460 --> 00:42:53,700 So completely throwing elimination out the window, just letting it run through, 385 00:42:53,940 --> 00:42:59,519 and then suddenly realising that that was going to lead to to overwhelming the hospitals. 386 00:42:59,520 --> 00:43:07,200 And we would have to backtrack on that. I don't think that many sensible by this were pushing for herd immunity. 387 00:43:07,650 --> 00:43:12,810 I think when I heard that phrase, you know, even with literally undergrads, 388 00:43:13,290 --> 00:43:19,590 that's 15 years old, having studied the concept of herd immunity and population models. 389 00:43:20,010 --> 00:43:26,790 You can see I remember doing the back of the envelope calculation and when they gave the note, 390 00:43:26,790 --> 00:43:32,670 I and I knew in about a minute that they were talking about more than 400,000 deaths. 391 00:43:32,910 --> 00:43:38,610 Right. And that's you know, that's with, you know, my basic decade undergrad. 392 00:43:39,840 --> 00:43:46,080 I think. Yeah, I, yeah, I think herd immunity was a terrible idea at the beginning. 393 00:43:46,500 --> 00:43:55,350 Um, particularly, I mean, for something that isn't nearly as dangerous as criminals, then there are arguments for it, but for something that can. 394 00:43:56,460 --> 00:44:01,350 Kill quite a lot of people. Just ethically unthinkable. 395 00:44:01,780 --> 00:44:09,330 Hmm. Um. I think, you know, the evidence around some things has changed as the data has come in. 396 00:44:09,870 --> 00:44:15,420 To its credit, the World Health Organisation is often criticised for this, changed its tune on masks. 397 00:44:16,240 --> 00:44:20,910 Um, and, but I think that is a very positive thing. 398 00:44:21,090 --> 00:44:24,920 It's not, it's evidence of egos. Right. 399 00:44:25,510 --> 00:44:33,299 Um, but it's, it's been a really tough job to know the evidence on things that, you know, 400 00:44:33,300 --> 00:44:40,770 I wrote and the Lancet called me very early on and one of the news editors and just said, 401 00:44:40,770 --> 00:44:45,630 Honey, you know, what do you think of some of the unanswered questions? And I said, Well. 402 00:44:47,140 --> 00:44:50,550 Again, this is very early on. It's like when can we get antibody tests? 403 00:44:50,560 --> 00:44:54,520 So these very, very difficult closure policies are coming in. 404 00:44:55,060 --> 00:45:01,510 We want to know, you know, when people are safe because it's only and we don't have any vaccines, 405 00:45:01,510 --> 00:45:05,379 you don't have any drugs, you know, but doctors have to be in the hospital. 406 00:45:05,380 --> 00:45:14,560 Right. So who amongst them is safe and who is? And I remember interviewing all the top people around that. 407 00:45:14,560 --> 00:45:22,980 And and at the time, you know, we thought the best evidence was size 1.0 thousand point zero. 408 00:45:23,000 --> 00:45:32,380 This one is. And that, you know, that that's 17, 18 years old and the frozen blood vials and the people who survived that are still immune to it. 409 00:45:32,950 --> 00:45:38,710 Right. So but so we thought early on that immunity would be very long lasting. 410 00:45:39,550 --> 00:45:42,460 But now we know that it's kind of not the case. 411 00:45:44,260 --> 00:45:51,460 We thought we didn't realise in the epidemiological modelling early on that you had asymptomatic transmission. 412 00:45:51,640 --> 00:45:58,150 Right. And that meant that an entirely different structure of models needed to be used. 413 00:45:59,620 --> 00:46:06,429 So I think the biomedical community has I mean, I can't imagine how stressful it is. 414 00:46:06,430 --> 00:46:13,600 I know what it's been like for us. But overall, I think they, you know, they have stepped up in a big way. 415 00:46:13,870 --> 00:46:18,849 And it's just been to too many unknowns. Yes, I do. 416 00:46:18,850 --> 00:46:28,389 70. Yeah. I think you know, a big change of public policy not just to do is disease t with climate change and you know, 417 00:46:28,390 --> 00:46:32,020 researchers in general have to communicate uncertainty to policymakers. 418 00:46:32,020 --> 00:46:34,360 And that's such a difficult job. 419 00:46:35,370 --> 00:46:43,380 When I worked for The Economist in the Science Desk, we used to have a bit of a joke that as science journalist we would report the average. 420 00:46:45,090 --> 00:46:49,350 The paper we'd be reporting from would have an average and a variant statistic. 421 00:46:50,010 --> 00:46:53,670 And then when we actually talk to policymakers and often MP said. 422 00:46:54,530 --> 00:46:57,830 Ask me at parties, you know, what should I do on this? You've written about it. 423 00:46:58,100 --> 00:47:02,710 They just want to know what direction, up or down, so much gets stripped away. 424 00:47:03,050 --> 00:47:06,510 And you try to have that conversation with them about variants. 425 00:47:06,920 --> 00:47:10,070 Right. And it's hard. It's really hard. 426 00:47:10,140 --> 00:47:13,510 Yeah, yeah. Yeah. I think it's. 427 00:47:14,700 --> 00:47:29,650 Yeah, yeah. So yes, is it possible at this stage to draw any conclusions about different policy approaches and their effect on this, 428 00:47:30,080 --> 00:47:35,090 either the spread of disease or pressure on health systems or the economic impacts? 429 00:47:35,300 --> 00:47:35,540 I mean, 430 00:47:35,540 --> 00:47:46,130 is it you've made some you've recently made some recommendations that have been taken up in the paper from the Global Preparedness Monitoring Board. 431 00:47:46,430 --> 00:47:54,470 Yeah. And so what were your recommendations? And so that paper is interesting because they asked us to. 432 00:47:55,950 --> 00:48:02,200 Really think about why a lot of policies that you know, there are many good policy recommendations. 433 00:48:02,340 --> 00:48:08,370 There's a sea of great PDFs out there. But implementation and pickup is very, very patchy. 434 00:48:08,910 --> 00:48:15,809 And so that conversation is really around trying to understand why so many of the recommendations 435 00:48:15,810 --> 00:48:20,820 that would have been very helpful if they had been implemented up to this point had not been. 436 00:48:22,040 --> 00:48:30,270 And so we sort of we looked into that, but really to reaffirm the ones that were most important and should be back on the table. 437 00:48:31,890 --> 00:48:34,560 So things like increased funding for the W.H.O., 438 00:48:35,070 --> 00:48:45,180 thinking about some sort of a international body or council to try and coordinate decision making, that kind of thing in terms of. 439 00:48:46,230 --> 00:48:52,040 Think about packages of policies at the things that we code, which is the question that very often get asked. 440 00:48:54,430 --> 00:49:01,700 I mean, it's. I usually say they work is a simple thing to say, right? 441 00:49:02,060 --> 00:49:08,870 But different countries use different combinations of different contexts, which is why it gets very complicated. 442 00:49:09,860 --> 00:49:19,110 On closure policies. I think the thing that I just you don't actually need to code to say you don't need all this data is. 443 00:49:19,410 --> 00:49:28,320 Well logically you know if people are in households and this you know people get COVID and passed on for about two weeks. 444 00:49:28,950 --> 00:49:33,060 So you have one person comes home with COVID. Imagine they just got it. 445 00:49:33,600 --> 00:49:38,940 So then they've got it for two weeks and and for two weeks. Then they spread it to someone else in the household. 446 00:49:39,180 --> 00:49:42,210 Who? Everyone else in the household, they can have it for two weeks. 447 00:49:43,200 --> 00:49:48,629 Right. So I think what you want to do with closure policies is to make sure that households 448 00:49:48,630 --> 00:49:52,890 aren't mixing and you want to keep them in place for long enough for that logical, 449 00:49:52,890 --> 00:49:57,750 roughly a month period to be done. And you want them to be effective. 450 00:49:59,660 --> 00:50:09,680 I think one of the things that we have not done well is we've had very long lockdowns that have been very costly to people in their everyday lives. 451 00:50:11,210 --> 00:50:15,380 And it's been done with poor messaging, and we don't code that. 452 00:50:15,740 --> 00:50:22,040 But there's a fascinating study from Italy. It compares two regions with the same policies. 453 00:50:22,550 --> 00:50:27,070 And in one region, they basically say, you know, we're doing this for as long as it takes. 454 00:50:27,290 --> 00:50:33,800 Keep going. And then another region, they set clear goals, you know, case rates, number of days. 455 00:50:34,250 --> 00:50:38,270 And in that second region, people adhere to the policy so much better. 456 00:50:39,220 --> 00:50:46,570 And it goes back to what I was talking about with the pandemic fatigue, with this logic of if you just tell someone to keep doing press ups, 457 00:50:46,720 --> 00:50:51,700 just keep going, motivated, they're going to be compared to counting them down, you know. 458 00:50:54,190 --> 00:51:02,020 So the the broad point I want to make and it's it's strange how you have to keep making this and I had to make it. 459 00:51:02,100 --> 00:51:11,020 A bunch of UK US legislators recently is the policies worked, they cut deaths, they cut cases, 460 00:51:11,380 --> 00:51:16,840 but they should be done intelligently because they are some of them are very costly for mental health, education and so on. 461 00:51:23,830 --> 00:51:33,850 So. Yes. I've been reading your papers and from what you've just said, I came to the conclusion that essentially you're arguing. 462 00:51:33,940 --> 00:51:39,640 Arguing for. A form of world government in this specific area. 463 00:51:40,610 --> 00:51:45,349 Which. Is itself a politically contentious ideas. 464 00:51:45,350 --> 00:51:49,280 And the idea that I mean, yes, we've got the World Health Organisation, 465 00:51:49,790 --> 00:51:56,389 but it depends on all the individual countries agreeing and so it tends to make policy will change 466 00:51:56,390 --> 00:52:03,150 policy rather slowly as you separate them off it to go from trying to to change its policy on masks and. 467 00:52:04,250 --> 00:52:07,010 But you seem to be suggesting there should be a short cut. 468 00:52:07,980 --> 00:52:15,299 That is a is a global organisation or globally trusted organisation that can that can move quite 469 00:52:15,300 --> 00:52:21,150 quickly and which people other people have to other countries or countries have to conform to. 470 00:52:21,630 --> 00:52:25,650 I'm not sure we call it a world government. I have a saying that slightly. 471 00:52:25,650 --> 00:52:39,770 But, you know, I think a you know, a strengthened, more nimble, a legitimate council would be would would be helpful. 472 00:52:40,440 --> 00:52:49,020 You know, more personally, I think and that's not in the report because it would have taken a lot more work than we had resources to do at a time. 473 00:52:49,470 --> 00:52:56,340 I think one of the issues, which is very hard to pull off in international relations in its current state, 474 00:52:56,970 --> 00:53:04,050 is that this has been essentially a regional pandemic, the way it's moved around the world. 475 00:53:04,140 --> 00:53:10,900 It was you know, it was in China and it was big in Europe and and in New York and then shifted to South America. 476 00:53:10,900 --> 00:53:14,760 And now you've got it in Africa. And you know, the. 477 00:53:15,790 --> 00:53:25,390 What the W.H.O. has done too much, given its budget of essentially a large U.S. hospital, is its annual budget has been going into this. 478 00:53:26,770 --> 00:53:30,430 And it should have pandemics. And it's really it's absolutely. 479 00:53:31,500 --> 00:53:35,760 But I think there is a case for strengthening regional bodies as well. 480 00:53:36,720 --> 00:53:43,590 But apart from getting the African Union and maybe to some extent ASEAN, nobody, China is not in that. 481 00:53:45,270 --> 00:53:56,320 It's not easy to pull off. Mm hmm. Yeah. But it's I mean, I think it's interesting because we've got other big problems globally. 482 00:53:56,890 --> 00:54:02,680 Climate change is the obvious one. And I wonder whether you think there are lessons from. 483 00:54:03,600 --> 00:54:13,110 The necessity of of global exchange of data in the pandemic that might actually help to shift the dial a bit on something like climate change. 484 00:54:14,220 --> 00:54:21,450 Well, I sincerely hope so. Actually, Tom Hale, my colleague, we all have this. 485 00:54:21,450 --> 00:54:27,390 He's building a and a tracker of policies for climate. 486 00:54:27,780 --> 00:54:39,809 It's very much the same model. But I think actually one of the things that I hope we get to do next year, so I have my fingers crossed. 487 00:54:39,810 --> 00:54:41,610 I don't want to jinx it by talking about it. 488 00:54:41,610 --> 00:54:52,079 Now that we get to run surveys in other countries regularly where we probe how linked people's thinking is, 489 00:54:52,080 --> 00:54:58,680 or how maybe their perspectives might have shifted on the value of international coordination. 490 00:55:02,270 --> 00:55:04,220 That's the you know. I hope so. 491 00:55:04,580 --> 00:55:15,680 I also think that many aspects of how the world is managed this pandemic have felt like a car crash in slow motion, particularly vaccine distribution. 492 00:55:16,730 --> 00:55:19,790 Um. Entirely predictable. 493 00:55:20,270 --> 00:55:27,840 Car crash information. And so because, because countries took a me first approach. 494 00:55:27,970 --> 00:55:33,270 Yeah. If you were to say okay well the world's structural inequalities are going to be how this ends up, 495 00:55:33,720 --> 00:55:40,330 you know, and essentially that is what has happened. So. 496 00:55:41,800 --> 00:55:46,210 Well, we'll see. In short, I mean, Glasgow wasn't. 497 00:55:48,480 --> 00:55:51,870 I heard the jamboree that we could have all hoped for. 498 00:55:51,870 --> 00:55:57,690 Kind of. Yeah. The emergence of a sense of global unity through a pandemic. 499 00:55:58,170 --> 00:56:01,590 We're emerging as the pandemic, but we might get there. 500 00:56:02,900 --> 00:56:15,130 I'd like to think if it was a possibility still. So I may be a little bit now to you more personally. 501 00:56:15,430 --> 00:56:23,230 So how did the provisions that were made in this country, lockdowns and so on, impact on what you personally were able to do? 502 00:56:24,250 --> 00:56:28,200 It was a. Wow. 503 00:56:28,290 --> 00:56:35,419 To start with that one. Um, well, I, you know, a lot of my colleagues have kids, 504 00:56:35,420 --> 00:56:43,220 so and home schooling happens what they were able to do personally in terms of their research, you know, I got blown out of the water. 505 00:56:44,000 --> 00:56:49,040 Um, and I don't know how they kept teaching. Um, I don't have kids. 506 00:56:50,240 --> 00:56:58,670 I. So I think it's probably worth saying the first lockdown, um, uh, so late March onwards, 507 00:56:58,700 --> 00:57:02,960 2020, I didn't have a meal with another human being for four months. 508 00:57:06,220 --> 00:57:11,170 It was actually fine. You know, if you say that, that's about to happen to you. 509 00:57:11,500 --> 00:57:16,570 I don't think anyone would particularly want to roll the dice on their psychology, just seeing how that one plays out. 510 00:57:17,170 --> 00:57:20,860 But I. I do think that. 511 00:57:22,180 --> 00:57:24,940 The track is in some ways being a kind of therapy. 512 00:57:25,720 --> 00:57:35,820 And in that I didn't really feel any sort of psychological impacts from that experience because it's every day, multiple times a day. 513 00:57:35,830 --> 00:57:46,330 I was seeing a huge group of people brought together by a sense of, you know, public service and. 514 00:57:47,540 --> 00:57:54,810 Um, also a sense of agency as well, when we all felt we didn't have any hope or it was being taken away from us. 515 00:57:54,840 --> 00:57:59,250 Yeah, actually. Everyday agency. Um. So. 516 00:57:59,640 --> 00:58:05,950 I have. Exhausted by this project, but it was being propped up by its owner personally. 517 00:58:07,910 --> 00:58:12,050 No, that's the idea. That's another question I had, which was whether. 518 00:58:12,680 --> 00:58:20,630 Yeah. Having something to do that felt worthwhile. Just a little bit to do with your own well-being? 519 00:58:21,290 --> 00:58:25,100 Definitely, yes. Yeah, definitely. Even when I wasn't sleeping very much. 520 00:58:26,510 --> 00:58:29,510 And did the the institution. 521 00:58:30,230 --> 00:58:32,050 Did you feel that that within that, of course, 522 00:58:32,150 --> 00:58:38,510 the institution that was in a sense that for colleagues who might be struggling, there was support there for them? 523 00:58:39,410 --> 00:58:39,950 Yes. 524 00:58:41,420 --> 00:58:52,400 I feel extraordinarily lucky to be part of the school because it's the youngest department in Oxford and it feels like a very, very modern workplace. 525 00:58:53,480 --> 00:59:00,620 We're constantly talking about, you know, different policies for different kinds of diversity. 526 00:59:01,700 --> 00:59:10,640 Different aspects of well-being. And those systems are there and they're prominent in our conversations before the pandemic. 527 00:59:11,980 --> 00:59:15,549 And so I think it's I wouldn't say it hasn't been a strain, 528 00:59:15,550 --> 00:59:19,720 particularly when we all had to step up and we had to move everything online for the teaching. 529 00:59:19,720 --> 00:59:26,590 And, you know, there's only so many staff and so many hours in the day, so people have felt the heat. 530 00:59:26,680 --> 00:59:30,670 But at the same time, I think this department has been incredible. 531 00:59:30,950 --> 00:59:35,590 Mm hmm. And the teaching, particularly because a lot of the students are from overseas, aren't they? 532 00:59:35,830 --> 00:59:43,660 Yeah. Did they were they here or did they stay in their home countries in a complete mixture? 533 00:59:44,780 --> 00:59:56,860 Um, you know, I, so when it came to Hillary 20, 21, I think a lot of the assumption was we were going to be teaching in person. 534 00:59:58,480 --> 01:00:03,760 And then as Christmas drew closer, you know, things looked much worse here. 535 01:00:04,420 --> 01:00:15,160 And then we had something like ten days before Hillary started to turn everything into online material and students couldn't come back. 536 01:00:16,120 --> 01:00:21,070 A lot of them. I remember taking a supervision with a student in Indonesia. 537 01:00:22,180 --> 01:00:28,270 I didn't know she was still in Indonesia and bless her, she'd stay up till three in the morning for the supervision, you know. 538 01:00:28,780 --> 01:00:33,310 But even that, she'd been locked in her little flats away from her family. 539 01:00:33,320 --> 01:00:40,750 She was in Jakarta. Her family were elsewhere. And, you know, just having a kind of. 540 01:00:42,150 --> 01:00:52,160 Um. She was just so down from being a tiny, sweaty little flat, trying to stay up for classes, moonlighting, basically. 541 01:00:53,230 --> 01:00:58,040 And yeah, it's been the human experience of Kobe just being. 542 01:00:59,410 --> 01:01:04,150 Off balance for everybody and completely different ways is how I feel about it. 543 01:01:05,700 --> 01:01:10,410 Yeah. And still no light at the end of the tunnel? 544 01:01:10,540 --> 01:01:19,220 Not much. Well, vaccines, I think, like keeping light. 545 01:01:19,760 --> 01:01:23,870 And in our mental health research, vaccines are really important for mental health. 546 01:01:23,930 --> 01:01:29,239 Oh, that's interesting. Would you like to expand on that? Well, it's unpublished, so. 547 01:01:29,240 --> 01:01:34,600 Oh, right. Okay. Yeah. Um, so I. 548 01:01:34,600 --> 01:01:42,770 I'm moving to a closed now. Has this experience, um, raised questions you'd be interested in exploring in the future? 549 01:01:44,010 --> 01:01:47,840 Oh, a little bit about a host of questions. Yes. Yeah, I think. 550 01:01:50,330 --> 01:01:59,240 So in terms of the the COVID work, I think there's a whole stack of research and thinking that needs to be done around, 551 01:01:59,690 --> 01:02:05,390 you know, what does preparedness really mean? If we were so bad, it's indicating it before.