 So Emily on behalf of the ECB and its executive board, it's a great pleasure to have you with us today and the floor is yours Great. So, um, so I am gonna I'm gonna talk a little bit today about What we've been learning about COVID and schools or what I've been learning over the last Really over the last year, but but more over the yeah over the last year. I guess it's been and And then I'm you know, I'm very happy to take questions I think that that we in the US are sort of Our schools are starting to reopen at this point. So we've but we've been through a long period with very extensive school school closures So just to kind of set the stage Although I think this is probably fairly familiar to to many of us I'm you know when we win things when the pandemic hit last year Basically all of the schools in the US and throughout almost all of Europe closed for in-person learning Sweden was an exception. It's not the only exception But you know largely everything shut down and I think that at that time There were very good reasons for that kind of extreme shutdown because if we think about in general how we The the way that respiratory viruses tend to work Schools do tend to be fairly significant sources of COVID of spread of general respiratory Viruses so if you think about something like the flu Or you know non respiratory things like like the chicken pox or the or the measles before before vaccines Those things are really highly spread in in schools So when there is an unknown pandemic of a respiratory Virus, it is very natural for an early for an early response to be like let's shut down schools because we think that kids are likely to Are likely to be vectors or be at risk and and so on As the pandemic even sort of actually even as early as maybe March or April really as we learned more from From from China And from some of the places that were hit early it became pretty clear that at a minimum Children were not a high-risk group And that's that's been true for this whole time And I think there's been debates about the their role in spread and so on but the idea that kids themselves are Both at least somewhat less likely to become infected With COVID and are particularly less likely to have serious consequences. That was true fairly early. That was true fairly early on And as that became clearer There was some move towards school reopening in Europe and in the UK and so you guys know this but There was there were some schools the UK reopened some of their schools in particular in the kind of late spring and summer Nothing in the US open through the end of the school year So for I mean that that's not like it's not like you literally could not find any open school But basically no significant amount of reopening occurred at all in the US throughout the entire sort of rest of that 2020 spring But then in the US as we got to kind of August and September even kind of some schools here open in some states in late July There was some school reopening it tended to be concentrated in what you know, we would call here the look the red states You know states that are sort of more traditionally On the Republican side of the of the political ledger So places like Georgia, Indiana, Florida Started to reopen and you know in a lot of those places they started to reopen kind of like what we think of as pretty regularly You know, so they kind of open for a full-time in-person learning And they did did it kind of did it look kind of pretty pretty normal But that but many places did not open and so when we got into the fall There was some more reopening and then there were some places that basically stayed closed and then there was Over the sort of fall there was kind of ebbing and flowing of opening and closing in in the US and some of that in Europe although I think you guys have been a bit more consistent with with opening in the US in the winter the sort of after like the January sort of after kind of the winter surge or maybe during it actually There was all of a sudden a much larger push for more reopening the Biden administration has pushed very hard for schools to reopen And so over the course of the spring we've seen much more kind of much more reopening at this point You know virtual learning rates are are much lower although some states still have fairly high fairly high rates And you know now we're starting to think already about the kind of 2021 2022 school year and what's going to happen What's going to happen? What's going to happen there? So in this talk I want to talk about sort of two things First I want to talk a little bit about kind of what we are learning about the consequences of school closures And there's a lot more that will be learned about that over a very long period of time But I just want to give a little bit of a flavor of what I think we are What I think we are we are getting so far and then I want to talk A bit more because this is the part that I've been working on really about sort of can schools be safe And to what extent are we learning about the the safety of Of schools and what makes them sort of safer in the US and and almost just how is the data kind of evolves there So let's start with the consequences of school closures. So um, so the The first thing I want to talk about is what happened when we closed schools abruptly last spring um, so in in the us When we when we monitor schooling most of the way we monitor isn't kind of like point-in-time large scale Like point-in-time high stakes testing. Oh, she actually didn't do much of last spring because of course it was supposed to happen And in the spring and the schools were closed and so some places tried to do it, but mostly they they didn't um, but So so we need to sort of look to other kinds of data to to see what was what was going on This is kind of this sort of graph, which I'll explain to you shows what I think is kind of my favorite Source of of data on this So this is a graph that uses data from an online learning platform called Zern, that's the name of the platform and Zern is a an online math like math platform that a lot of schools use it's linked up to the most common Curriculum the common core curriculum, which is what a lot of schools in the us use Use for their for their students. Um, and so zern is basically like a Like a way that instead of you teaching the math Zern teaches your kids stages the students in the class the the math through some like really nicely curated thoughtful videos It's actually a totally amazing resource if you're trying to homeschool here kids um But what they what they have what's sort of nice about these data Is that they have very detailed tracking they can see Basically, how much kids are online at every sort of moment they can see when when an individual kid logs on Um, and they can see how much they are they are doing and they measure Learning in something called badges And so what this graph is showing is that basically we have a set of classrooms That are using this online program before the shutdown So this is a part of their kind of normal classroom thing So if you sort of look at this at this space here the kind of january february part of this Uh, we're kind of seeing like, you know, there's a little bit of cyclicality in in how many badges kids are Are earning at any given week and some of that is just you know, winds of vacation weekend and so on But but you know, you're kind of seeing this truck along here And then this is when schools close and so what we see when schools close Is that there is a huge drop-off in these badges that students are are earning And that drop-off is is very much larger in school districts Or in schools that are in districts with lower with lower income families So if you sort of read this here, this says basically when we stop when kids stop going in person school There is something like a 50 to 70 percent Percent decrease in the the basically amount of math that they are doing So that kind of says like more or less they just kind of stop doing math There is also some decrease in these high income districts, but it's much smaller So this illustrates the what I think is the broad picture of what happened in the spring Which is that basically learning was in the there was no people stopped doing school much And schools were really poorly equipped even very fancy Pants private schools like my kid's school were just like very poorly equipped to move to some kind of online learning thing It was just like nobody had any idea what they were doing everybody was terrified locked in their houses with their kids It was not a good situation to be running an online school And that was particularly deeply felt by by lower income students And you know, I will say I think this is if anything kind of an understatement of how bad Of how bad this was because this is a an online platform This is a thing where literally they were already using the computer to do the math So you would think of all the things we could port over into online learning An online learning platform that we were already using seems like the best thing to port over You know forget about like I'm trying to teach some first grader to read by teaching them with some little Some little books that's going to be a lot harder to to port over and so if anything I think this probably understates the kind of aggregate losses In the fall students Students come back and now we have a little bit better Like a little bit better Testing a little bit more sort of sophisticated Testing so one version is something called a map test is a lot of these tests. This is math scores And this shows kind of by grade on the sample that's in this math test Sort of like what's happening in fiscal year 19 and fiscal year fiscal year 20. So Clearly, we are still seeing here reasonably large declines in kind of how students are doing sort of percentile ranks on these tests But just to be clear the scale here is 40 to 60 So, you know, these declines are I'm size one. I want to understate them, but they're not, you know, 70 percentage points And when we look at reading scores on something like this actually not affected much And I would say the kind of general picture of the learning losses in the fall Is a dialed down version of the of the picture in the spring So it is still the case that the learning losses are bigger in lower income school districts And that there are some learning losses But they are they are attenuated and I think that's in part because we actually learned a little bit In part it's attenuated because some kids are at school But in part it's attenuated because we've learned something about how to do this kind of Learning and that in fact, you know, you could do a slightly better job at it Even if it's not as good as being at at the school The bigger issue I in some ways Which bigger and poorly understood are all of the other sort of follow-on consequences That I don't think we know very much about yet that we are going to learn more about as we go forward So one one very big one is just mental health of kids There are a huge number of kids in the u.s. Who are not going to see an inside of the school classroom between march of 2020 and at We hope september of 2022 2021 so that's you know kids who basically have been in their houses Learning for I don't know six hours a day on zoom And it's not, you know, just 14 year olds. We're talking about seven year olds We're dying at seven year olds who have spent the entire year more than a year only learning on zoom Not really seeing other people in person the the mental health consequences of that I think we're starting to see some we're seeing increases in emergency department admissions for For students for for suicide for depression anxiety All these things are coming out But we don't have a great infrastructure to to measure those and so I think that's something we're gonna Be getting more of over time There are clearly also big impacts on demon labor force participation We saw this very sharply in the fall Where uh, you know in september there was a huge drop in in labor force participation about a million people of which about 800 that that was almost it was like 800,000 So it's we're seeing drops in labor force participation and those drops being sort of very concentrated in in in women and actually one of my Undergrads this semester wrote a paper looking at basically how those drops Have linked up to to schooling and showing indeed in places in which schooling was less in person Those declines in female labor force participation were bigger So not surprisingly it looks like the sort of if your kid is not at school Then you it is hard for you to be at a work, which I think would not surprise any of the parents on this call Then sort of there's a there's a kind of two other big pieces here, which is just there's a lot of Parts of the school learning environment that we are Um, that are not directly measured with things like test scores So kids dropping out particularly in you know among high school students kids who are delaying school entry Unclear how big a deal that is exiting school early totally abandoning public education So in the u.s. One of the patterns, um is that private schools have largely been open uh for in-person learning even as um as sort of What would be what we call public schools like sort of the The regular schools and most people are in have had not been open to we've seen places where people have basically just Um, just kind of given up on on public schooling and and gone to a private school or an archdiocese Catholic school That you know, those are of course disproportionately people with more resources If those people start abandoning the public school system That's really going to change the kind of power dynamics around a lot of these A lot of these these school districts and of course there's like a general exacerbation of of inequality across students and and schools All right So now I want to talk a little bit um in the sort of second part of this about uh the kind of school safety because we we sort of In a sense like I think I'm trying to make a case of like the benefit of having in-person school is really big In the u.s. We've been sort of fighting against The question of okay, we all agree that the benefits are large But if this is spreading covid to everybody, then we need to um, then we need to to sort of try When maybe we can't have this and and that like There's trade-off, right? Um, and so so when we think sort of think about school safety We kind of want to ask the question like does covet spread in schools and I said at the beginning I think there are reasons we might think yes Um, and that is because generally kids, uh spread respiratory illnesses Early data basically suggested no so we get kind of some evidence out of uh out of the the european data Um suggesting that there was relatively little spread when schools were open. It's sort of generally fairly reassuring Um, there, uh, there are sort of open Related open questions of what kinds of mitigation are important So should we just open schools like totally regular? Should we do should we be having kids undertaking? um different kinds of prevention and what would that what would that look like? um, and so I my sort of Way into this is I've been doing a lot of data collection. Uh around this in the us and I kind of got into this Uh, actually not as a researcher. So my main job is you know being an economist or a parenting author or some other thing um, but not being a person who collects data on on schools and I think in in the us We might have thought this would be the responsibility of um of the government of the we do have a federal government here Um, we have an organization called the center of disease control. We have an organization called the education department Both groups might have been um responsible for such for such data collection I mean, you know unfortunate issue was that um until you know recently the uh federal administration over here wasn't I'm really interested in science uh, and they They do not have any efforts to uh try to do anything that would be helpful It'll be helpful. And so what happened is in august school started to open they opened in georgia They opened in indiana and a lot of us looked out at this and we're like, oh Okay, well, it's not obvious. That's a great idea Like these this was opening right in the middle of like this sort of summer surge in these places And they were opening these schools and putting all the high school students in and nobody was wearing a mask and Everyone was like, oh my god. This is going to be awful But it was like, all right. Well now they did it like I hope that we're going to be able to learn from this And it became really clear that there was kind of no Data effort to to try to learn from this um Nobody was collecting any information And when we came to this we sort of came with kind of three pieces of of question Um, you know, one is just our school's like super spreaders, right? Are we going to see enormous outbreaks associated with uh associated with these high school Reopenings or these elementary school groups? Is this going to be the thing that like puts pushes everything? You know, is this the biogen conference, which was like our big super spread event in in boston Um, the second question was okay. Well, if there is some super spreading Are there some places that don't do it and you know, which mitigation factors are going to are going to matter Um, and then, you know, third kind of what where is the spread? Who is spreading it? Is it students staff like what what's if we're seeing spread? What's what's going on? And so we wanted to design some data that would at least get us a little bit at some of these some of these questions um, so So this is like a lot of the u.s. COVID response. It's a sort of cobbled together effort by just random people So I got together with something called the school superintendents association. So To be clear in the u.s. Um, there are school districts. They are very very autonomous Um, so you can sort of think about like there's a chicago There's a school district in chicago. That's like one school district But then there are like in texas. There's 1200 individual school districts And often these will be like A district that has, you know, two elementary schools, a middle school and a high school And so they're and they are they were basically responsible for making all of their own choices in general and also in this space of COVID So decisions about reopening were were basically made by these individual districts Which left these districts in a really crummy position Because now there's some school some superintendents some person who oversees this district You know their normal job is to is to kind of oversee two elementary schools a middle school and a high school Often they also have some other job if it's small that maybe they're the principal of one of the schools Sometimes they're like also the bus driver, you know These are like pretty like small places and now you're telling them Okay, you need to be responsible for whether we open schools in the era of COVID Like if you make it do it wrong people will die. Good luck Um, and I think that that was a really high stress situation And so part of the motivation for this sort of their this association involvement here Was that the school districts really wanted the answer to some of these questions I got some principal association. We got a data science team from qualtrics We have my team, which was just me, you know, it's some other people Um, and what we did was initially like super simple We just went out to school districts and we said, hey, can you just tell us like how many people do you have Are your school in person? If any, how many staff do you have in person? Are you requiring people to mask or you're requiring them to distance to do anything about ventilation? And then we came back to them every two weeks and we said how many COVID cases do you have? That's it. That was like our initial Basic plan. It was sort of very very simple and initially we just did it with a kind of opt-in sample of schools and And districts and then we ultimately added to this a bunch of data from states that had sort of consistent data over time And we sort of kept adding Over time but but unlike, you know research project our goal here was not To like wait till everything was perfect and then, you know, do it our goal was to be like literally We need to know some of this information right at this very moment And so we are just going to get this and we're going to post what we have And we're going to say what it is and we're just going to put it out so we can try to You know help speak to this conversation So, uh, so just to give you a sense of kind of where we are in terms of numbers, um, we started this effort Basically the sort of first time you posted data was like early september the mid to late maybe late september and we had about I want to say like 700 thousand In 700,000 Enrolled students of whom about half of them were were in person during this period Um, we added a few more, you know in the first part of september and then we started pulling in full states And so in the kind of for most of the school year Our data set has been about 12 million 12 million enrolled students of whom, you know, basically about half of them are Are are in person to give you a sense the sort of overall us K 12 populations about 56 million students. So, you know, we have some we don't have all Um, here is what so and then we we get these like infection rates And so here is what these look like and so a lot of different ways to to put these numbers together and talk about what they what they mean Um, but but here is the simplest ways you're just asked like what share of the students or staff Have coveted over a two week period. So they sort of given two weeks period. So if there's like So this is like our sort of top number here is like like 1.3 And so what this says is in this that's like the first two weeks of january in the first two weeks in january of a of a thousand staff We would expect on average to see Like what is it 13? Um, yeah 13 No, 13 So i'm off by motors of magnitude or it's a little early in the morning here We'd expect to see 13 13 infections Now just to be like crystal clear what we are collecting in these data Is information on the number of students or staff who are affiliated with the school who have covet Uh, not people who got coveted school. We do not have any information about who got coveted school There's like a flag a little bit of stuff at the end about this But in general what we were aiming to do was just to find out how many people Around your school have covet and the reason we went with that approach Partly because it was the only option But the other piece is if you if I sort of think about how we were motivated early on and what we really wanted to know A lot of our questions were just in this space of like It does everybody have covet, you know, is the school overrun with are we seeing these sort of big super spreader events? Like what are what are we seeing and I think that for that? This is kind of sufficient, right? So if you if you sort of saw Hundreds of people in a given school that would be a real, you know, that would be a real signal and that's something we'd show up We'd show up here so So this is and so this is kind of the pattern here is really mimicking what happened in the us over this period We're seeing a kind of increase. This is sort of this winter surge and then we're seeing now things are much Things are going much better It is also true in general that we see staff infection rates much higher than student infection rates That is reflecting very likely reflecting simply age patterns in Covid infections. Um, so kids are less likely to be affected than Then than adults the fact that we're now seeing these rates Kind of come come together is reflective of the fact that teachers in the us were prioritized for for vaccinations So as of kind of February a lot of teachers were already vaccinated and certainly by the current period You know a big share of of teachers have already been sort of fully vaccinated There are some states where we see pretty consistent data Or we see very consistent data over this whole period in particular new york, texas and massachusetts So we can sort of graph Graph this and and I actually should say I mean I can we can put it somewhere at somewhere in the chat All of the data for this is public and so there's like a public dashboard. You can poke around in it and see what's see what's going on Um, and so that was a big part of our goal was to try to actually make this sort of possible for people to to play around with One interesting thing that comes particularly out of the new york data is an ability to look some at um at The the kind of density of cases because you might have said look Okay, maybe overall case rates are low, but maybe we're seeing You know some concentrations of cases that look like there's a lot of spread in schools That's actually really not what we saw in these data. So this Says, you know, if we look at um, kind of This is new york at some particular point You know 70% of the school is in this in this bi-weekly period have no cases But then of the places that have cases a a large share of them are having kind of one or two Are having sort of small numbers of cases It really looked like those are very likely to have been acquired elsewhere to have been acquired kind of outside of The school and then there's just a person Affiliated with the school has who has coveted. It's not that we don't see any clusters So there's one but it's it's like like I can pick out a particular example So like there's one Religious high school in new york that has 15 as a cluster of 59 cases. Obviously something happened Um at that in that example, but there's very few of those It's not like that's something that's showing up every week That's something we see a couple of over the course of like the entire school year So finally a couple other things I wanted to I apologize that these are a little A little small But in the u.s There's been a lot of focus and I'm happy to talk more about this because I think this has been sort of very different from the european case But there's been a lot of focus on masking and distancing in schools Um, and in particular on the on the role of those things in lowering Lowering trans transmission rates And so here we kind of divide this sample I think most interesting is to look at at high community transmission, which is this Like the time periods in states locations where there's a lot of Transmission and and what we are seeing is that um in places where Masks are required. We are seeing particularly for staff somewhat lower Somewhat lower rates. So sort of things that are kind of looking like Uh, like the masks are protecting the staff from from infection A second very common mitigation in the u.s was requiring distancing This is a little tricky to think about because in fact if you have ever been to like my kids school Says that they do six feet of distancing and you know, I frequently receive Emails full of pictures where the kids are like wrestling with each other So obviously there's a difference between saying that your kids could be in principle be six feet of distance But in fact, uh, definitely are are not um, but when we look across those perhaps for that reason When we look across these groups, um for no distancing three feet six feet Actually, generally not seeing that that's mattering very much for for a case rate So I think the kind of distancing piece of this is looks like it's less less important than the than the masking in this in this period um, so uh, so final final slide is just to well not final one more thing Two more things to say but final data slide is just to uh to look at One of the things we would try to figure out sort of like How Does it look like schools are are higher risk than than other places and the answer is basically know that that kind of community rates Uh, and school rates are really tracking each other very um are really tracking each other very closely Um So there are a lot of limitations, uh to our uh to our data But I think actually rather than telling you about those of which there are many, um, let me sort of highlight, uh to um To some other data and that sort of and say something about about what we're sort of learning from from kind of more Almost more academic data. I guess I'd say that's come out over the last Over the last few months because one of the things I think our data added was that we were out very early So in a place when sort of people who were like trying to do Careful contact tracing data collection. They those papers kind of started to come out in december I think we were we were sort of early on in saying look, you know Our school super spreaders know and saying that in early october When you know, there was maybe more time for some for some opening But when better data came came out it does seem to support this sort of general picture So, uh, so one is uh, there's like a large data set from north carolina where they followed about 90 000 Students and staff in person school for nine weeks There were 773 People affiliated with the school who were kind of like had covet would acquire somehow got covet But when they sort of dialed down to like, okay all of those 773 to what extent were there school infections? They saw only 32 of them Um, and so that's a I don't know like a secondary attack rate of like 0.04 percent. It's like a very very small number um And importantly there and this is a very consistent pattern They did not see any transmission from students to staff. So when we are seeing infections in school It is largely between staff members Uh, when it was not between staff members, it was either staff members two students Or it was students among each other. Um, and again that last group is actually the the smallest Um, but we did see some we do see some, you know kiddos Uh affecting affecting each other. Uh, there's also some data from from wisconsin Which is showing um was also sort of showing very limited transmission We are in the u.s. We're very into sports youth sports And there has been an increasing amount of evidence suggesting that youth sports Are a source of some infections. So there's some cdc stuff about wrestling um And you know, I think what we learned from that is that like an indoor wrestling unmasked wrestling tournament is a good way To spread coven and I think if you sort of think about the mechanisms of coven and you think about how wrestling works It wouldn't be very surprising that a sport where you literally put your face next to another person's face Um, might be a good way to spread a respiratory uh respiratory virus. Um, so we're probably not having any wrestling tournaments um So what do I think are the kind of big takeaway takeaway lessons here? So I think one is that schools can be operated safely and then in fact this is not a high risk environment One of the most frustrating aspects of this whole discussion in the u.s Is that there have been many places where we have seen, you know bars and restaurants and gyms and You know indoor water parks open and not schools And it has I think you guys have had less of an issue with that But I think for for many parents in the u.s. It sort of felt like Oh, like are you kidding me like this is not only is this very important But it's also pretty safe and so relative to some of these other things It feels like our risk-benefit calculus has been really has been really off Um, I think there is some evidence that that masking may may matter particularly for for staff that may The importance of that may change as more of the staff are vaccinated youth sports are kind of a problem um community rates matter for school rates As we at least move into the fall I think that there's a lot of questions about kind of how can we make this school normal? I think there's a lot of fear. Um, it's a lot of fear Uh, and uh, and we're going to have to kind of dial that um dial dial that back. Um, but Uh, I think we we can we we can get there or I hope we can get there particularly as As adults are vaccinated a community rates have started to go to go down So I'm going to stop there and I'm excited to hear questions Great. Um, thanks so much. Emily. Um, so I have a one question on the consequences uh that you talked about in terms of the widening gap between private and public schools in the u.s Which of course is a long-standing issue as you said for inequality and all that Um, and you you you've talked about how this gap has increased further from a learning perspective And then later you spoke about action rates the question is whether There are also sizable differences in affection rates between these two different types of schools So is there something you learn about either the behavior of children Features in those two different types of schools and do you see it in the numbers? Yeah, so so it's a little bit of a tricky question to to sort of thing But I think that if you want to think about how to organize this data The way that we or the way like the easiest way to organize this data is to imagine that there is very little in the school transmission And that differences across schools in how much cover there is are mostly driven By the infection rates in the community and in the sort of groups of people who come to the school and so it is Is it true that we've seen I mean I was trying to think like in general It's I mean it probably rates are a bit lower in private schools, but But not in any, you know, very consistent not in any very consistent way And I think it really It is because you know the kinds of precautions everybody's kind of taken pretty similar Pretty similar precautions And there be and their infections are largely being driven by what's happening in their community in their in their communities outside Okay and another question is about the the infection rates that you showed in schools with this interesting difference between teachers and And the the pupils So I guess there's sort of two sets of questions one is sort of this data jumps out to to kind of Throw a lesson to say, you know protect the teachers stupid kind of no and It happened in the u.s. But very late in the game it seems so is this something that may I mean What's your view on that should have happened earlier? but The teacher needs your vaccinations Yeah, or some other way of protecting teachers. Yeah But indeed when the the other question is in terms of vaccination So it's very interesting to what you showed that after vaccination of teachers The two curves kind of converge so the teachers and the students for infection rates Um, so could we take away from this is the question that if we since for children it's low anyway That if we vaccinate children that may not it may not make such a big difference. I mean basically the biggest You know the biggest thing to do was to vaccinate the teachers. Is that one lesson you would take? Yes, okay, so let me answer both of those. So I think that on the first thing You know, I think we understood the the importance of of protecting the staff and that that was an important group To to work with and I think they bore in many places. The staff were prioritized for vaccinations over, you know It's like that that was a priority group in a lot of places in the In the u.s. There clearly is more we could have done in the fall And so I think many of us felt that we didn't do enough testing that that was kind of a big fall down Was that we uh, we should have been doing more like surveillance testing in the fall focused on on the staff And you know that was partly it was kind of expensive and partly, you know There was some resistance and so and so there clearly were things that we could have done better If we had under if we had approaches differently I think the issue was not so much that we didn't understand that it was a problem, but that it was You know that that it was we just like fell apart in a lot of different in a lot of different ways The other kind of interesting piece of this is that you know, a lot of the places A lot of the places that did open there like there sort of was a more general kind of I don't know COVID denial is just Pejorative but like many of the places that had the most aggressive openings were also places where people weren't necessarily taking COVID like other COVID things super seriously And so it was it was like there was a rare place like I the place I live So I live in Rhode Island. It's a very small state next Massachusetts And uh, and and we are kind of among the very few sort of democratic states Which did open a lot of most of the public schools and so there there, you know, I think there was a lot of effort to kind of Try to do the things that we would need to do to protect teachers and and so on I think there were other places where there was more like, yeah, everybody goes to work like you go to work This is your job go to work like wear a mask. It's um, but we could have done we could have done more There are a lot of discussions at the moment in the u.s. Around now the teachers are vaccinated like okay, like Like what's the big you know, what's the big deal? Like let's just like have it kind of regular um, and I I think that uh, that there's a lot of there's a lot of disagreement there part of what's happened is You know people have gotten The people are still worried about kids. There's a lot of you know, a really heated debate here around you know kids and vaccines and okay, those aren't going to be ready for like a while I think there are people who are like well who cares kids are a low risk group anyway Like let's just like act regular now that we have adults vaccinated isn't that what we were waiting for I think there are other people who say well, you know Even one kid in america gets coveted and and dies which of course could could happen Um, they don't won't have been worth it. And I think that you know, we're struggling a little bit here probably you are also with how to trade off these um How to appropriately deal with these risks given the amount of real fear that is sort of That has become the covet has been wrapped up in Yeah, I see now very clear. Um, so There are also two questions around The staff testing intensity numbers that you showed so one is a more technical question of whether This testing intensity is constant or whether it could otherwise be adjusted for if it's not constant And the other Related question is whether the data that you presented also covers the period with more aggressive mutations the variants Yeah, so right. So we don't so the testing thing is actually very um, so we we don't really Um, no they they sort of short answer is we don't actually know how much testing there is we know That this matters in the sense of like if you look at the new york data Um, we know that in the in the like winter new york is doing a ton of testing of teachers And their estimates suggest like teachers were 10 times as likely to be tested as the kind of General population. So of course that's going to kind of drive case rates Up some because there's testing there's testing more It would be better if everybody were tested if for covid all the time Like we all sort of all this like science people agree agree with that that hasn't happened We're sort of limited to to reporting out what we um What we have I think that you know, this is why we try to give these comparison rates We try to sort of talk about comparisons to community rates Or at least if the testing of this group is consistent with the testing and the rest of the community Then those rates are then those rates are comparable. Of course, if they're if they're not, you know, I mean Again, I kind of I kind of often fall back in this on saying look, there You know, there are a lot of really detailed things that we would that we would like and for which these data are not That well suited If your question is are we seeing huge outbreaks in school? I think our data are really well suited for that I think ultimately there are some other things about the data that will be helpful and kind of long term Long-term things but these kind of like Like people be like, what about lunch? Like my data is not well suited to tell you like how students should have lunch You know, we don't we don't have enough enough information there um On the variance variance, um, you know, we are still collecting data now So we're seeing the you know, we're seeing a data now when there are some places where there are more, um, you know There's there's kind of more variance You know people have sort of talked in People have talked about the idea that the variants are relatively more infectious for kids My sense of the latest uk data was that that's actually at least at least for b117. That's actually not really true It's just that they're generally more infectious and so of course they're relatively more infectious for everybody Either there's been a lot of rhetoric here around. Okay. Well, you know in this latest wave, particularly in michigan We're seeing more younger people You know relatively more younger people relatively more kids It's really unclear whether that has anything to do with the variance or whether it's just that old people are vaccinated And so the kind of people who are left susceptible are younger people and because the old people are vaccinated We kind of opened all kinds of stuff and then people started going out and doing stuff Not not particularly going to school But you know going out and going to bars and restaurants and Doing this and the other thing and then some people got got COVID So I think it's gonna it's hard The data is making it hard to separate out the role of variants from from everything else at least in the in the us Thank you Next question is do we have evidence that the consequences of closing schools can be overcome in the future? Most people seem to think that they will so I think there's there are Of some things for sure, which we cannot overcome some things which I think we are likely to be able to overcome And some things we don't know so I think on the first piece sort of things we can't overcome There are choices that were made in this year that will not be easily undone So there are students high school students who dropped out who aren't going to finish high school And they're not going to come back and finish it next year. Like we're not getting those people We're not getting those people back. Those are real long-term consequences that we will see I'd also put in that group, you know the category of kind of people who left their jobs Like parents who left their jobs who may of course return But we know that when you take a gap in in the in your resume, you're less likely to return You return in a different in a different way like we're going to see some some real losses there. I think those will not be made up Then then there's the more basic question of like look some kids didn't really learn the second grade Like, you know, are they going to be able to catch up and like learn the third grade? And they are kind of more optimistic because I think that we if we have that at least with appropriately used resources We could catch up a lot of that. You know, we know that one-on-one tutoring can be helpful for Can be helpful for kids if you think about a lot of these early stage things like Kids need to learn to read they need to learn to like add and subtract We kind of know how to catch people up on those when they're when they're behind it if we get enough resources in place I think we'll be able to do that The place where I think we're least sure is this piece of kind of like mental health sort of long term like socio-emotional like trauma basically trauma scarring of you know, this was not you know Even for people with a lot of resources and kids whose families have a lot of resources and have a lot of computers and have a parent home and whatever This has been really really challenging Um, you know, I think it's just not it's not the way that people want to be and I think people are seeing a lot of those kinds of um A lot of those kinds of consequences Um, we're just we're going to have to see how they play out over time Yeah, uh, actually the the previous speak in our series spoke about mental health and it was a pretty Alarming situation indeed. It's bad for the children and the parents and basically everyone so I think it's bad for people's relationship with their kid. I mean, you know, this is not the relationship that many people wanted to have With their um, you know with their kids yeah Now I know it is very striking how uh, also the what you said about It's not necessarily the math skills, but it's the social Interaction skills that was Could you say a little bit more about that? Yeah, I mean, I think, um I think it's sort of interesting because I see this partly because the sort of people I know the people I talk to the most you sort of I see this from kind of the the standpoint of I mean of the people who in some ways are the most privileged Which is like people who you know have a lot of resources We maybe have the time to take some time whatever with their kids and I think One of the things that that happened for many of those people and I sort of candidly for for me is that if we had this um this sort of idea that like If your kids are like we're like homeschooling our kids like they'll learn so much math You know and like we can like really like use this opportunity to like I don't know like not this opportunity But like people sort of thought like okay, my kids aren't my school doesn't push them enough in math and I'm gonna push them more and I mean one of the things people learn is like actually that Sometimes you can do that right like actually there are some good online platforms for learning math And it is possible for your kid to keep up in those in those things or maybe even just sort of move beyond but that That what was really really missing Is the part where you like interact with other kids and where you go to school and that piece of school is So much more important than I think many of us recognize And when we think about school as kind of okay, you go to school to learn to read Yeah, that's something people learn at school. Um, but you know, you you could teach your kid You know how to read you can probably teach your kid to read But the the thing you cannot deliver is the environment of like learning to interact with other kids And learning to help to problem solve and I and they think as people's kids have gone back to school They have realized that that they're kind of like they missed that right? They sort of missed a period of that And that that is a huge part of what we are doing with with education is teaching people to interact in a society And to work out problems that they have with other people because ultimately as adults Like that's actually a really important part of being a person and When people have asked tomorrow the question people sometimes ask is what do you think will be? What do you think are some long-term changes that we'll see, you know around sort of society around After this and or around schools rather I think we will see I think the two big things we're likely to see Are one a move to more individualized online instruction for some things So actually one really good learning here for the for the us is that there actually are some really good ways that you can Help differentiate kids and help kids sort of Who are behind ketchup and who are ahead, you know move beyond Using these kind of online online learning platforms. I think we will see much more of that But I think in exchange we will see a lot more focused time spent in school on some of these socio-emotional curricula or where they actually try to teach you How to problem solve in social situations where they try to sort of teach you the kind of like how to how to be a person In the same way you would try to teach like how to add And I I think we'll see a little bit of that shift so So then there's one question which is very us specific, I guess Which is the stark difference that you Showed in terms between again the public schools and the private schools That also makes with the socio-economic kind of background. So What from a european perspective seems a little bit strange is that there wasn't sort of a much stronger pushback to To keep schools open because at least the way we look at the us is that the social safety net is weaker Daycare facilities are not available for everyone You have very few vacation days. So if you are a working parent and the school is no longer there, how do you manage? um This was difficult here as many colleagues can and Echo but To us it's like I mean I guess sort of it's sort of like what's behind the numbers the data that you show Must have been awful Yeah, I mean I I think there's a there are There are a sort of two things that are that are probably very different. I don't know exactly how important they they are But I think that sort of they're they're kind of two I would highlight So first of all, I will say a lot of people are extremely angry about this So it's not like everyone was like, oh, this is a great system like everything's going great. Like people were very mad um, but I think the two things that have made it difficult to um make progress on this are One I think actually a lot of families um are we're scared So and I think that our our approach to our kids. Um, I was talking to a sort of european colleague about this and He said this and I I think it's kind of right all under empirical evidence. I think that then in the us we have a much um more like Safety oriented like we're like much more cautious about children, you know It's like there's a there's a rule like, you know kids under There are rules like, you know, you can't let a kid play outside by themselves Into like if they're under 12, right? So we have this sort of set of things which I think are a bit different than some of the norms in Europe around kind of child independence and I think reflect not not things about how much you like your kid but just like like Aspects of physical safety that the us sort of takes a different approach to than some other places So I think some of what happened is just there wasn't as much push to reopen as you might have thought Because people were afraid that their kids would uh would die even though that was like not realistic Or you know, not very really really really really tremendously unlikely There was just like a lot of a lot of fear And that's one piece But I think a much bigger pieces than in a lot of places the teachers unions opposed reopening. Um, and the kind of Uh, that that's just like that's like a factor in the us And if you look at kind of where school is open and where they didn't open a lot of that has to do with You know where the teachers unions were um were were powerful That interacts with the second with the first piece rather because I think some of what happened is teachers even when schools were going to start to open Teachers actually in some cases would say to their students, you know Or to parents like don't send your kid back because the district was trying to kill us Like the district is trying to kill the teachers. I think that's continued to be it's not all like It's not all like by any means all teachers And I think this has been a big issue within these unions that people do want to go back and some people don't but that's uh, The union piece of this is clearly a school factor That's very interesting. Yeah, that's of course, that's very different here indeed So a related question is how do low income families deal with the issue of lack of hardware? Uh to the homeschooling in the u.s. And this was an issue here in europe Almost all the districts so this is like the one thing that people did in the fall Is basically a huge amount of money was spent by districts to give every kid a laptop So we basically went to like sort of most of these four urban districts went to kind of one to one um, every student gets a laptop or an ipad um And they just delivered them to them and then actually a lot of places cities And municipalities set up wi-fi um Set up wi-fi like so now everybody had wi-fi So that was like one thing we did that was okay good um So a different question here is actually something I haven't talked about and I don't know if you Have done research on yourself or maybe you have seen research by others, but basically universities I mean you're yourself of course at the university and we we think of university students as adults Do we do we think of them as adults? Okay So, um, how much of what you have to say sort of basically applies you think also to the university almost, yeah So I think one of the things that has been so striking About these discussions is that like while we are simultaneously saying like, you know We're not seeing a lot of outbreaks in schools We are seeing like in the fall. There was like huge outbreaks in university. It's continued to be true, right? so I just often I will say like it is in some ways really Astonishing that we didn't see bigger outbreaks in high schools because if you like look at a map of pennsylvania You can see where cop like we're penn statements Penn state is a is like a status penn state college Like you can see that on the map of the covet map because it's like a bright red chunk Right, they just like have like hundreds and hundreds and hundreds and hundreds of cases And of course most of these are the students are fine Like there's a little risk group But we saw just huge outbreaks in in universities and the places that managed to kind of keep those under control So like my you know my university pretty we did pretty well, but we still had some, you know reasonably sized outbreaks particularly in the In in the winter and that was in the context where we tested every single person who was on campus Twice a week since august And so every day every two twice a week I go and get tested with like a pcr test and process and we're doing that for everybody and still we had some outbreaks And so, you know in some sense like the university environment in terms of I don't know I don't think we have a great understanding of why but some aspect of this has meant like these are just very high risk Um People environment something and and they're not this was so Zero correlation Yeah, I mean, I think in you know here in the u.s. When people come back to cut to a university in the fall They're a big every place is going to require vaccines for students Well, Emily, um, so I've run down all the questions and anyway, we are at the end of Timing Very good timing indeed. And so thank you so much for doing this I I hope we can share your slides and anyway videos available So, um, we're sure the best Good luck with the good luck with your schools Yes, same to you. Okay. Bye now