Thanks for reading! The best way to make sure that you don’t miss any of these conversations is to subscribe to The Good Fight on your favorite podcast app.
If you are already a paying subscriber to Persuasion or Yascha Mounk’s Substack, this will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!
And if you are having a problem setting up the full podcast feed on a third-party app, please email our podcast team at leonora.barclay@persuasion.community
Frances E. Lee is professor of politics and public affairs at Princeton University. In addition to In Covid’s Wake: How Our Politics Failed Us, she is author or coauthor most recently of The Limits of Party: Congress and Lawmaking in a Polarized Era and Insecure Majorities: Congress and the Perpetual Campaign.
Stephen Macedo is the Laurance S. Rockefeller Professor of Politics and the University Center for Human Values at Princeton University. His books, in addition to In Covid’s Wake: How Our Politics Failed Us, include Just Married: Same-Sex Couples, Monogamy, and the Future of Marriage, and Diversity and Distrust: Civic Education in a Multicultural Democracy.
In this week’s conversation, Yascha Mounk, Frances Lee, and Stephen Macedo discuss school closures during COVID, why Republicans and Democrats reacted differently to the pandemic, why institutions failed, and why as a consequence institutions lost the public’s trust.
This transcript has been condensed and lightly edited for clarity.
Yascha Mounk: You’ve written a book that looks in quite thorough empirical detail at our response to the COVID pandemic. Take us back, if you will, to 2019. At that point, if somebody had said we’re about to have a big global pandemic of a respiratory virus, what was the playbook supposed to be? What was the response that the CDC and other institutions had planned for years and decades supposed to look like?
Stephen Macedo: They would have found that there were decades of pre-COVID pandemic plans in 2019. In fact, they would have found a fresh one from Johns Hopkins University over the summer of 2019, surveying and asking questions about the whole range of non-pharmaceutical interventions that were considered a few months later, and finding that there was very weak evidence to support the effectiveness of all of these non-pharmaceutical measures that would soon become familiar, including contact tracing, masking, social distancing measures of various sorts, school closures, business closures, and so on. Indeed, if they had waited a few months, they would have found that in November 2019, the World Health Organization would have issued yet another planning document surveying these non-pharmaceutical interventions in much the same way. And they would have found there too that the World Health Organization found that the evidence to support the effectiveness of all of these measures was quite weak.
In fact, they would have read that the World Health Organization said that there were some things that were not recommended under any circumstances, including contact tracing, border closures, quarantine of exposed individuals. They would also have found emphatic statements that these measures would be costly, that they would impose costs, especially school closures, on children, and the less well-off. If they had looked further back, to the George W. Bush administration, they would have found debates with mathematical modelers who were more optimistic about these measures based on their models—which involved a lot of speculation about behavioral changes—and they would have found that these modelers were much more hopeful about targeted layered interventions of various sorts. If we could just pile them all on top of each other, then maybe they would work.
But they would have found in the Institute of Medicine a letter reporting on the empirical evidence behind those optimistic accounts, warning that the evidence for these measures was not good, the costs would be high, and finally that political leaders would be tempted to implement these measures regardless, in order to show that in a crisis they were in charge and that they had things under control. So they really would have found quite a comprehensive assessment of both the effectiveness of these measures, their costs, and the risks that there would be political temptations to engage in these tough measures to show that political leaders were in charge. So an absence of evidence, an absence of good evidence for the effectiveness of measures that would soon be implemented on a large scale.
Mounk: Why did that all go out of the window so quickly? There was a playbook that had been researched and debated and prepared for such a long time. You then have obviously this very scary situation where this virus is actually starting to spread and you do see emergency rooms being overloaded in northern Italy, in New York and so on. It's understandable that this is a very scary situation for policymakers to try and orient through. You're in the fog of pandemic war, but all of the preparation seemingly gets ignored at that point and all of the measures that these plans had said wouldn’t be effective became the conventional wisdom in a period of a few weeks.
Frances Lee: I'd say there are two critical factors that induced that shift towards much greater optimism about the effectiveness of non-pharmaceutical interventions. First was the lockdown in Wuhan and the immediate fall in cases and deaths from the disease after the lockdown had been implemented. There was a World Health Organization delegation that was sent to Wuhan for about a week. There were a couple of Americans on it. The report that's produced out of that trip endorses the Chinese approach fulsomely and recommends it to the whole world. There isn't any skepticism raised about the effectiveness of the measures in that report. It's sort of strange to go back and look at it now, having gone through the pandemic, that at no point does the report address how we already knew that pandemics come in waves and that one would have expected there to have been a decline. That's just how the pandemics unfold and every place has experienced multiple waves of COVID. But at any rate, it was seen to have been effective.
Second then was the lockdown in Italy, which was the first nationwide lockdown and that set an example for other democracies. The measures were popular. There was a clamor for the imposition of these measures. When they were imposed in the United States in mid-March, 87% of Americans supported them, including large majorities of both parties. Then, in addition to those developments cascading from the Wuhan example, there was another important report produced by Imperial College London by Neil Ferguson, who had been a longtime collaborator of the mathematical modelers that Steve mentioned, dating back to the George W. Bush years—in other words, one of those voices that had long been optimistic about the use of non-pharmaceutical interventions. That report laid out very clearly for policymakers that this was going to be a devastating disease. In fact, it projected that in the United States there would be 2.2 million COVID deaths by August of 2020. It also provided differing projections if social distancing measures were implemented, which we could expect to significantly reduce those tallies. I think laying that out for policymakers in a stark way made it very difficult for them to resist the push for broad-ranging closures.
Mounk: One of the very interesting things that happened in the space of a few short weeks in 2020 is that the meaning of these NPIs, these non-pharmaceutical interventions, and in particular of social distancing, changed in two ways, very quickly, I think. I wrote an article called “Cancel Everything” for The Atlantic on March 10, 2020, which went kind of hyper-viral. I think that had some actual impact. When I was publishing that article, there were still giant indoor concerts planned for the next weekend, there were still sporting events indoors and outdoors going ahead, and the Biden campaign was still planning giant rallies for the following days. When I wrote this article, what I was trying to say is, let's avoid having these events which are not fundamentally important while we figure out what's going on with this virus and what's going on with this pandemic. We shouldn't be creating these potential mass spreading events just in order to have a campaign rally.
The other thing is that the slogan of those times was: flatten the curve. We were seeing that health systems had become overwhelmed in Wuhan, that they were being overwhelmed in Northern Italy. I have a good friend who's an emergency room doctor in Italy who was in touch with a lot of his colleagues at those hospitals. The goal here was simply to make sure that we don't have such a surge in cases so early on that people who could easily be treated with relatively straightforward medical interventions are going to die because no doctor is able to go see them because the ICU is so overloaded.
Now, very quickly, the idea of let's not have concerts indoors turned into let's close down schools and let's have government regulations tell people you're not allowed to leave your house unless you have an excuse and so on and so forth. The idea of flatten the curve turned into make people stay at home indefinitely. That is a really striking development that happened within a span of a few weeks in March 2020. How did that come about?
Lee: One thing I would say on that is that the recommendations not to hold large-scale events, that sounds more like the policies used in Sweden. That basically the canceling of large-scale events, but keeping schools open, sort of trying to find a balance with the aim of flattening the curve, as you describe, that was more consistent with the pre-COVID pandemic planning. Of course, that was quickly treated as a departure from the rest of the world—that Sweden was taking the experimental approach was how it was covered. When it was really the rest of the world that was adopting completely new approaches, really inspired by the extraordinarily stringent lockdowns that had been employed in Wuhan and in Italy.
Macedo: The message quickly over the summer became that we can only reopen safely when there's sufficient contact tracing, on a massive scale that we didn't have the capacity for, so that virus spread could be reduced to a few hotspots that could be detected somehow and then suppressed. As Frances said, we seemed to move from this mitigation strategy aimed at flattening the curve to a kind of suppression strategy aimed at reopening only once it was safe and keeping mandatory measures in place until that time.
Lee: It's also worth noting on that contact tracing point: that was one of the measures not recommended under any circumstances for a respiratory pandemic, because it was understood that contact tracing could not be implemented at the speed necessary to successfully wrestle this type of virus to the ground. It spreads too quickly. Contact tracing cannot catch up. Even when states hired thousands of contact tracers, they were overwhelmed by these efforts. So that was understood not to be a fruitful way to approach a pandemic of this kind. It was never attempted in Sweden. So it's quite interesting how this also became universally embraced among policy leaders in the United States, without confronting the reasons why previous pandemic planners had rejected this approach for a respiratory pandemic.
Macedo: It's also worth pointing out that in March 2020—so shortly after the time you mentioned—Yascha, when people did start closing down and after China had done so, there was a lot of pushback initially for a brief period of time in op-eds and major newspapers like The New York Times and The Washington Post, from a variety of very well-known public health officials—David Katz at Yale, Justin Lessler at Johns Hopkins, Tom Frieden, former CDC director, and Michael Osterholm at the University of Minnesota—who said things like, the Chinese must not be getting very good advice. This is not something that has ever been successful before. Anthony Fauci said that. Michael Osterholm made the point in a couple of different op-eds that the virus is spreading in society, and there was very little that could be done to stop it, that the best strategy would probably be focusing protection on the vulnerable and allowing most people to go about their business, more or less as normal with some voluntary measures in place. And that those who were relatively healthy, who had low risk, would thereby contribute to herd immunity by this happening while we protected the vulnerable.
So there was some genuine realism in mid to late March, in a number of places, and a kind of acceptance that the virus that was already spreading could not be slowed. But then that seemed to disappear by April and May, when we no longer heard that kind of dissent being welcomed.
Mounk: One of the interesting things that I think it's easy to forget in retrospect is how quickly the debate oscillated in the early days of the pandemic and how quickly the political fronts were changing. I mean, you go back to January and February of 2020 and the people who are warning about the pandemic tend to be more on the political right—Tom Cotton was going on quite early on about the threat of the virus—and it tended to be Democrats who were assuaging. I think Nancy Pelosi in Chinatown in San Francisco towards the end of February 2020 basically said, if you're worried about this pandemic, you're really subject to misinformation and it might be motivated by an irrational fear of Chinese people, et cetera. You really shouldn't worry about it. Then, of course, later in the pandemic when Donald Trump started to worry about the damage it might do to the American economy and perhaps in part his chances of reelection, it became the left who said, no, we’re the party that really wants to continue to social distance and lockdown and so on and so forth.
The same is obviously true of masks. We now remember masks as being kind of a cause célèbre of the political left and of Anthony Fauci and it's the right that are anti-maskers that didn’t believe the public health guidance, et cetera. But you go back to February and early March of 2020 and those political fronts were less clear, while Fauci was saying, look, these masks work for trained professionals in hospitals for doctors and nurses. They really don't work for anybody else, in part because at that time, personal protective equipment was so scarce, but many doctors and nurses didn't have access to it.
However you feel about what I believe was a kind of “noble lie,” as they thought of it, they seem to have thought, the only way we can make sure that these doctors and nurses have access to this PPE is to discourage the general public from trying to buy up these stocks, so we're really going to lean into the evidence that masks don't work. Now, there's an additional irony here, which is that early on, they say that the masks don't work in what seems to me like a political communications exercise. Then it becomes a very polarized issue, where part of being a good person who believes in the institutions and who believes in public health and certainly in more left-leaning milieus has to very very thoroughly wear masks—which I did for much of the pandemic—but in retrospect it's not clear that masks worked in the first place, so tell us a little bit about that debate.
Macedo: Let me just say that, in addition to what you just said, Dr. Fauci was asked in 2019—with no sense of COVID on the horizon—by an interviewer, what should we do in the event of another pandemic? The interviewer said, should I wear a mask? Dr. Fauci immediately interrupted and said, no, no, no, no. Avoid the paranoid stuff. Just get a good night's sleep. Don't drink too much. Get some exercise. Do the normal, healthy things. So before COVID, he was not confident about masks working. And indeed, that would be entirely consistent with those pre-COVID pandemic plans that we talked about before.
Mounk: Steve, so just to pick you up on that point, because there's been a good bit of debate about that and I'm not quite sure what to think. So, when Fauci said in February 2020, don't go out and buy a mask, they're not going to help you. Do you believe that he was truthfully stating his opinion at the time or do you think he was to some extent knowingly misleading the public?
Macedo: We don't really know what was in his mind. The fact is that that was correct and it was consistent with the studies of masks, which found that masking was not effective at controlling respiratory viruses. When he has reflected on this more recently, just in the last few months, when asked whether masks worked, he said, yes, of course masks work. But then he said, if you wore an N95 mask, these very tightly fitting respirators that medical professionals might wear, very tightly fitted, breathed through only the N95 and did so very, very conscientiously, then that would work. But that wasn't what was being recommended when cloth masks and other sorts of things were being recommended in 2020. So I don't know what he had in mind or when he was being sincere and when not. There was a lot of incredible flip-flopping around mask wearing.
Lee: I'll just say that when Dr. Fauci did not recommend masks in February of 2020, what he was saying was consistent with the pre-pandemic plans, that they were skeptical of masks. They often never even mentioned masks. The U.S. plan didn't mention masks, for example. Others would mention it and say, well, it's plausible, but we lack evidence. And there had been studies as Steve mentioned by the Cochrane Library, which is a highly regarded institution that takes stock of scientific evidence with regard to many medical interventions. So I think what Dr. Fauci was saying in February was more or less this conventional wisdom. I also don't think that the “noble lie” explanation works very well, because cloth masks were what the CDC recommended from April that year. There was never a shortage of those for purposes of instructing people on how to protect the community or themselves. So I think it was a desire to give people something that would reassure them. And they thought, well, maybe it'll help, but there just wasn't an evidence base that they could rest such recommendations on, much less mandates.
Mounk: That's an interesting aspect of all of this, thinking about it in a more sociological, perhaps even anthropological vein, that it feels like all of the deeply emotionally charged debates we ended up having over how to behave during the pandemic perhaps had the function of making us feel like we're in control. Like the incandescent anger at somebody who let their mask hang down below their nose and mouth where it didn't really achieve anything. In retrospect, the anger at that was in part the desire to feel like, if I do the right thing, I'm going to get through this scary period. It was actually a kind of misdirected form of self-reassurance.
Macedo: And showing to others that you care, that you would also prevent other people from contracting whatever you might have. So I do think it became a symbol of concern about others. Once Donald Trump stopped wearing them properly, and famously took one off on the balcony of the White House, then it became a partisan symbol as well. And as Frances said, in those reports it does describe some of these measures as mechanistically plausible, when thinking about droplets and so on. For an aerosolized virus, there was no evidence of effectiveness and that had been longstanding. But nevertheless, it did become a very powerful symbol and I think what you're saying is correct.
Lee: One of the unfortunate aspects of the pandemic response is the way in which these recommendations drove deep divides into society, dividing friends from friends and family members from other family members around measures, when there was never a good evidence base on which they should have been recommended so strongly or mandated. The same thing, of course, proved to be the case with the vaccine mandates, where you're going out and telling people in the summer of 2021 that if you get this vaccine, you'll become a dead end to the virus, that you can protect your loved ones if you get this vaccine. And then everybody discovered in short order that it didn't stop you from catching COVID and from transmitting COVID. The CDC even did a report in late summer 2021 after an outbreak in Provincetown, documenting that vaccinated people were spreading the virus and contracting it. Yet they still went forward with those vaccine mandates in the Biden administration.
This was very divisive among people in society. We already suffered from deep polarization, and then just to layer on additional divides without a good evidence base—this is an unfortunate feature of how the pandemic unfolded. I don't think it was an intentional set of actions that leaders took, but it nevertheless shows that taking measures that you sort of hope will help, but don't have a strong evidence base for believing that it will help, can be reckless. You can wind up doing more harm to society than good.
Mounk: It's interesting how many different controversial elements there are of this broader controversial topic of a pandemic. We've talked a little bit about social distancing. We've talked a little bit about masks, and now a third big topic, vaccines. Obviously we’ll get to others, including school closures, perhaps the origin of the virus and so on. When it comes to vaccines, it seems to me that this is the one area in which the technocracy, the experts, and the science triumphed. Because we did manage to get these incredible vaccines, which significantly reduce the risk of having adverse consequences from COVID for people, particularly those who are more vulnerable, or who are older and so on. We managed to get out of this pandemic much faster than we would otherwise have done because these vaccines became widely available and widely distributed so quickly. So this is the one point in which we might say that actually humanity triumphed.
But as you're pointing out, vaccines also then became this hyper-polarized political issue in part because if these vaccines had, as initially advertised, stopped your vulnerability to the virus and also stopped transmission to the virus, then the policy would have been very different. What turned out to be the case is that it gave you quite good protection against adverse consequences for yourself, but not so good protection against transmitting it to others. That meant two things. First of all, that meant the justification for requiring people to get vaccinated was much lower. If it were true that my getting this vaccine would stop other vulnerable people in the community from likely contracting COVID, the case for requiring me to do so would have been stronger. Perhaps it would not have been strong enough to allow the state to impose that kind of treatment on people—we can have a debate about that—but it would certainly have been much more plausible than it was.
Then the second consequence is that it sort of turned this into a classic question of liberal political theory about the Nanny State. If what's involved really is reducing your likelihood of adverse consequences for you, do we have the right to make you do this? The classical answer from John Stuart Mill on down would be, of course not. But the answer that governments took, and particularly the then left-leaning government in the United States under Joe Biden took, was, yes, we are going to do that. So how should we think about this, both in terms of our willingness to recognize this empirical fact that this life-saving vaccine does not, in fact, stop transmission, even though that was originally advertised, but also that normative debate was kind of implicit in this grappling with how to deal with a vaccine that has those particular properties?
Macedo: Well, let's just say a couple of things about that. I basically agree with everything you just said. The self-protection point is not quite as simple. We require people to wear motorcycle helmets in many states in the United States and the main concern there is people injuring themselves, but also that society bears a cost from medical expenses and so on because we're not going to allow people just to bleed in the street. So it's not entirely self-regarding even as far as that goes.
But I think you're right. The vaccines were a game changer. We agree with that. They were very important, especially for the vulnerable parts of the population. There was some misleading messaging around vaccination and natural immunity. There was not full transparency about the side effects, which were real for a small number of people, and there should have been more openness about that. The other thing is that, later on, recommendations on boosters in the United States went far beyond those in Europe. We've been recommending boosters here for people over the age of six months when in Europe they were being recommended for people over the age of 65 and the evidence base for those recommendations—which are very extreme by global standards—is not well established.
The other thing around this that's related is that, prior to the vaccines, the World Health Organization and other agencies were denying the reality of natural immunity from getting the virus and recovering. The World Health Organization actually changed the definition of herd immunity on its website over the summer of 2020. It had said what was common sense among epidemiologists, which is that herd immunity can come from people getting the virus and recovering or from being vaccinated. They changed it over the summer of 2020 to say it just comes from vaccination. Of course, they were called out on that and then they changed it back. There were problematic aspects too, and whether, when there were vaccine requirements—which may not have been well justified, as you suggested—having had the virus should have counted. Now that may be a little harder to verify, but there should not have been this misleading messaging. I think with the World Health Organization and other agencies, they were trying to encourage people to not get the virus, to wait for the vaccine, which was arguably absolutely the right thing to do, but they shouldn't have misled people about it. As you say, this was another “noble lie,” but in any case, people have felt misled about various aspects of this. But we're not vaccine skeptics, and Frances’ chapters show that the vaccine made a big difference.
Lee: There's a strong relationship between vaccine uptake across the states and mortality from COVID in the post-vaccine period. So at the end of the pandemic, in January 2023, which is when our data leave off, there's a marked difference between Democratic and Republican-leaning states in terms of their overall COVID mortality, in which Republican states—which had lower vaccine uptake—do worse. In fact, Democratic states wind up doing quite markedly better by the end of the time series. But all of that difference between Republican and Democratic leaning states emerges in the post-vaccine period.
If we just look at the period before the vaccines rolled out—the initial vaccines were administered to nursing home residents that got underway in December 2020—prior to that, there was a lot of variation in how the different states in the United States approached the pandemic. Yet at the end of that time period, there's not a notable difference between Republican and Democratic-leaning states in COVID mortality per capita.
Mounk: So the anti-vaccine rhetoric really does seem to have had a very, adverse consequence.
Macedo: In the Republican states, yes—lower vaccine uptake, higher mortality once vaccines are available.
Mounk: Tell me about the debate over one particular element of social distancing policies which probably had one of the most lasting socioeconomic impacts, which is school closures. How was the decision taken originally to close down schools? Was the intention originally to keep them closed for a long time or was it similar to this logic of flattening the curve that we're to do this for a few weeks to figure out what's going on, then we're going to reopen? Why is it that in some parts of the United States in particular, schools ended up being closed down for such a long period of time, which again was quite abnormal from an international perspective?
Lee: It's one of the ironies of the pandemic that schools were the first to close and the last to reopen out of our major social institutions. The school closures got started even before the coronavirus task force recommended closures. On March 16, there was a press conference where governors were instructed to close schools and other businesses where people congregate, both indoors and outdoors. Many governors acted to close schools after that. Schools had already started closing before that, beginning on March 5, so there was a cascade underway. Now, for those of us who had any association with universities or educational institutions during that time, initially we were told that this was a short-term closure, that this would be a couple of weeks and then we would come back. Those initial closures of schools were also announced on that kind of basis.
It was only later on in the spring when governors then announced that schools would not be reopening at all in the 2019-2020 school year. Wyoming and Montana saw some schools reopen in spring of 2020, but aside from that, all the public schools stayed closed, serving 50 million young people through the entirety of the balance of 2019-2020 school year. Now, schools began to reopen in large swaths of the United States in the fall of 2020. So across the South, across the Intermountain West and in the Plains. As we look at it in the book, the best predictor of whether schools would stay closed or reopen is the partisan lean of the jurisdiction. It’s more important than anything else. It's more important than any epidemiological facts about what's happening in any particular location. And it's more important than teacher union strength, though of course that interest group politics also plays a role. But it's primarily a story of partisanship. Half of public schools in the United States would still be closed in March of 2021. In the bluest jurisdictions, schools would not reopen in the 2020-2021 school year. In many cases you're looking at schools not reopening until the fall of 2021. So this was extraordinary and much longer than other advanced democracies. The length of the US school closures looks more like the developing world and looks more like Latin America than it does to our European peers.
Mounk: What's the evidence both about how rational that was at the time and what kind of impact that had in the long run? To play devil's advocate for a moment, I can understand why people might say, look, the most important thing is to protect young children. The worst thing that could happen is that a lot of young people die because of the pandemic. Of course, at least very early on, it was hard to know exactly how differential the impact of COVID might be on different age groups. Now, my understanding is also that older people were at greater risk of dying from this virus, which was not true of all viruses. In the Spanish influenza epidemic in 1918-1919, it was actually young adults who were particularly strongly affected. So we got very lucky in that respect in the COVID pandemic. But it's not just that older people were more affected, that they were orders of magnitude more affected than young people, and that really there were very, very few deaths among young people, particularly those without very serious pre-existing conditions. But at what point could we confidently say this? Are these mistakes that are obvious with the benefit of hindsight, or should they have been obvious at the time?
Secondly, what's your best estimate as to how much long-term learning loss and other adverse social impacts those school closures had? Was there a very high price to be paid for that in a more permanent way or is that going to be made up in the years of in-person schooling that those kids are hopefully experiencing now?
Lee: Let me just say on the age gradient of the disease, that was known very early. By late January it was known, because it was clear who was dying in Wuhan, and then it was clear who was dying in Italy. So there were already good indications that young people were not showing up at hospitals with severe outcomes relating to COVID. That information was there. Sweden did not close schools at all. In fact, there was a publication with Swedish data pointing out that Swedish kids in schools still open at full capacity, with no social distancing or masking, had not had worse outcomes compared to young people in places where schools were closed. And teachers had not suffered worse outcomes than other professionals. So they had that evidence. Then the EU schools began to open in summer of 2020, and EU ministers had a couple of meetings in the summer of 2020. At both of those meetings it was announced that when schools had reopened there was no evidence of increased community spread. So there was no evidence that young people were suffering, no evidence that teachers were suffering worse outcomes where schools were open, and yet it was still impossible to get schools in so many places in the United States opened in the fall of 2020. So one of the most striking features of pandemic policymaking is how these fixed ideas that got set around the narrative in early spring 2020 just were not revisited as new information came in.
Macedo: I agree with all of that. We know too that the course in those mathematical models that were optimistic about these non-pharmaceutical interventions did see school closures as the crucial thing because children couldn't be kept apart. I don't think that was all that important once COVID got started, but school closures were regarded as the crucial measure in the pre-COVID modeling exercises that did play a role very early on in COVID. But Emily Oster did a study of daycare centers, some of which were closed, some of which were open over the summer. The government wasn't gathering information on these places, looking at the differences in transmission between places that were open, daycare centers and those that were closed, but she did some calling around and so on and found no differences in rates of disease across those who were working from home versus those who were working in person in daycare centers.
So there was a real failure to learn from evidence that was being gathered. Even when the Republican states had largely opened their schools over the course of the fall and into the spring, there wasn't an uptake somehow in Democratic states. The governor of Massachusetts at least allowed local schools to make their own decisions on these matters, which provided some grounds for comparison too. Catholic schools were open in many places and so on. There would have been ample evidence, certainly by late fall going into the spring, and yet there was great resistance to learning about these matters.
Lee: It's interesting that governments were not collecting data. We had to rely on the private firms' data on which schools were closed and open. Those data were not being collected. And you've got independent researchers like Emily Oster going out and collecting data. It should have been a top priority of CDC and NIH to work on compiling a superior database of data on which decisions could have been made, and yet they just didn't do it.
Mounk: That's remarkable. I think that comes from a kind of sanctification of particular policies. This is the right policy that anyone who is moral must want. Therefore there's no trade-off involved here. One of the things I was really struck by in those debates about school closures was not just that we didn't systematically collect evidence about whether they had a positive effect on public health in the United States, whether opening schools really did lead to community spread and other negative things. But also that there was a denial of the idea that they might have genuine costs. I remember a number of really striking op-eds published in the time in mainstream publications, which said things like, actually, this is going to be positive for children because, rather than having this boring school learning, they're going to be learning from their families in these other important social ways. And actually it's going to lead to superior development. They often even said this when talking about families where the parents actually did have to go to work, who might have had manual work that was not canceled by the virus. These parents weren't working from home and so could pretend to call into the office while actually playing with the kids.
So just on this point, what do we now know about the extent of learning loss for children that were kept out of school? And do we have a decent estimate of this point about whether they might be able to make that up in the rest of their schooling, or is this just going to lead to many individuals having a permanently lower skill set, being less good at reading and writing and doing math than they might have been if we hadn't pursued those policies?
Lee: If we draw on the best data that we have on student performance—that's the National Assessment of Educational Progress, the NAEP—we see that pandemics and school shutdowns are associated with the largest drops we've ever seen by that indicator in both reading and math. There hasn't been significant recovery since. Closing schools had other effects in addition to the cost associated with just keeping schools closed. It also broke the habits of school attendance for young people, so you saw a rough doubling of chronic absenteeism in school attendance all around the country. That's nationally. It's very hard for students to make up for lost time when in fact they're missing more school than they used to miss. The closures were inequitably distributed, so urban schools were closed longer, schools in high poverty districts were closed longer, and the effects were worse, as you might expect, because it amplified inequalities that already existed in society. These were also not parents who were in a position to be able to spend the day working with their kids to teach them reading and math. So students who were academically lagging fell farther behind. It's not just a uniform drop, though there was a uniform drop, but it was bigger for some than others. Then as it did so, effects tended to widen pre-existing inequality. Students who were lagging academically suffered more.
We don't have evidence that things are improving, so the effects are long-lasting. They were even evident among young people who were just entering school at the end of the pandemic. Because of their lack of socialization—time with other kids in the years leading up to school—they were less prepared, less school-ready, less able to deal with other kids, less able to hold a pencil and do the sorts of things that you sort of expect a kindergartner should be able to do. There was a long tail on these effects.
Macedo: This was perfectly predictable. There's evidence from teacher strikes that was available at the beginning of COVID indicating that these costs would be there and other measures. All of this should have been well known.
Mounk: Just stepping back for a moment, I have a question about how all of this played out politically. It's obvious that we had a very large degree of polarization in the United States in which, a few months into the pandemic, your ideological leaning predicted how you felt about many of those policies and the partisan identity of a state governor predicted to some significant extent how severe the lockdowns they ordered would be, how long they would last. As you were saying, the partisan lean of school districts helped to predict whether those schools would stay closed or not. Is that partisan link the same in the United States as it is in other places? Is there an obvious logic rooted in political theory or political psychology in that?
I'm struck by the fact that when I read one of my favorite books in political psychology, The Righteous Mind by Jonathan Haidt, Haidt argues that conservatives play on a broad set of moral emotions. Liberals often focus in particular on two, which are fairness and equality, so therefore they underplay things like purity and the fear of contagion and so on, which conservatives also are able to exploit for political purposes. Now, if you took that logic seriously and you think about it in 2019, you might think, well, who's going to jump at the pandemic as the dangerous virus that's going to enter our community and spoil our purity and spread this contagion? It should be conservatives who would be particularly reactive to a global pandemic, who would say, do anything to defeat this threat to our security, to our safety. But that's not at all how it played out. So is this all contingent? Does it just have to do with who's in power and what decisions particular newspaper editors made at a particular time? Or is there a more fundamental set of political attributes about this period, which if you run the pandemic a thousand times, then 999 times you're going to get the left wanting longer school closures and more extreme social distancing and the right saying, no, do away with all of those measures?
Macedo: It is clear, I think, that the United States was more polarized on these issues than Europe. Mark Woolhouse, a Scottish epidemiologist, has a very good book called The Year the World Went Mad. He makes it clear that there was more learning over the summer of 2020 in the UK—and of course, European schools generally reopened in the fall. So that was not as not as polarized. In 2020, Donald Trump's presence on the ballot and its being an election year made a big difference. He was probably on his way to being reelected had it not been for the pandemic. I think that there was a strong polarization effect of the scientific establishment seeming to be much more cautious on these matters of reopening and much more concerned about preventing the spread of disease. Trump contributed to the polarization, and it was a very strong sort of partisan effect. I think that made things worse. I'm not sure what I think about this reflecting deeper aspects of liberalism or conservatism.
Lee: I'm skeptical that we can trace this down to differing psychology of Republicans and Democrats, conservatives and liberals. Though I did note how the partisanship seemed to be reversed based on what one would have expected, given conservatives’ greater concern psychologically or what's been documented as a greater psychological focus on purity and on avoiding disease. So to the extent that there are those differences, they're not huge. That didn't seem to be determinative of the response. I really do think it is highly contingent and grew out of a reaction to a president that Democrats did not trust, that on a daily basis would go out and give press conferences, so everything about the response to the pandemic became associated with him. It was seen through that lens.
Of course, Democrats were looking to mount a case for turning him out of office in November of 2020, so they are going to want to say that he's not doing a good job. They laid at his feet every COVID death as if it could have been stopped. That was the rhetoric. Then you had the Biden campaign pointing to closed schools as a reason why Trump should not be reelected. So there was opportunistic use of the pandemic for purposes of developing a rationale for not reelecting Trump. That's part of the politicization. But you could also say it fits in with predispositions among progressives that the pandemic is a societal problem and if we all pull together and act collectively, we can solve this problem together. That's a line of thinking or an approach to social problems that is more recognizably progressive. And so I think it sort of fitted in intellectually with the way progressives think about how social problems can be solved.
Macedo: There's a good point there: we do find that in April 2020, fairly soon in the pandemic, a wartime rhetoric develops around COVID. There's a call for national unity—universities and independent foundations and government and the media all pulling together and being on the same page. Now again, at least in the past, I wouldn't have associated liberalism with the sense of needing national unity in a way that implicitly meant not being welcome to dissent and debate. But that did manifest itself in April and subsequently. So however one regards this from the standpoint of political psychology or political philosophy, the partisan nature of it led those on the left to double down on strategies that were seen to be the opposite of what Trump said. The same thing happened with the origins of the virus. Once he said, it probably leaked from a lab and came from China, then many, many people took the opposite point of view and stuck with it in the face of competing evidence to the contrary.
Mounk: This is one other way of thinking about it, which is that a fundamental fact about America in 2019 was that the left had started to have much more trust in institutions and experts in science and public health authorities than the right, which is itself an interesting deviation from what you might have said about American politics in 1960. It was clearly well established before the pandemic, and then accelerated by it. One element of this is, well, why is it that those public health authorities, which the left trusted to a higher degree than the right, made the decisions they did?
One of the things that's just really striking in retrospect is the amount of group think that you get very, very quickly in those institutions, where topics that are subject to genuine debate, including the social distancing measures, on March 1, 2020, become barely questioned anymore by May 1, 2020, or where an open question about the origin of the virus in January 2020 becomes—and not because there's very strong scientific evidence to rule out an alternative hypothesis—by consensus a natural origin story rather than a virus that was created by gain-of-function research and then, most likely through some kind of accident, escaped the lab.
Why is it that institutions that are meant to have mechanisms for internal debate, that are meant to operate on the principles of academic freedom in some cases, that are meant to actually follow scientific procedures which kick the tires of any empirical claim to see whether there might not be an alternative explanation, failed throughout the pandemic in this remarkable way?
Macedo: The “why” question is hard, but your description, I think, is correct. A consensus developed—and I think you're right, part of it was that the Democratic Party has become more associated with educated elites. The percentage of Democrats with a four-year college degree has increased considerably in the last 25 years, where Republicans have tended to remain about the same at about 30% with four-year college degree. Democrats have more than doubled over the same 30-year period. I think that is part of it, but—again, with respect to the COVID pandemic plans and masks and so on—the sense was in part that these things are only going to work if everyone complies and we need compliance for the sake of the social good.
On other matters like the origins of the virus, why that should have been turned into a partisan issue is harder to see, except that the scientific establishment took the position early on that the virus had a natural origin. We now have lots of information about how that consensus came about. But the scientific establishment took that position, whereas Trump and others have suggested that the virus may well have originated in the Chinese virus lab that studies these viruses and that experiments in making them more dangerous.
One of the things we found is that public health seems to be a very hierarchical discipline. There's a single funding source, the National Institutes of Health, which is hugely important for people working in this area. There seems to be a reluctance, I think as compared with political science, to question and challenge others, even on matters where it seems as though mistakes have been made. So we have found that this is a hierarchical discipline. Actually, the former dean of the Boston University School of Public Health, who is now at Washington University in St. Louis, Sandro Galea, wrote a book himself along these lines called, Toward a More Liberal Public Health. It’s his memoir, and includes the COVID years, and he suggests, just as you've said, that his own field is subject to an unfortunate degree of groupthink and intolerance of dissent.
With respect to the issue of the suppression of speech on social media, we now know that the Biden administration contacted social media companies secretly and asked them to de-boost messages at odds with administration policy and to change the algorithm so that it would boost messaging supporting government policy. We don't know of a single law school conference that's taken place on these matters. The Supreme Court decided that the plaintiffs didn't have jurisdiction so they never got to the merits of the issue (federal appeals court and district courts found that these measures violated the First Amendment). This certainly is a First Amendment concern and it's very surprising to us that in the broad academy many of these controversies around COVID are not being more widely debated. There are few academic conferences on the greatest global crisis since World War II, even though there are so many interesting and important issues concerning free speech and science and public debate and the performance of the media that need to be asked.
Lee: I don't think we have a great answer in the book as to why this happened. In fact, our main goal in the book was just to tell the story, to get it down as a first cut at history. But I was struck all through the pandemic by the lack of skepticism being applied to government policy in this area, whereas as a political scientist, it's just typical that we ask questions about whether government policy is effective and whether it might have unanticipated consequences. I mean, those are organizing questions for the field. We tend to approach those types of questions carefully with a considerable degree of skepticism and great attention to measurement. I was just mystified by the lack of publications comparing the response to the pandemic around the country and documenting the outcomes.
When Steve and I wrote the chapter on the laboratories of democracy, we asked, why aren't there many more publications along these lines in our journals? I know that other social scientists collected these data and they're not terribly difficult data to collect, yet there's just no writing on it. My suspicion is that this is a file drawer problem—that people run these analyses and they discover that there's a lack of association with the measures. Then they regarded it as not being an interesting story to tell. But the reluctance and failure of the academy to sort of confront these obvious and important questions about government policy during the pandemic is really striking.
Mounk: What do you think this is going to do to long-term trust in the institutions? How has it changed your trust in these institutions? I have to say that my political values today are roughly the same as they were 10 years ago. But the degree to which I trust the institutions of the state to be able to have an effective and rational response to a major crisis like a pandemic has suffered tremendously. The amount I trust in the consensus of experts has suffered tremendously because we see with example after example in the pandemic that things—which with the benefit of hindsight now appear to be based on erroneous evidence, or appear to have come to the wrong conclusion—were celebrated as the great wisdom of “Science.” This doesn't mean that I'm skeptical of the scientific method, or skeptical of the importance of studying things in an empirically rigorous way as you indeed do in your book. But it does mean that claims that something is based on “the science,” or claims that some policy is needed because it is favored by so-called experts, seem to me to be much more dubious than my instinctive response would have told me 10 years ago.
Macedo: I very much agree with what you just said. My confidence in liberalism and science has not been shaken, but I have found myself more skeptical and more disappointed—I guess I would put it that way. On many matters scientists and academics and journalists are not willing to ask questions when the valence of the question and the possible answers are politically unwelcome. There's a very good article by Cory Clark et al. in the Proceedings of the National Academies of Science on pro-social motivations for bias in scientific work in general. They find that there's a tendency to moralize scientific questions and to take one side and to be unwelcoming to people who come from another side because of what are seen to be the political or the social or the moral implications. Obviously we are concerned about the moral state of our society, but empirical evidence should proceed open-mindedly and dissenters should be listened to even when they seem to challenge the partisan inflection of a position that we might prefer. So we have been very concerned about that. Another way to put it is we're publishing this book at a time when both science and universities are under considerable threat, now even more so than just a few months ago.
But I think our feeling is that the premise of those attacks is that these institutions are politicized. And the only way to show that's not true is by being willing to ask hard questions and being willing to pursue the answers where they lead. Our hope is that by pursuing these questions and trying to address them in a fair-minded, objective way, and taking seriously the arguments that were coming from the Republican side to a greater extent, that might restore some measure of trust in universities and science and related disciplines.
Lee: My experience post-pandemic is similar to yours, Yascha, in that I find myself questioning any time I read a claim in the newspaper or a news magazine that the science is settled on X or Y. I just don't trust those claims in the way that I did prior to the pandemic. I feel that I need to do more research on claims around climate change, that I don't trust this in quite the same way that I did. I know how scientific communities can circle the wagons and make certain types of questions unwelcome and disincentivize scholars from asking them. And I also know how newspapers and other news outlets can maintain an editorial line so that obvious questions don't get asked.
It was evident by the end of the summer of 2020 that the speed with which states had locked down had not made a difference in COVID outcomes. You could see that by that point. Yet there was no reporting on that, even though there had been all those articles in the lead up to the lockdown saying that every day you wait the exponential spread will have all these dire effects because models suggested that. The models were oversimplifications of society and one thing they didn't take into account at all is voluntary behavioral change. Where was the revisiting of assumptions from back in the spring? There was just a complete shutdown of all those sorts of inquiries. The idea that these measures might not have been effective, that keeping schools closed might not be making a difference—and what exactly is the evidence around population masking for control of the spread of a respiratory disease? Why were there not many articles on this in our major news outlets? The fact that I saw those kinds of group-think dynamics unfold during the pandemic has left me feeling very cautious about any kinds of claims about what science says for purposes of policy. I feel that it's necessary to go and do some study before I'm prepared to accept on faith any testimony from experts as reported in the news.
Mounk: That is genuinely sad. I think this is one of the areas of our society where the failings of one institution or to some extent one political side then reinforce the failings of the other institution or the other political side. Clearly there were these pre-existing structural problems in the public health community—that it's far too hierarchical, that perhaps it’s too driven by certain dubious ethical principles like the precautionary principle, that it's not actually open to genuine debate in a moment of emergency. That has then led to attacks on those institutions that are often exaggerated or misplaced. And now in the form of the Trump administration, it leads to just outright hostility to the great universities in this country and to institutions like the NIH.
But that then in turn leads to rallying around the flag, in which a lot of people in our spaces are saying, no we have to defend the NIH exactly as it is, we have to defend the universities exactly as they are. I feel even in these last few months in my own academic circles—in my own more left of center circles—the space for criticism of genuine failings of universities and more broadly the scientific communities closing down. And yet, when I speak to friends who are in the hard sciences, they certainly want the federal state to put a lot of money into scientific research and they believe, I think rightly, that it is of great medium and long-term benefit to society. But they’re also deeply frustrated even beyond these political issues about how the NIH gives grants based on very small core research that is often somewhat ideological. But even when it isn't, it's just based on these sort of very small incremental steps in ways that really misshape the research agenda and make it really hard to do ambitious science. So what are some solutions here? Are there some ways in which we can reform the university, reform the public health institutions, reform the funding agencies like NIH to actually drive a bigger positive impact in the long run and to avoid these pitfalls?
Macedo: Well, let me just make a couple of comments. I'm not an expert on the structure of the NIH. I do think you're right that there are many matters in universities and in science that are worthy of much more open discussion and debate. We need to have those discussions and be much more open about criticism. I just think one fundamental thing is that we need to remember that the institutions of science and the academy depend upon certain basic values and value commitments on the part of the practitioners. That includes scrupulous adherence to the scientific method, to following the evidence where it leads, and to being open to criticism. I mean, I have an article here from the Washington Post in December 2020 talking about John Ioannidis, an evidence-based researcher at Stanford who was taking some unorthodox positions on COVID. One of his own colleagues, talking to the Washington Post reporter, says about scientific debate that it's very important to have debates in science and public health about things like the lockdowns and the closures and so on and so forth, but if conducted in public, the rules change. These debates can confuse people and undermine the consistent messaging needed for good public health. Politicians can misuse these debates to undermine public health and so on and so forth.
Well, we have to be more open to allowing these debates to break out in public and not try to hide things from the public. One of the books that we found useful in thinking some of these issues through is Sissela Bok’s book on lying. One of the things she finds is that educated elites, for quite some time, have tended to be skeptical of the capacity of the public to understand the nuances of their debates and therefore have a tendency to mislead systematically in order to nudge the public in the right direction.
Mounk: So there we are back in the realm of the supposedly “noble lie.”
Macedo: Exactly right. Or noble obfuscation. Sometimes it's not outright lying, it's just not telling the whole truth, or saying something misleading, or hiding the truth. But absolutely, some variation on those themes. And I think it's very destructive. People have ways of finding these things out. Paul Offit at the University of Pennsylvania has written on some of these issues quite usefully. He was involved in the vaccine trials and with respect to the question of side effects for some small number of people, he makes the point very strongly that we need to engage in truth-telling with nuance. That the public will find out if they’re being misled, people will resent being misled, and that will lead to terrific distrust. So I think there are many matters around COVID and controversies in universities in general on which we need much greater frankness and willingness to acknowledge disagreement and to acknowledge past mistakes.
In the rest of this week’s conversation, Yascha, Frances, and Stephen discuss the origins of the COVID-19 virus, what could happen in the next pandemic, and how institutions can regain trust. This part of the conversation is reserved for paying subscribers…
