Five years after the COVID lockdowns, the performance of government and policy experts is not looking great in retrospect. Stephen Macedo and Frances Lee find that policymakers dispensed with years of pre-pandemic planning that suggested the tools used to fight COVID would not work. Experts did not sufficiently consider the costs of their preferred approaches and spoke publicly of consensus while privately admitting limited evidence. Policymakers and experts deterred alternatives and suppressed dissent, leaving us with today’s increased distrust of health and political authorities. The second Trump administration is now empowering the skeptics and taking advantage of Americans’ distrust of expertise.
Guests: Stephen Macedo and Frances Lee, Princeton University
Study: In Covid’s Wake
Transcript
Matt Grossmann: How Policymakers and Experts Failed the COVID test, this week on the Science of Politics. For the Niskanen Center, I’m Matt Grossman. Five years after the COVID lockdowns, the performance of government and policy experts is not looking great in retrospect. What lessons can we learn? Can we guide policy with evidence without rule by the unaccountable and overconfident? Can we make quick decisions that consider benefits and costs and learn from early data to refine our approach, or perhaps we’ll overlearn from our mistakes Next time, dismissing the possibility of independent expertise.
This week, I talked to Stephen Macedo and Frances Lee of Princeton University about their new Princeton book, In COVID’s Wake. They find that policymakers dispensed with years of pre-pandemic planning that suggested the tools used to fight COVID would not work. Experts did not consider the costs of their preferred approaches and spoke publicly of consensus while privately admitting limited evidence, policymakers and experts deterred alternatives and suppressed dissent, leaving us with distrusted experts and political authorities. The second Trump administration is empowering the skeptics and taking advantage of American’s distrust of expertise. So it’s a good time to review how we got here, even with some uncomfortable lessons for the expert class. I hope you’ll benefit from our conversation. Stephen, tell us about the major findings and takeaways from the new book.
Stephen Macedo: Well, thank you, Matt. Happy to do so. I would say that one of them was that we found a lot of group think among elites, not enough questioning of strategies that were adopted, an intolerance of dissent that developed over the course of time, unwillingness to question assumptions and failure to learn as more and more evidence was available. So we’ll talk more later on about policy mistakes that were made and costs of policies that were adopted. But at a sort of high level, I think group think among elites, and obviously among the general public, there was a lot of panic and fear that was understandable, but that was fed by journalists very often who emphasized preventing deaths from COVID. And again, they’d also did not ask hard questions. There was a lack of leadership. Politicians didn’t want to step out of line and question strategies very often. And so, that was another one.
And as far as academics go, there’s been a remarkable lack of curiosity given the magnitude of the problem, the magnitude of the response, and the massive consequences of what we did. Even around free speech issues, there hasn’t been much by way of law school conferences or debate. So one of the major themes I think was aside from the policy failures, which I know we’ll get into, and they were things that are well worth talking about, the way we think about it is there are truth-seeking departments of liberal democracy, the academy, science, science journalism, who are tasked with objectively pursuing the truth of big important public questions, and they just didn’t perform as well as they should have.
I think we find that parts of it were the influence of partisanship and undoubtedly the polarization that we have in the country these days. I think COVID provides a window onto the degraded state of public debate in conditions of polarization and partisanship. There’s more to be said there as well. I think there’s also been failures of truth-telling on the part of some public health officials, and we can get into that. But I would say those are the big takeaway themes politically.
Matt Grossmann: Frances, we’re at the five-year anniversary of COVID and you all not only want us to go back to it and learn some lessons, but relitigate every decision that was made along the way. So why is it worth doing that? Why should we go back through all of this?
Frances Lee: I appreciate that. This is a painful topic that not very many people want to return to, but these are among the biggest, most consequential decisions that government has ever made. I mean, just looking at the budgetary impact alone, we’re talking about the largest crisis response in US history as gauged by share of GDP. The COVID response in 2020 was roughly on the scale of both the New Deal and the 2009 stimulus combined. 2021 COVID Aid was another new deal all over again. So we’re talking about massive investments, all of which were added to the public debt. And so, we’ll be born by future generations annually in higher interest payments on the national debt. So we continue to bear the consequences in a fiscal sense that society bears the fiscal consequences well beyond what we can tally up in financial costs.
I mean, there’s long-term effects on education and on child development stemming from school closures. There was health deterioration of elderly people in nursing home care due to isolation, where their dementia became more severe, their mobility problems became more severe. There were effects on crime in the US, the largest ever surge in the homicide rate in 2020. It took a couple of years for that elevated crime rate to decline.
I mean, the consequences are so broad ranging, it is hard to even survey them all, that we feel like in our book, we barely scratched the surface and that there’s a lot of room for more research in this area. So the consequences call out to us for a look at what we did right and what we did wrong. It’s a shame we have never had any kind of COVID commission looking back at these decisions, trying to find some societal consensus. Because what has happened over time is that the interpretation of the pandemic has become more deeply divided in society, more deeply divided along party lines. And so, to figure out what we should do as a democracy next time we face this type of crisis, next time we face a crisis that is understood or presented to us as an existential threat, that crises that have that character are likely to give rise to some of the same kind of flaws that we document in the COVID response.
Matt Grossmann: Stephen, you come from a political and legal theory background and have analyzed several other policy areas, but you also caution experts to stay more in their lane. So tell us about how you became interested in these COVID debates, and what background do you bring?
Stephen Macedo: Yeah, well, thanks. Like everyone else, of course, I’m concerned about the state of American democracy, the apparent fragility of our institutions and the influence of partisanship and polarization on them. I started out actually with a broader book in mind on several policy areas where I thought progressives weren’t paying enough attention to the arguments coming from the other side of the political spectrum on the merits. And that was going to include immigration, which I have written on, and where I think there are more downsides to high levels of immigration at important parts of American history for working class people than progressives have often allowed, academic progressives, certainly. And abortion, which I think is a harder question. It used to be regarded as a harder question, and I was going to do COVID as an additional topic. But I soon realized having gotten started on COVID, that was a massive problem itself as Frances has just mentioned.
So I decided to focus on that. And then after I’d gotten started on it and organized a workshop with Frances, because I’d been talking to her from the start, she decided to join on, which I was delighted by. But my core interests have always been in liberalism and democracy, liberal democracies, and the core value of liberal democracies is equal freedom, of course, but also a commitment to public argument and public reasoning, public deliberation in a fair-minded way, openness to dissent and criticism, freedom to criticize, and a commitment to trying to solve our disagreements on the basis of evidence and reasons.
I mean, that’s the ideal. We often fall short, of course, but COVID policymaking discussions around it, the history of the emergence of the virus and the deliberation that took place around it provides, as I said before, a window onto those aspects of liberal democracy not functioning nearly as well as they should. So I come at it as a political theorist interested in public policy questions, looking at a situation where we face an existential crisis and one in which elites are called upon, educated elites are called upon to play a central role and debating, deliberating about providing information about what’s going on, and fall prey to a surprising degree to groupthink, partisanship and polarization. So it’s a cautionary note for the state of these institutions of more broadly, I think.
Matt Grossmann: Frances, this book is quite a bit more scathing than some of the previous books that I have read from you. And I know that you’re known for trying to give policymakers the benefit of the doubt as much as possible, as well as researchers. I was just figuring out, try to figure out how the COVID experience changed your perspective and if there was any radicalizing moment here in living through these debates.
Frances Lee: That is such an interesting question, Matt. I saw my reaction to the COVID crisis as a piece with my approach to studying other policy debates and policy decisions over the course of my career. I remember when the lockdowns began, when we were sent home from the university at midterm in 2020, just noting how far the virus had spread in the US and around the world, and thinking to myself, “I just don’t think this is going to cut it. I don’t think this is going to work.”
And as I’m thinking that through, I’m wondering how long it will take all our institutions and society to sort of catch up to the reality that this virus had spread too far to be stoppable. It’s going to be like a cold or a flu or any other respiratory virus that society’s plagued by. I mean, I didn’t think of it at any minimizing terms. I was afraid of what was going to happen, I mean, terrified at times, but I just didn’t think it was going to be stoppable. And so, I thought there was a folly there in not recognizing the possibility of failure and for revisiting questions, revisiting decisions in light of new information as it comes in.
I see this skepticism about whether government action works or whether it might have unanticipated consequences as going all the way back to my dissertation work, which looked at problems stemming from equal representation of states in the Senate and how that biases federal policymaking in important ways. I don’t think our book has a muckraking tone. It’s not accusatory with maybe some exceptions around we don’t think that there was a sufficient transparency around what was known about the lab leak, potential for a lab leak, the lab leak hypothesis. But in general, I think that the tone of the book or accusatory instead, it’s just like standard political science. Sometimes government action doesn’t work and sometimes it has unanticipated consequences. I mean, that’s a normal theme for us. It was just on a really grand scale in the COVID pandemic.
Matt Grossmann: Stephen, you go back to the beginning, look at these pre-COVID pandemic plans and find that they were very skeptical of lockdowns or mass testing and tracing, and many of the policies that we ended up implementing. And the story seems to be mostly that we, the West as a whole, was copying China’s earlier policies rather than relying on those prior plans. So what do you see as the most credible alternative path that we could have taken in that early going, and why was it that everybody copied those policies?
Stephen Macedo: Well, it’s just, as you say, it’s remarkable to go back to the couple of decades of pre-COVID pandemic planning that was available. In fact, the World Health Organization updated its pandemic plans in November 2019, just a few weeks before the onset of the COVID pandemic, and as you say, was very skeptical about the evidence supporting the efficacy of the whole range of non-pharmaceutical interventions, from masking, to various social distancing measures, school closures, business closures, and so on. Johns Hopkins University similarly did a survey of all the available evidence that came out in the summer of 2019 and emphasized both of those as well as others that not only was there a lack of evidence, a weak evidentiary basis for these non-pharmaceutical measures, social distancing measures, but that it was quite certain that there would be great costs if schools were closed and businesses were closed and society was disrupted. So the message there is largely quite skeptical. And then the question is to why we deviated from them. Yes, the World Health Organization sent a team to China. They spent a week in China in late January, I think it was, and-
Stephen Macedo: In late January, I think it was, and issued a report after spending five or six, seven days there very quickly with a fulsome endorsement of the Chinese lockdown strategy. They unequivocally endorsed the strategy, said that China had stopped the virus, suppressed the virus, and that every country in the world should copy the China model. That was one influential factor. There was also a dire report done by a mathematical modeler, and I should say that there were mathematical modelers who dissented from those Covid plans that I just mentioned. They were influential when the George W. Bush administration started studying pandemic responses because President Bush was concerned about the possibility of another pandemic. These modelers had been optimistic based on their modeling projections of the efficacy of school closures and other measures.
They did not have the general support of those who studied these things closely, but they were optimistic. And Neil Ferguson in Imperial College had been involved with these from the start. And he put out a document on March 16th, 2020, arguing that there would be, I think it was 2 million deaths in the United States by August 2020 and half a million in the UK if we did not implement a severe set of social distancing measures, school closures, lockdowns in the economy, and so on. So the Imperial College report on March 16th, which Deborah Birx and Anthony Fauci showed to Donald Trump that day before a consequential news conference, endorsing lockdown measures was another important finding. And then there just seems to have been a kind of cascade effect.
Italy was actually the first country to institute a nationwide lockdown measures in early March, and I think they demonstrated to western European countries and others that, well, it was possible something that had heretofore been thought impossible was possible, and the citizens of a Western democracy were willing to tolerate this kind of response. So from there, there was a cascade effect around the rest of the world, a lot of copying, a kind of herd effect. The evidence shows that countries tended to enact lockdown measures when countries proximate to them were also enacting lockdown measures. The lockdown measures were not responsive to increases in viral spread or increases in mortality so much as they were to proximate countries implementing these same kinds of measures.
So exactly why we did it is hard to say. I mean, the various influences I’ve just mentioned for sure, and a considerable amount of group think also, I think, on the part of leaders. Now, in terms of what the alternatives were, well, there were some dissenters from these lockdown measures in March. We quote them, prominent epidemiologists and public health scholars, Michael Osterholm at the University of Minnesota, David Katz at Yale, Tom Frieden, former CDC director, Justin Less ler, and others all sort of responded to China by saying, “This is a bizarre strategy. They must be poorly advised. It’s unlikely that this will be effective and it will be very costly, these lockdown measures, and we can’t imagine these measures being adopted in western democracies.” Anthony Fauci said something to that effect.
So there were dissenters early on, but a strong consensus developed among elites in April, as we see. There was a first national report, national pandemic planning document put out by the Safra Center at Harvard with the support of a number of national nonprofit institutes, the Rockefeller Foundation and others. And the urging was that there’s only one path forward. This is the equivalent of war. In wartime, people need to pull together. There needs to be vital unity. And at that time, in March and May, dissenting voices seemed to wane and disappear. And of course, messages that dissented from government policy started to be removed from social media sites. So that seems to be what happened and we can talk more about the intolerance later on.
The alternative strategy was what was mentioned by some of these dissenters in March to protect the vulnerable, not to lock down the entire society and economy, but to recognize that there was a tremendous age gradient that was already apparent in March and April 2020 that the elderly people were much more vulnerable to Covid than others. That large parts of the population bore little vulnerability to death from Covid. And there could have been a more stratified approach to focusing protection on the vulnerable. Now, is it guaranteed that this would have worked? Well, we could have certainly done more to protect people in nursing homes and other vulnerable parts of the population.
But the thing that we’re most confident about in the book is that these alternative strategies should have been debated more thoroughly and that people should have been more open to alternative strategies and should have recognized earlier on that general lockdowns and school closures were not being effective and were exerting or causing significant costs on society.
Matt Grossmann: And for instance, there wasn’t a lot of internal variation in the US either. One thing polarization is supposed to give us is at least two different strategies to debate between, but national and state policymakers quickly followed the pronouncements of the expert class. Every state had put forward these public health officials who people hadn’t previously encountered Trump, who’s not known for being deferential to expertise in his administration, went along with it even though there were pretty early indications that the politics might favor a different approach.
Now, eventually, obviously we did get polarization and polarization on expertise and the role of experts as well. But how did we get to the point when Trump and Republicans went along with all of these things and then polarized afterwards? And what about the Democrats? Why wasn’t there an effort to learn from the limited evidence of early effectiveness?
Frances Lee: It is one of the most striking features of the Covid response that it began in such a non-polarized way. So I mean, I think policymakers at all levels felt the pressure of the crisis and that they needed to do something in response to it, that every day of delay potentially might inflict harm on their populations. So I think an… Aware of that, the galloping nature of the crisis, they accepted the recommendations coming from the coronavirus task force at the White House. March 16th is a key date in that that is the… So Steve referenced that as the day that the Imperial College released report number nine, which offered dire projections about the deaths to be expected in the US and in the UK and other places, and the prospects that that death toll could be reduced through the use of non-pharmaceutical interventions.
So after the decision was taken to recommend the use of those measures in the White House, then there was a press conference. And at that press conference, there was a two-page document circulated that laid out the strategy that said that in states with evidence of community transmission, governors should close schools and should close businesses where people congregate. And so you begin to see the lockdowns get implemented within days of that. The first statewide lockdown was California on March the 19th, and the last was three weeks later in South Carolina. So it was a wave after wave of states announcing stay-at-home orders and specifying on those orders who were essential and who were non-essential workers. And all the restrictions that were in place by late March.
And where you begin to see the divergence in state policy on party lines is in the reopening process. Republican states stayed closed for about 30 days on average, Democrats more like 73 something, I think that’s the average. Democratic states two and a half times as long as Republican states in terms of maintaining those stay-at-home orders. Over the course of that subsequent year before the vaccine rollout occurred, there was substantial policy divergence between Republican and democratic states. Republican states reopened schools more quickly than democratic-leaning states. It’s the best predictor of the state policy is the partisanship of the state, the partisanship of the jurisdiction even if we’re looking below the state level to school districts. The best predictor of whether schools would reopen is the partisanship of that locality.
So it became highly party polarized, but that was a sequential process and evolving process of divergence rather than something that was baked in at the very start of the crisis. Very interesting story of how polarization can shape policymaking. Why wasn’t there learning from this policy divergence? And that’s mysterious. I would have thought that there’d be enormous interest in what’s happening in these places doing things differently. You would think that there would be, but we couldn’t find evidence that there was such interest. That where states were studying outcomes in places taking a different approach. Instead, you see sort of states benchmarking themselves against other similar states. I mean, I guess if we look at the political science literature on policy diffusion, maybe that’s what one would expect.
But with an unknown problem like this, you’d think new data coming in all the time that there might have been more policy learning. But instead what we find is that states that start off more stringent in their Covid response, stay more stringent than other states all through the crisis. There’s not a reversal of their early decisions. The democratic states remain more stringent than Republican states through the crisis. Even as reopening occurs, that reopening happens in a way where you can still see that sorting on a gradient that tracks partisanship. And the governors were popular in all the states, and in 2022 governors that had taken wildly divergent approaches to the crisis all got reelected.
So you might say in that sense, democracy worked in Covid in that Americans got what they wanted from the Covid response and were happy with it no matter what it was, or as equally unhappy with it. They certainly didn’t seem to blame their governors. The governors seemed to do quite well politically through this crisis.
Stephen Macedo: I guess it could add that there was one outlier internationally, Sweden, which was invoked as a kind of pariah example of not locking down sufficiently. And they did have higher death rates early on. Now there were various reasons for that, including that they had had a mild flu season the year before. But they, along with the other Nordic countries, took less stringent measures in general. But within a year they were doing very well by international standards. And the evidence now suggests that they performed as well as any other place in Europe and did extremely well.
So the Swedish case was held up as a very poor example, but there wasn’t a lot of follow through and people were not very critical about it. And even now people seem very reluctant to allow that the evidence shows that Sweden did very well. Francis and I have both encountered a fair amount of skepticism about that possibility from people who were reporting quite otherwise during the pandemic.
Matt Grossmann: And Francis, we also know that policy can cause social behavior differences, but it can also be an effect of them. And one of the things that you find is that some things that might be attributed to partisanship leading to different policies, like differences in vaccination rates and associated outcomes, were actually more about the social behavior of individuals. I guess, how do we think about that, the effect of policy versus the effect of things that might lead to policy differences?
Frances Lee: So we do our best in the analysis to sort of take account of what’s known about individual preferences relative to mandates or policies that the states impose clearly both matter in the Covid response. Now it’s less of a problem for the analysis than one might expect in the sense that democratic states were both more stringent in policy terms and all the evidence suggests that democratic populations were also more compliant with the recommendations. So those two things don’t cut against one another. Democrats were more compliant with mask wearing expectations.
Frances Lee: … compliant with mask wearing expectations, polling consistently showed that. Cell phone mobility data showed that Democrats stayed home more than Republicans did. We’ve got evidence that behavior aligned also with partisanship. With the vaccine story, all states had a vaccine rollout funded by the federal government where people could get vaccinated for free. There was some variation across states in how quickly the vaccine was made available to all adults. But the way that worked is that it was Republican states that tended to open up vaccine availability to everybody more quickly. Democratic states had a heavier crush of demand and had a more gated process.
But what polling showed from the very beginning of the vaccine rollout was that Democrats were more eager to get vaccinated than Republicans and that the difference between Democratic and Republican states began to emerge immediately in terms of vaccine uptake. And so it’s hard to trace this back to policy because the policies really weren’t that different across states in terms of making vaccines available. And if anything, they were more easily available in Republican states, including you didn’t have to wait as long to get an appointment. And so I think you have to trace those differences back to individual preferences rather than to policy.
Matt Grossmann: Stephen, you criticized the laptop class, which we’re all a part of for not realizing that the burdens that were being imposed would be harder on other people. But as Francis said, a lot of these policies were quite popular across the board there. Only sometimes did a class divides materialize, and obviously professionals were also involved on the other side in trying to get the schools reopened more quickly, for example. In what way was this a class war or a failure of one class to represent the burdens they were imposing on others? And in what ways was this just a national over caution?
Stephen Macedo: Well, I think there was probably some of both, but it’s definitely the case and it seems like people were not sufficiently cognizant that in order for much of society to be able to stay home, locked down, keep their kids out of school and so on, a third of the workforce needed to keep right on working in person in order to keep us supplied with electricity, food, wine, and all of the other comforts that allowed us to continue working. Food had to be processed, food had to be delivered, food had to be farmed and so on in the first place, truckers, police, firefighters, other utility workers and so on, so a lot of people kept right on working in person. Many of them were vulnerable. No particularly special measures were taken to protect the vulnerable members of the so-called essential workers. And if the schools were closed, they would’ve borne special burdens in terms of how to deal with their stay-at-home children during the pandemic as well.
We think there was a disproportionate burden borne by essential workers on a class basis, even though as you say, there was much popular support for the lockdown measures. We should also point out that the special burdens borne by essential workers, working class people, disproportionately minorities, is something that had been emphasized in the pre-COVID pandemic planning documents. They were all quite emphatic on that score that many people would have to keep working, that disproportionately poor minority people would be disproportionately burdened. Indeed, the American Civil Liberties Union in 2008 itself wrote a pandemic planning document arguing against conflating the values of law enforcement and public health and arguing that mandatory measures would tend to have a disproportionate effect on members of minority groups. We do think there was a disproportionate burden. We do think the benefits flowed disproportionately to the laptop class, able to work from home. People were actually able to do home improvement projects to have new couches or new refrigerators delivered.
And indeed, there was an uptick in home improvement projects during the COVID pandemic, people upgraded their kitchen appliances and so on. It’s actually quite remarkable to think about just how many people were working along in person in spite of all that. Yes, we do think that there was definitely a class element to this and that should not be ignored. Children of course also didn’t have the right to vote, so they’re also disenfranchised. And there’d be other political scientists who have argued this before, but that was another way in which those who are relatively politically powerless, this could be said of the right elderly as well in nursing homes, were not as well represented in public deliberations, and their interests were not as well represented, I think as those of better educated knowledge workers. Indeed, some of the dissenters from COVID policy described the policies we adopted as a policy of focused protection for the laptop class.
Matt Grossmann: Francis, we can excuse some of the public as just being scared and wanting to do something, but you don’t excuse the experts because they didn’t just err on the side of caution in your view. They actively inhibited alternative views from being considered, they considered the benefits rather than the costs of their policies. And in some cases, they spoke publicly of a consensus despite knowing that there were big holes in the evidence and that there wasn’t a consensus behind the scenes. Are these things that we should consider inevitable outcomes from linking expertise with policymaking that has to be done in a crisis mode, or did something in particular go wrong here?
Frances Lee: Thank you for that very insightful question. I do think that the COVID pandemic offers a cautionary tale about the role of experts in policymaking of technocracy. We might want to broaden it out in that using that term that this was policymaking led by people who claimed expertise in public health or in epidemiology, that those were the experts that policymakers turned to. When they formed these advisory bodies to develop COVID policy both here in the US and around the world, like around the coronavirus task force, they would bring in virologists and public health bureaucrats and then generalist government officials and no other kinds of experts were weighing in. You often hear the COVID response described as erring on the side of caution, the precautionary principle. Let’s stop everything and figure out what’s going on, that that was the cautious thing to do. Well, it’s cautious with respect to the virus, but it’s not cautious with respect to the rest of society that there’s a recklessness in policymaking where the costs are not considered.
You can’t only look at one side of the equation when you’re making policy decisions. You can’t just look at the benefits that you’re hoping to achieve. You have to also consider what are the costs. And the really striking feature of COVID policymaking is that the benefits were uncertain. The prior pandemic planning literature is very emphatic that the state of the evidence on non-pharmaceutical interventions was poor, so we did not know that those measures would yield benefits over the course of the pandemic, but it was obvious that those measures had large costs. And so I do think we need, as political scientists, as people who study ethics and public policy, we need to grapple with the dilemmas that policymakers were facing in the pandemic. It was a crisis, a fast unfolding crisis, and there was a lot of uncertainty about what to do. And there was uncertainty about whether the options they had available to them, whether it would do a lot of good or not.
And then it was completely clear that the measures that they had available were going to be very costly across numerous dimensions, not just financially, in fiscal terms, but also in terms of ethics and morals. To impose outsized burdens on some share of the population to protect another. The young bearing burdens for the old, there were winners and losers in COVID policy. And so we can’t think of this as just a decision that benefits everyone equally or that was the decisions that they were taking had equitable benefits and equitable costs. No, the costs fall more heavily on some populations and the benefits fall more heavily on others.
Stephen Macedo: It might be worth adding that all of these things were oversimplified or in a way brushed aside with phrases like follow the science and the deference that was paid to relatively narrow experts in epidemiology and indeed public health. All policy choices, as we argue in the book, involve valued decisions and valued trade-offs, weighing risks, policy alternatives, involve the distribution of risks across different parts of the population and weigh things like the value of liberty, the value of attending church on Sunday versus the risks of doing so even when social distancing measures might’ve been available to people attending church. These decisions should have not only involved a wider range of experts, but should have involved more public deliberations since it’s told me the public that had to live with these policies. And we should have also recognized that they were not just matters of scientific expertise, but matters of weighing value judgments. And as Francis said, burdens and benefits spread across the population unevenly. And unfortunately simplistic slogans like follow the science made it impossible to recognize these things.
Matt Grossmann: But Stephen, the backlash is here. Your book is coming out at a time when the new Trump administration is going after the experts, including expert-filled agencies as well as universities appointing some of the people who were COVID dissenters. And clearly not the whole story, but a big part of that justification and that move on the right was partly in response to the same failures that you’ve identified in COVID. To what extent is this an outcome of the COVID experience, and what about the risks of the other side? Are we going to move just as far in undoing the influence of experts and indeed the creation of the evidence base in reaction?
Stephen Macedo: Matt, that’s a very good question, and indeed it’s something that has concerned us, that friends colleagues have urged us to be more careful about the ways in which our arguments might give, as it were, aid in comfort to those who want to simply undermine expertise and universities and the knowledge-seeking departments of liberal democracy, as I mentioned before. We are of that tribe and we believe in expertise. We believe in the great good work that’s done by universities and the need for journalists and science and science journalism in a liberal democracy to allow us to get good information about scientific matters, public health matters, and to inform public debate and public discussion. And indeed, of course, there are many academics who are focusing on the dangers of populism, authoritarian populism in particular. And we decided instead to focus on elite failures because we just think that in this time of fragility for liberal democracies and threats, we need to make sure that our own house is in the best possible order.
And in fact, we do find evidence of what we feel are shortcomings that have been made, and we think that course correction and are doing better could be a part of helping to strengthen these institutions. Now, we’re not soothsayers. We can’t predict the future, but if you look at things like the recommendations on vaccines that were made during COVID, and here we draw on the work of Dr. Paul Offit at the University of Pennsylvania Medical Center, the Children’s Hospital of Pennsylvania in particular, he was on vaccine advisory committees. And he worries that evidence about side effects for some parts of the population from some vaccines, young men in particular were downplayed.
And he feels that that undermined trust in science and the vaccine recommendations because people do have access to other sources of information. And the fact is that there are some rare but arguably significant side effects for some segments of the population. Similarly, there was other misleading messaging around the role of natural immunity in acquiring herd immunity that people can acquire immunity to infectious diseases, not just from vaccination, but from getting the disease and recovering. That was strangely obscured, it seems intentionally, by major scientific institutions during the COVID pandemic. In order to encourage young people in particular, I think to avoid getting COVID and instead wait for the vaccine. Now, that may have been a good message to convey, but it was an obvious falsehood to deny as people seem to do that-
Stephen Macedo: … to deny, as people seem to do, that getting COVID and recovering from it wouldn’t contribute to your immunity. That’s a silly thing to have suggested. So I just think we don’t think that these kinds of kind of blatant obvious manipulations do any good for people.
Frances mentioned briefly the lab origin issue, and I know we don’t have time to get into that, but it was always a possibility that the COVID virus emerged from the Wuhan Institute of Virology, China’s leading bat coronavirus research institution. Many Chinese cities have wet markets, but only Wuhan has the Wuhan Institute of Virology, which experiments on coronaviruses.
To have as vehemently denied or tried to deflect attention from that possibility as occurred, obviously seemed to be manipulative and political, and I think it’s been perceived by people as such, only a small minority of Americans believe in the natural origins theory.
And I might just mention one other data point on that. On February 21st, 2025, Nature Magazine published an article, and I read the title here, What Sparked the COVID Pandemic? Mounting Evidence Points to Raccoon Dogs.
That article quotes six scientists, six scientists who are well-known partisans of the natural origin theory, who participated in communications early on in which some of their Slack messages and email messages, which are now public, suggest that they had much graver doubts about a lab origin than they were willing to admit in public. The article quotes no scientists on the other side of the debate that increasing evidence points to a lab origin.
The only thing I’m going to say about that is it’s an obviously unbalanced essay in Nature Magazine in February, 2025, and obviously represents a kind of politicization of the issue, suggesting that it’s important for various reasons to deflect attention from a possible lab origin.
So I think that at least we tend to judge that failures of candor, frankness, nuance, allowing that we can have strong recommendations for vaccines while allowing that there may be some evidence of some small numbers of side effects. That truth-telling with nuance is the best strategy, to be frank with people. And after all, experts have not been authorized to mislead the public for their own good.
So we just believe that noble lies will be seen through, in a democracy they’re disrespectful, and that academics and scientists owe the public the truth to inform public decision-making, and then allow deliberation to take place, values to be weighed in a more open and public way. So that’s our hope that frankness candor will contribute to the strength of academic institutions, and that we’ll have a better chance of being more trusted if we behave in a way that’s more trustworthy.
Matt Grossmann: Frances, we might learn the lesson successfully from COVID, but there are also potential risks for overlearning things that were particular to this pandemic. We are currently facing measles outbreaks and bird flu. We might face Ebola. There might be different circumstances than the particularities of COVID. Just to cite one example, there might be something that is more problematic for the young rather than the old. There might be something that is more deadly, or might be caught in the beginnings stages rather than spread so fast.
So it could be that we didn’t learn those lessons from COVID, so we won’t for the others, but it could also be that the politics now for the next pandemic are that basically nothing will be believed or could be implemented in any location because of our experience here. So how should we think about using these lessons for the next time?
Frances Lee: I think that the big generalization one can draw from the COVID response is that policymakers need to be open to emerging evidence that they can’t let the sunk costs of their early decisions then lead them to refuse to reconsider, that they have to be more open, pragmatic, willing to consider emerging empirical evidence.
The problem with the crisis is that the particularities weren’t uncovered, and learned, and taken on board for policymaking, that there was a blunt approach adopted at the start, and then even a disinterest in learning about the particularities, such as the much greater risk to the elderly than to the young, that the COVID pandemic reached the U.S. Months before the lockdowns. I mean, there’s evidence in blood banks that it was here in December, 2019, which if it’s already being picked up in blood banks in December, 2019, when did it get here?
We don’t know. But that should have been something that experts were trying to track down right away, to try to understand how far it had spread. Why wasn’t the CDC doing seroprevalence studies to figure out how far it had spread in early 2020? They were very slow to get going with studies of that kind.
So I think that there was a real slowness of response in terms of policy learning over the course of the COVID pandemic. And that one lesson we can draw from that is that next time we face a crisis where there’s a lot of uncertainty about what to do and what the nature of the problem is, study it as best you can in real time, and adjust in response.
Matt Grossmann: Stephen, we have no shortage of policy problems where experts believe that they have evidence that should be heeded in policy debates, but are probably susceptible to many of the same incentives for overconfidence, and for taking advantage of any opportunities of influence that they have. So what are kind of the realistic steps you would give experts in these areas for productively being involved in policy debates?
Stephen Macedo: Yeah, so one thing I think is to be more alert to our own biases, and we all have biases, but at a time when higher levels of education tends to be more strongly associated with Democratic Party affiliation, I do think there’s a worry about our sort of even-handedness with respect to arguments, evidence coming from both sides.
There’s a very good article in the Proceedings of the National Academy of Sciences, the lead author is Corey Clark, it’s called Pro-Social Motivations for Censorship in Scientific Studies. And it details a variety of forms of evidence for the proposition that scientists and social scientists errors tend to be in the direction of their preferred outcomes, and that there’s a kind of systematic bias towards political values in arguments being made about empirical evidence.
So I think we need to be more alert to our own biases, more open to the uncertainty that’s implicit or part of many scientific findings for sure, and more aware, very importantly, that all policy recommendations involve value judgments, which we should try to be clear about, and not assume that other people necessarily agree with, that ought to be publicly debated, that we ought to be explicit about.
So I do think we find evidence of a certain amount of hubris, overconfidence, unwillingness to acknowledge uncertainty, and partisan bias running not just through expert opinion, but also a fair amount of the academic commentary on these matters.
Matt Grossmann: And Frances, policymakers aren’t going to be able to avoid the experts in many future policy debates, but there certainly is skepticism in some people saying that these people are always acting in their ideological interests and should be ignored. And some of the tips… not tips. Some of the implications of what you say could lead people to find the one dissenter in a consensus area. So what are the best strategies for policymakers to reach the happy medium here?
Frances Lee: Well, skepticism of experts claims are certainly warranted, that policymakers bear the political responsibility for decision-making, and they have to understand the basis on which they are making these decisions. And so it’s incumbent on them to ask questions of experts, and to seek out experts with divergent opinions, and to find out why the bulk of experts or scientists in an area disagree, that falls to them to take that role. It’s not an easy role.
But you can’t pretend that just because there’s majority opinion in a field that that means that that stance is correct. Sometimes the minority opinion is correct, and so you have to hear it out. And so they shouldn’t be ashamed of bringing experts before the bar of public opinion and asking them questions, and seeing how well they fare.
I mean, this is one of the more important roles for legislators, that the hearing process is just designed for this. And so to hear experts out, to understand the nature of the disputes that exist within an expert community, where the soft spots are, group think in expert communities is not an unusual phenomenon. So where they have shared assumptions that warrant probing questioning, that needs to happen before policymakers make decisions that affect the whole of society.
And in the case of the COVID pandemic, you’re talking about whole of government, whole of society. You might remember that lingo. So the responsibility was all the more great under circumstances where one is making decisions of that scope.
So policymakers should take it upon themselves to recognize that they can’t just defer to experts, that the responsibility lies with them, and that they shouldn’t hide behind experts. They should instead hear them out, question them, and then bear the responsibility directly, as we expect in democratic politics.
Stephen Macedo: The way that it’s been put is experts should be on tap, but never on top, is the line that was attributed to Winston Churchill.
I might just add one other thing to what Frances said, following right up on it, which is that we live in a time in which many on the progressive left seem to have less confidence in freedom of speech, and freedom of debate, and freedom of discussion than in the past, is much more concerned with misinformation, disinformation traveling online. But one of the things that we find in looking at the COVID pandemic is that there’s a tendency to confuse disagreement and dissent with misinformation, that those who had a minority opinion about the right policy path were cast as purveyors of misinformation. Their good faith was challenged.
When has that ever been the proper way of conducting science to censor dissenters, the minority viewpoint, the minority voice? So I think we just need to remember some of the core values of our own disciplines in social science, and liberalism, that being open to dissent and criticism is the only basis we have for the confidence we have in our own views, the standing invitation to the whole world to refute them. So we did see evidence of both groupthink, and intolerance to dissent, which I think can undermine the integrity of science, social science, and public deliberation more broadly.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available bi-weekly from the Niskanen Center, and I’m your host, Matt Grossmann. If you like this discussion, here are the episodes you should check out next, all linked on our website:
Lessons from the COVID-era Welfare Expansion
Policymakers Follow Informed Expertise
Conspiracy Beliefs Are Neither Increasing Nor Exclusive to the Right
The Politics Of School From Home
How Anxiety And Crises Change Our Political Behavior.
Thanks to Stephen Macedo and Frances Lee for joining me. Please check out In COVID’s Wake, and then listen in next time.