Flavio Toxvaerd: ‘We can’t tackle Covid unless economists and epidemiologists talk to each other’

This is part of a series, ‘Economists Exchange’, featuring conversations between top FT commentators and leading economists about the coronavirus economic recovery

Two years since the coronavirus pandemic first engulfed the world, there is still no consensus on how best to marry effective disease control with economic success. China insists a zero-Covid strategy is best, while the mantra in most advanced economies is to maximise vaccination rates and then learn to live with the disease. Many emerging economies would love to have these sorts of choices.

Ever since the pandemic started, the economic community has wrestled with the fundamental possible trade-off between lives and livelihoods at the same time as worrying whether controlling the virus was best for both. Most in the discipline had not thought about these problems. From the scientific and medical side, there was a similar lack of understanding over the economic impacts of the measures they proposed. The two camps had not spent much time talking to each other.

However, some academics have devoted their careers to precisely the nexus between these issues. One of the most prominent in the field is Flavio Toxvaerd, associate professor at Cambridge university. He first became interested in the field after reading a paper on the economics of rumours, by the 2019 Nobel laureate Abhijit Banerjee. A footnote in the paper highlighted similarities in the proliferation of both rumours and infection, and Toxvaerd was intrigued.

More than a decade ago, he joined a small band of economists and epidemiologists who had studied links between the spread of a disease and strategies to control it through vaccines, treatment and social distancing. As the Covid-19 pandemic has gathered pace, Toxvaerd has never been busier.

Chris Giles: When you’re combining economics and epidemiology, what is the big insight?

With Covid-19 we know that behaviour is very important

Flavio Toxvaerd: First of all, it’s useful to think about different kinds of diseases. There are some for which behaviour is not important. Something like athlete’s foot is an endemic disease and people could take measures to protect themselves, but on the whole they don’t go out of their way to change the way they live their lives in order to avoid it. So, this is a disease in which behaviour is not very important.

With Covid-19, unfortunately that’s not the case and we know that behaviour is very important. People take measures to prevent infection. So I would say that there are two major ways in which economics can inform disease modelling and public policy. The first is to better understand the role of behaviour and use a toolkit of economics to better inform policy innovation. The second is to use an analytical framework with which we can evaluate different policy measures and determine, quite frankly, whether any intervention is worth it in the first place.

CG: Let’s stick with Covid-19 for now. What do you think we know about modelling people’s behaviour in the face of the virus and what does that suggest for the best way to think about policy interventions?

FT: If somebody were to ask me before the pandemic, what are the things we need to think of and how can economics help us think through those different choices, I would suggest making a credible cost-benefit analysis. In the same way that if, for example, we want to build a new bridge, we would go through a specific set of analyses that will help us guide that choice.

For example, how can we value lives, how can we trade off lives and livelihoods, how do we deal with uncertainty and so forth? In most countries, there is already a very well-developed framework but for some reason, it has not been applied to Covid-19.

CG: As the pandemic escalated in early 2020, how would you frame that cost-benefit analysis? What should we have been thinking about at the time?

FT: Ideally there should have been an urgent recognition that rather than looking at health versus wealth, which was kind of the trade-off that people were touting and using on either side of the debate, it would have been useful to think immediately in terms of the health and wealth consequences of some decisions versus the health and wealth consequences of other decisions.

Both the economy and the pandemic are extremely complicated, dynamic objects of analysis. What you do to change either the economy or the pandemic today will influence the world you face tomorrow, and therefore will affect what we can or should do in the future. So, rather than thinking at each stage, “we need to safeguard the economy”, it would have been useful to realise that this is a very integrated system and policies that might look, on the face of it, to be useful for the economy today, can backfire and make things worse tomorrow.

CG: Are you saying that allowing the spread of the virus to be very rapid at the start led to a worse economic outcome than clamping down on interactions quickly?

In most countries, there is already a very well-developed framework [for cost-benefit analysis] but it has not been applied to Covid-19

FT: Right. When we’re building a bridge, the cost-benefit analysis is the wealth improvement of construction compared with the impact of doing nothing. In the case of a pandemic, we have a rapidly changing situation and so if we do nothing, people are obviously going to behave differently. They are going to have incentives themselves to self-protect and that might itself cause a lot of destruction to the economy.

So, unfortunately, people have been caught out by saying things like “lockdowns have created such and such damage” without realising that the real benchmark was not the world we faced before the pandemic, but a world in which we did nothing and the disease would spread and people would voluntarily change behaviour and that itself would also cause damage.

I think a useful metaphor would be a patient who’s undergoing chemotherapy. You can say that chemotherapy is bringing terrible side-effects to a patient relative to a healthy individual but that’s, of course, not the right comparison. The comparison is, what would the patient be going through in the absence of treatment?

CG: Can we take this to the real world during the pandemic. There were several experiments going on, with countries doing things differently. Sweden versus Denmark would be one. What did these governments get right or wrong, in your view?

FT: It’s extremely difficult to compare across countries for the obvious reasons. But it’s useful to take an example like Denmark and Sweden, which are similar, both theoretically, culturally, historically and also in terms of size – both have very small populations.

There are two things that I want to point out. Let me just preface it by saying that Denmark eventually put in quite strict lockdown measures whereas Sweden notoriously did not. The first thing to note is that in the first week or so before the lockdown measures were imposed in Denmark, the behaviour response from people in Sweden and Denmark were very similar.

We know from looking at Google mobility reports that the time spent in public places like supermarkets and public transport went down both in Denmark and in Sweden. So that is already strong evidence that there was actually a voluntary change in behaviour, even before lockdown measures became relevant.

What we’ve seen since the lockdown measures were imposed is not controversial. I think it’s fair to say that Denmark has done better overall, not just in terms of cumulative deaths per head of population, but also economically. There are papers that have looked at expenditure decreases in both countries after the pandemic started, and they are actually comparable for Denmark and Sweden, which goes to show that a big chunk of the change in economic activities is actually attributable to changes in behaviour rather than to lockdown policies.

CG: When you’re thinking about the advice governments received, particularly early on in the pandemic, how could the structure of that advice have been improved?

There is strong evidence that there was a voluntary change in behaviour, even before lockdown measures

FT: As soon as it became evident that the disease was sufficiently serious to change behaviour and, therefore, have really numbing impacts, of course the government, here in the UK and in other countries, realised they had to manage both the disease and the economy.

And so economists were brought on board to try to manage the economic impact. My complaint has been from the outset that those people did not seem to talk much to each other, and that gave rise to unforced errors because people who were looking at the problem from a purely economic perspective might have made perfect sense but from an epidemiological and public health perspective, their ideas were just a non-starter.

Unfortunately, economists have not communicated clearly enough what it is that they do. So many people thought economists were going to say, “oh, we have to get the economy back to functioning normally”. Instead, there has been a quite broad consensus from economists that the best way of managing the economy was to get the pandemic under control.

CG: That was the initial policy choice before we had vaccines. How does the arrival of vaccines change the picture and policy recommendations?

FT: You can think of vaccines as a substitute for social distancing measures or for lockdowns. And the reason is that they are both instruments or policies that slow down the spread of the infection. Now, as we know, lockdowns are incredibly costly. They have social costs because people can’t see their loved ones. They have costs in terms of missed schooling and, of course, economic disruption.

So even though lockdowns may be necessary in some circumstances, I don’t think anybody should consider this is a good instrument. It’s a very blunt instrument. And what vaccines allow us to do is to substitute for those measures that are very costly and instead allow us to resume some kind of normality.

What they also do, though, is to change individuals’ incentives to behave differently according to shifts in risk. What we find is that when people get vaccinated, it affects their behaviour in ways that are not entirely beneficial. For example, there was one study that showed a sudden spike in infection in those who had had one jab. That was ascribed to people being under the illusion that protection would kick in immediately.

CG: Does this mean there is a risk, especially if the efficacy of the vaccine against onward transmission is low, that vaccination could spur a change in personal behaviour that might actually increase the spread of the virus?

FT: The interesting thing is that it can induce behaviour change, both in people who are vaccinated and those who are not. So the people who are vaccinated might think they are no longer at risk and let down their guard. But people who are not vaccinated might, likewise, think, “If everybody else is vaccinated, I am less likely to be infected and so I can also relax.”

The people who were strongly anti-lockdown are saying now we have the vaccines, we should just open up, because they are thinking exactly along those lines. And this can be dangerous.

CG: How does the way of thinking change when we now have some effective therapeutic medicines to combat Covid?

Vaccination can induce behaviour change, both in people who are vaccinated and those who are not

FT: There are two main differences between therapeutics and vaccines. First of all, vaccines are given to people who are not yet infected, whereas therapeutics are given to people who are already infected. So, that difference raises a number of issues.

With therapeutics, the individual calculation is different because once you have the disease, it doesn’t really matter what other people do because it’s just you who has got the disease and you have to live with the consequences. The interesting thing about therapeutics is that it may lead to something called behavioural disinhibition, which means if the severity of the disease is lower, you have a weaker incentive to try to avoid becoming infected in the first place.

This is a concern that has been raised in many other contexts. So, for example, when the morning-after [contraception] pill was released, there were voices that were complaining, “hang on, people should be protecting themselves in the first place and not rely on the morning after pill to interrupt a pregnancy” and therefore that could actually give the wrong incentive. The mechanism for how therapeutics could change people’s behaviour and incentives is clear, but whether this is something that will actually happen is still not known.

CG: How does this help us consider questions such as whether China should stick with its zero-Covid policy?

FT: China has a very strong zero-Covid policy and it’s inevitable that at some point they will have to open up. The question is really how do they do that. I think it’s fair to say that Australia will probably rely upon high-quality vaccines like the ones that are used all over western Europe and the US — Moderna, Pfizer, and AstraZeneca.

China seems unwilling to use those vaccines and has relied on its own vaccines that are less effective. And so it’s going to be scary but interesting to see how it manages that transition, because if you have effective vaccines, you roll out protection, and you can then open up the country because the consequences of spreading the disease are smaller.

If you have an imperfect vaccine, that calculation becomes much, much more complicated and because they have so many people that are susceptible to the disease, that could become, if mismanaged, an unmitigated disaster.

CG: Another issue that’s controversial around the world is vaccine mandates. How should countries decide between carrot and stick effect when encouraging people to have a vaccine?

FT: There are several considerations here. There are purely economic or incentive considerations, and there are moral and political considerations that overlap. From a purely disease-management perspective, it’s one more tool in the box to try to avoid contacts between people who are potentially infected.

Now, people object to vaccine incentives for reasons that are not entirely clear to me. What countries have been doing is to provide indirect incentives to encourage people to adopt certain behaviours. And this is something we do all the time. We banned smoking in public places because we recognised that there were important health impacts on third parties. I don’t quite accept the premise that people have absolute freedom to decide everything about their own lives.

This is not true in a democracy and I also think that it’s misplaced to think it applies exclusively to the rights of people not to vaccinate themselves. Vaccinated people also have a right not to be exposed unnecessarily to infection. So, I think the old maxim that they do not want to force people to do things unless they actually hurt third parties is perfectly applicable here.

I think it’s a good libertarian principle that we should, as much as possible, stop people from hurting others and I think that refusing to get vaccinated can be seen in those terms, although the narrative is usually not presented that way.

CG: When you think about how economics and epidemiology interact, clearly you are able to make public policy recommendations. Does this mean you’re always going to get them right?

It’s inevitable that, at some point, China will have to open up. The question is really how do they do that

FT: Certainly not! Somebody recently said that economics is a science in the same way that medicine is a science. We know, for example, that smoking will make people worse off but that doesn’t mean that if you stop smoking you will not get lung cancer. So, we cannot always predict exactly what will happen, but we will be wrong fewer times than otherwise would be the case. So, what economics does at a minimum, is that it avoids us shooting ourselves in the foot.

For example, one ill-conceived policy that I think most economists were quite unhappy with was the UK’s Eat Out to Help Out policy. This was brought in, of course, to help the hospitality sector because it was in such a bad way. So the Treasury thought, “well, let’s support them” and the way they thought of doing so was to incentivise people to go out to restaurants.

Now, any economist who had thought through this problem might have said to them, “maybe you should just send them a cheque or just insist that it’s takeaway only, because we don’t want people to mingle unless it’s necessary”. So, that, I think, was an unforced error that could have been avoided if they’d thought more like an economist would.

CG: When you’re looking at the global situation now, how optimistic are you in terms of both the disease and the economic situation? What’s your assessment of how things are now?

FT: I think there are still very difficult problems resulting from the fact the vaccine rollout hasn’t been equitable, it hasn’t been uniform across the world and I think that is a decision or an omission that will come back to bite us. Because my understanding is that the more infection there is, the more likely it is that new variants will develop. I still think we are not yet out of the woods. But I think we are much better off than we were just a year ago.

We have tools in our toolbox that were not available to us before, but it’s way too early to declare victory. There are still very difficult things to navigate, so I’m cautiously optimistic but I’m not unrealistic either.

The above transcript has been edited for brevity and clarity