Is there a trade-off between health and the economy?

Throughout the pandemic, people have claimed that there is no trade-off between health and the economy. This claim has usually been made in the context of whether to impose lockdowns, the argument being that lockdowns are good for both public health and economic output. Here I want to argue that the claim is true only under very specific conditions, and that it is unclear whether any of the hardest hit Western countries would have been able to bring those conditions about.

Note that I am not arguing against lockdowns, or that the economy is more important than health. I am merely arguing that most Western countries have faced a trade-off between health and the economy. (My argument is therefore consistent with the belief that one should favour lockdowns on the grounds that any costs imposed on the economy are outweighed by the benefits to public health.)

Furthermore, this article takes as a premise that lockdowns do benefit public health in the short-to-medium term insofar as they reduce viral transmission, and thereby prevent healthcare systems from becoming overwhelmed. I recognise that this claim is contested, but for the sake of argument, I will assume that it is true. (To me, it seems very difficult to believe that lockdowns do not have at least some effect on viral transmission.)

Quotes from proponents of the argument

Various academics and journalists have argued that there there is no trade-off between health and the economy. What’s more, in the UK, both Rishi Sunak (the Chancellor of the Exchequer) and Chris Whitty (the Chief Medical Officer) have made this argument. Back in April of 2020, Sunak said:

The single most important thing we can do for the economy is protect the health of our people … It’s not a case of choosing between the economy and public health – common sense tells us that would be self-defeating … The absolute priority must be to focus all our resources … in a collective national effort to beat this virus.

Likewise, in an interview with The BMJ in November of 2020, Whitty was asked:

Recent data published by the Financial Times show that countries that have done worse in terms of covid deaths have also done worse economically, with the UK scoring badly on both counts. This suggests that it’s a false dichotomy to talk about health versus wealth and that we have to control the virus to restore our economy. Is that how you see things?

And he replied, “That is absolutely how I see things.” Hence it is not only political commentators, but also key advisors and decision-makers, who have claimed there is no trade-off.

When the argument does work

A lockdown may be bad for the economy in the short run. However, it could be good for the economy in the medium run if it reduced viral transmission to such an extent that the reproduction number fell below 1. Once this had been achieved, and the number of cases had been brought down significantly, further viral transmission could be prevented by testing and tracing infected individuals. New cases being brought in from outside could be intercepted at points of entry. After the virus had been contained, the lockdown could be lifted, and economic activity could continue.

As a matter of fact, several countries have managed to contain the virus using more-or-less these measures. For example, New Zealand has not had more than 31 new cases per day since early April, while Taiwan has not had more than 27 cases per day for the entire duration of the pandemic. By contrast, France reported more than 100,000 cases in a single day in early November.

However, the argument only works – so far as I can see – if the lockdown enables you to completely suppress the virus. Once complete suppression has been achieved, the lockdown must be combined with a well-functioning test-and-trace system, and a well-functioning system of border quarantines. In the absence of these additional measures, a new epidemic will almost certainly emerge once the lockdown is lifted. (And if the lockdown is not lifted, then it will be bad for the economy in the medium run, as well as the short run.)

Evidence that lockdowns are bad for the economy

One might say it is obvious that lockdowns (of any duration) are bad for the economy. However, it has been argued that the reductions in economic activity over the past year would have happened even in the absence of lockdowns. This is because, so the argument goes, people would have changed their behaviour to the same extent regardless of what policies governments imposed. (Of course, if this argument were true, lockdowns wouldn’t really be necessary.)

As a starting point, I checked to see whether there is a negative relationship between GDP growth and lockdown duration across countries. Data on real GDP growth were obtained from the IMF. And data on lockdown duration were obtained from the Oxford Blavatnik School’s COVID-19 Government Response Tracker. Specifically, I used the number of days of mandatory stay-at-home orders, and the number of days of mandatory workplace closures (for some sectors). In both cases, the number corresponds to days in which the relevant measure was imposed on any part of the country (either one or more regions, or the entire country).

Note that the contraction in GDP last year was much larger than the contraction associated with the global financial crisis. In 2009, world GDP contracted by 0.1%. Last year, it contracted by 4.4%. In 2009, advanced economies contracted by 3.3%. Last year, they contracted by 5.8%. In 2009, UK GDP contracted by 4.2%. Last year, it contracted by 9.8%. These figures are subject to revision, but you get the idea.

The chart below shows the relationship between GDP growth and days of mandatory stay-at-home orders in 2020. The left-hand panel corresponds to all countries, while the right-hand panel corresponds to countries with GDP per capita > $25,000. Two outliers were removed from the analysis: Guyana, which reported GDP growth of 26.2% (the next highest was 4.1%); and Libya, which reported GDP growth of –66.7% (the next lowest was –25%). Both relationships are negative and significant (left-hand panel: r = –.23 p = 0.003; right-hand panel: r = –.46, p = 0.001).

The chart below shows the the relationship between GDP growth and days of mandatory workplace closures (for some sectors). Once again, both relationships are negative and significant (left-hand panel: r = –.21 p = 0.007; right-hand panel: r = –.31, p = 0.024). Note that the circles are proportional to countries’ population sizes, but the slopes are unweighted. Of course, I’m not claiming this is anything other than a purely associational analysis. But it is noteworthy that the associations are all negative, as one would expect if lockdowns reduced economic activity.

More compelling evidence can be found in the scholarly literature. Indeed, numerous papers have found that lockdowns reduce economic activity, some using designs that allowed the authors to distinguish the effects of lockdowns from those of voluntary social distancing. Moreover, in a recent survey, academic economists were asked about the economic impact of lockdowns in the UK, and a majority said there is some trade-off. The first question was:

How much of the decline in GDP experienced to date would have been avoided in the absence of any lockdown measures or other policy interventions (such as fiscal support)?

27 respondents answered the question. Of those, 44% said “A small portion of the decline”, 19% said “A moderate portion of the decline”, and 15% said “A substantial portion or the entire decline”. By contrast, only 19% said “GDP would have been lower absent lockdowns”. The second question was:

How much will the new lockdown measures introduced on Thursday November 5 hurt UK economic activity this year relative to a counterfactual with the milder measures adopted over the summer?

26 respondents answered. Of those, 54% said “Small damage”, and 23% said “Large damage”. By contrast, only 19% said “The economy will benefit from lockdown”, and only 4% said “No additional damage or benefit”. The third question was:

Using not only the policy tools that have been part of the UK policy mix thus far but also policy tools implemented in other countries, to what extent does the government face a tradeoff between saving lives and preserving livelihoods?

34 respondents answered. Of those, 35% said “Small tradeoff”, and 30% said “Large tradeoff”. By contrast, only 35% said “No tradeoff”. Hence, in this (non-representative) sample of academic economists, most believed there there is at least a small trade-off between health and the economy. Note: I am not claiming the trade-off between health and the economy is necessarily large, only that it exists.

Could Western countries have suppressed the virus?

I noted above that there might not be a trade-off between health and the economy if locking down enabled you to completely suppress the virus. In most large European countries, this almost certainly would not have been possible during the first wave, since case numbers were already too high by the time governments decided to impose lockdowns. (Which, again, is not to say that imposing lockdowns was necessarily the wrong policy.)

However, daily case numbers in countries like the UK, France and Italy were in the triple digits for most of the summer, so it is conceivable that complete suppression might have been achieved at that time. But could it have been sustained? In other words, could Western governments have actually prevented new epidemics from emerging once suppression had been achieved. There are several reasons to believe that they could not (although the possibility cannot be ruled out).

In a previous newsletter, I noted that – among rich countries with large elderly populations – only 9 managed to keep COVID-19 deaths per million below 100 last year: Taiwan, Singapore, New Zealand, South Korea, Japan, Australia, Norway, Iceland, and Trinidad and Tobago. In others words, no Western country other than New Zealand, Australia, Norway and Iceland – all of which enjoy obvious geographic advantages – managed to contain the virus.

In addition, there is at least one rich country that had both closed borders and contact tracing in place for much of 2020, but still didn’t manage to contain the virus. Specifically, Israel implemented a total border closure on 12 March (which lasted all the way until 15 August) and introduced comprehensive contact tracing on 17 March. (These figures were taken from the Oxford Blavatnik School’s database.) Despite reporting two-digit case numbers for most of May, another epidemic had already gotten underway by early July.

The fact that every country in Europe and North America other than Norway and Iceland has reported several hundred COVID-19 deaths per million suggests it may not have been possible for major economies like the UK, France and Italy to contain the virus at any point. One potential counter-argument is that the four East Asian countries mentioned above did manage to contain the virus. So perhaps if the UK, France and Italy had replicated their policies, complete suppression really could have been achieved.

Of those four East Asian countries, Japan is the most puzzling. During 2020, it had zero days of mandatory stay-at-home orders, and zero days of mandatory workplace closures. It did not introduce comprehensive contact tracing until 17 November, and did not implement a total border closure until 22 December. Despite this lax approach, it has reported only 40 COVID-19 deaths per million as of 24 January – less than every country in Europe, including Norway and Iceland.

Cases and deaths have been on the rise in Japan, so the situation may change going forward. But given how little was done until recently, the country’s “success” in dealing with COVID-19 may owe to factors other than government policy. The various possibilities include: more favourable climate; less individualistic behaviour; more pre-existing immunity; and lower vitamin D deficiency. (These factors may also partly explain the success of South Korea, Taiwan and Singapore.)

Japan’s experience with COVID-19 suggests that the “East Asian model” may not be replicable in most of Europe and North America – particularly in large, highly connected countries like the UK, France and Italy. On the other hand, the UK is an island. So perhaps if it had implemented a total border closure (or extensive border quarantines) as well as a comprehensive test-and-trace system, it would have been able to contain the virus from the late summer onwards.

Conclusion

Various people have argued there is no trade-off between health and the economy, meaning that lockdowns are good for both public health and economic output. However, this argument only works – so far as I can see – if locking down enables you to completely suppress the virus. Only once complete suppression has been achieved can economic activity continue. While some Western countries (such as Australia and New Zealand) have managed to contain the virus, it is unclear whether this was ever a realistic goal for major economies like the UK, France and Italy.

Except perhaps at the end of the summer (when daily case numbers were below 1,000) most Western countries have faced a trade-off between health and the economy. Hence those that imposed lockdowns ended up prioritising the former over the latter. This may well have been the right decision, since the health benefits may have been large, and the economic costs small (relative to a counter-factual where the pandemic still caused a major recession). However, those economic costs were not zero. And they may have been quite sizeable.

Image: Fritz Melbye, Ships on stormy seas


COVID-19’s lethality in England

Yesterday, I wrote a piece for Quillette comparing COVID-19’s lethality in England to that of the Spanish Flu. Some of the points will be familiar to regular readers of this newsletter, but the piece does include a new graph. Here’s an excerpt:

In order to separate out the deadly effects of the Spanish flu from the deadly effects of the First World War, it is helpful to look at countries that remained neutral during that conflict. One such country is Spain. In fact, the disease became known as the “Spanish flu”—a misnomer—precisely because Spain’s leaders, unlike those of other large Western nations, did not feel compelled to censor early reports of the disease as a way to maintain morale.


Thanks for reading. If you found this newsletter useful, please share it with your friends. And please consider subscribing if you haven’t done so already.