Rejoinder

The response by David Goldsmith to our recently published Economic Affairs article (Andersson & Jonung, 2024) is puzzling. Most prominently, Goldsmith stresses that the correlation between the stringency of lockdowns and excess death rates is close to zero. This is exactly the main point in our article.

If lockdowns were a clear success, there should be a significant correlation between the extent of lockdowns and excess deaths. But there is none. The least one should expect from lockdown supporters is that they would be able to show how the non-existing significant correlation we find can be turned into the significant negative relationship they claim to exist. Goldsmith has arguments but no analysis on this point.

With this insight, we hope that he shares our message that Sweden was a success story in the sense that the use of more far-reaching lockdowns in other countries than in Sweden had no apparent benefits.

Goldsmith raises the question: what constitute a successful pandemic policy? Many argue that the only way to success is to arrest the spread of the virus by the use of lockdowns, ignoring all other social, political and economic consequences. We do not share this limited view. When evaluating public policies we must consider their full impact on society. Nevertheless, even if one adopts a limited approach, draconian lockdowns were obviously not a success as they had no significant effect on excess mortality.

In our analysis we adopt a broader approach by looking at the effects of lockdowns on the macroeconomy as well. Here we find a significant negative impact of lockdowns. For example, real GDP in the United Kingdom fell by 20 per cent in the second quarter of 2020, compared with 9 per cent in Sweden. In the United Kingdom, public finances and budget deficits are now a matter of general concern as a result of the fiscal expansion during the lockdown period. In Sweden, the public debt accumulated during the pandemic has been paid off. Instead, there is a fear that the public debt may have become too low, the contrary case to the UK.

Finally, perhaps the biggest weakness in Goldsmith's argument is his failure to recognise the role of voluntary behaviour. There are two alternative ways for a government to respond to a pandemic. It can either advise the public on how to adjust its behaviour and trust citizens to act responsibly, or mistrust citizens by enforcing mandatory restrictions backed up by law-enforcement agencies. The latter approach assumes people are ignorant and incapable of making independent decisions.

If the next pandemic has a case fatality rate of 10–20 per cent, as Goldsmith speculates, people will go out of their way to avoid contact with others. This is a key point from the Swedish experience, where the authorities relied to a large extent on voluntary adjustment. It is a crucial factor in explaining why Sweden could display low excess mortality and low costs to society of its pandemic policy. As each citizen's life situation is different, the voluntary approach is likely to be more efficient in combating future pandemics. Each individual could design his or her own measures to reduce social contacts while maintaining as normal a life as possible. The alternative approach of social control from the top down is far less efficient as it is based on the view that the government knows all.