Up until April of 2020, the basic case for lockdown was still intact: there’s a virus going round with a 1% mortality rate; it will continue spreading until two thirds of the population has been infected; if we don’t lock down, hundreds of thousands of people will die; therefore we must do so.
Now, I’m not saying this case was strong. After all, it completely ignores the issue of civil liberties – even if locking down had saved hundreds of thousands of lives, doing so might have still been wrong given the massive constraint on freedom it entailed. Another weakness is it assumes the only way to stop people dying was by locking down; in reality, we could have tried focussed protection.
Nonetheless, the premises at least appeared sound. This changed at the end of April 2020, when Covid deaths in Sweden began trending downward (left-hand chart below).
Aside from closing high schools and banning large gatherings, Sweden did almost nothing to halt the spread of the virus. There was no stay-at-home order, and no forced business closures. Despite this, daily deaths began falling at exactly the same time as in Britain – which had gone into lockdown at the end of March.
It’s often claimed that Swedes locked down voluntarily by drastically reducing their time spent outside. But this simply isn’t true, as Google mobility data shows (right-hand chart above). Swedes reduced their time spent on retail and recreation by only a quarter as much as Brits.
Sweden’s experience clearly undermined the second premise in the case for lockdown: that Covid will continue spreading until most people have been infected. After Sweden, it was no longer possible to argue that lockdown was necessary to halt the spread. You could still argue it made a difference, but not that it was the only thing standing between us and armageddon.
Which raises the question: why did so many people continue calling for lockdown? Why didn’t it dawn on them that the cost/benefit ratio was much greater than they’d initially believed? And I’m not just talking about joe public, who massively overestimated the risk of dying. Many ‘experts’ who were perfectly aware that the risk was low continued banging the lockdown drum.
Three reasons come to mind.
First, the benefits are concentrated (on those who are protected), whereas the costs are widely dispersed. Second, the benefits are immediate, whereas the costs are largely delayed (inflation, government debt, worse educational outcomes). Third, measuring the benefits is easy (or at least, was presented as being easy), while measuring the costs is somewhat harder.
For these reasons, I suspect, the costs were more difficult for people to “see” – in the sense of Frédéric Bastiat’s essay ‘What is seen, and what is not seen’. It’s not just the magnitude of costs that affects decision-making, but also their ‘visibility’.