The economist Lyman Stone, a Twitter acquaintance and Institute for Family Studies colleague of mine, has an interesting piece up at Public Discourse called “Lockdowns Don’t Work.” It has a lot of data in it, and the question it raises is a good one: What marginal benefit do we get from forcibly shutting down business and social life, above and beyond the effects of other strategies?
No minimizer of COVID-19, Stone argues not for letting the virus rip, but for closing schools, banning large gatherings, wearing masks, limiting travel, and, crucially, using “centralized quarantine orders, where individuals who test positive or individuals who have had contact with COVID-infected people are forced to be quarantined for seven, fourteen, or twenty-one days in hotels or special-purpose spaces.” These measures seem to have squelched the outbreak early in Hong Kong, where Stone lives.
It’s possible this would have been a better approach, or that it would make sense as a next step for us. But I’m a bit skeptical of Stone’s headline claim.
First of all, while said headline declares that lockdowns don’t work, the piece waffles between that assertion and the weaker one that “there’s no evidence of lockdowns working.” Early on, Stone asks,
Why should I have to prove that lockdowns don’t work? The burden of proof is to show that they do work! If you’re going to essentially cancel the civil liberties of the entire population for a few weeks, you should probably have evidence that the strategy will work.
I don’t think this makes sense in the context of the situation we faced (and to an extent still face). We saw a new threat that, in Stone’s own estimation, could kill hundreds of thousands if left unchecked. If extreme measures can’t be tried until proven effective in such a scenario, no country can ever be the first (or, pending studies, the second or the third . . . ) to try them. We need to experiment to know what works, starting from common sense.
And the commonsense case for lockdowns is obvious, requiring no more than a grade-school-level understanding of how infectious diseases work, and sophisticated epidemiological models back it up. This is a virus that jumps from person to person. We can’t generally keep people from spreading it within their own households (if my experiences with my kids’ colds are any indication), but we can stop it from spreading between households by limiting interactions among members of different households. That’s what a lockdown is, basically, a ban on unnecessary interactions outside the home.
And then it turns out there is evidence they work. Stone writes:
The best evidence [for lockdowns] comes from Wuhan, where China imposed a strict lockdown in late January. Researchers have found that this lockdown, combined with other prior policies, as well as population behavioral changes, may have reduced the reproduction rate of the disease from 3.9 to 1.3: a big change, but not enough to prevent the spread of the disease. In Wuhan, the reproductive rate did not drop to 0.3, a level at which the disease can truly be beaten, until after centralized quarantine measures had been put in place. It was centralized quarantine that beat COVID in Wuhan, not lockdowns.
“Lockdowns can help reduce the rate of spread by two-thirds, but they aren’t enough to break this virus on their own” is a far cry from “lockdowns don’t work.” I would add that there’s also this study on stay-at-home orders from the U.S., though it focuses on case growth rather than deaths. (And I should concede that I share Stone’s concerns about the quality of the Chinese data.)
Then he looks at the death curves in various countries that have locked down. Deaths usually start to drop about two weeks after a lockdown, but Stone argues, based on numerous medical sources, that it takes 20 days for an infection to result in death — so this striking pattern can’t actually be an effect of the policy. I hope the experts he relies on have this number right, because if not, the graphs in the piece imply something very different indeed.
It’s also possible, of course, that these government lockdowns came on the tail of voluntary social distancing that started to turn things around early. There’s strong evidence of this from the U.S., including some that Stone recently posted on Twitter. (Here in Northern Virginia, it also matches my personal experience that businesses were closing and events were being canceled well before the official stay-at-home order.) If that’s the explanation, the lockdowns could have reinforced the trend and kept people in the habit of social distancing when they otherwise would have stopped, or they could have just been redundant. In the latter case, the measures were unnecessarily coercive, but lifting them won’t do much for the economy until people are willing to come back out again.
Lastly, Stone provides some general details about a model he built that predicts U.S counties’ per-capita death tolls as of April 19 based on some characteristics of the counties and what government policies they had in place. The results suggest school closings and mass-gathering bans work while full lockdowns do not. I’m inherently skeptical of these kinds of exercises, and I’m especially dubious that a model can reliably tease out the independent effects of interrelated policies that rolled out in fast succession throughout the country, rather than just confirming that the policies worked in conjunction with each other. But I hope he shares more details about the model elsewhere so it can undergo at least an informal peer review on social media.
Anyhow, at this point, practically speaking, perhaps the question isn’t so much whether lockdowns work but what the next step is when we (very soon) get sick of doing this. Even if Stone is wrong about the effects of our current strategy, his suggestions for the next one are worth considering.