On "Two Weeks to Slow the Spread"
It wasn't a valid strategy for defeating the coronavirus, but it did succeed in not overwhelming the American health care system.
We're one year into the COVID pandemic that caused governments everywhere to flex the limits of their emergency powers to impose restrictions on business operations, travel, gatherings and more. (If you want to be formal about these restrictions, you'll call them non-pharmaceutical interventions, or NPIs. Colloquially, people have referred to all forms of NPIs as 'lockdowns,' which we'll do here.) If you're like me and poisoned by internet-media brain, you'll have seen a phrase that circulates when news about some of these restrictions hits: "two weeks to slow the spread." This is a reference to how a lot of these lockdowns were initially sold as temporary restrictions - perhaps two weeks - so that the government could get a handle on the pandemic, and we could go back to life as normal.1
Fifty-two weeks later, a lot of different kinds of lockdowns are still in place throughout much of the country.* So "two weeks to slow the spread" looks like foolish optimism at best and at worst an evil manipulative lie by America's leading health authorities.
"Slow the Spread". Or "Flatten the Curve".
As COVID spread through the United States in February and March, the message presented by health authorities was "flatten the curve." Presented as below, the idea is that a short, quick spike would result in more total deaths because health care infrastructures - hospitals, beds, ventilators staff and more - would be overwhelmed and wouldn't even be able to treat the treatable cases due to the volume and velocity of infections. Stories from Italy suggested this was indeed the worst-case scenario, as their health system did get overwhelmed and they were for a long time the worst-hit country in the world.
To pick on someone whom I greatly admire and respect, Jim Geraghty at National Review this week wrote about the failure of "two weeks to flatten the curve." His is merely the most recent but it echoes an argument I've seen elsewhere: that flattening the curve was a failure when it abjectly was not.
The United States' lockdown and mitigation strategy did, by the measurement of flattening the curve, work. New York City and the surrounding areas were hit hard, but much of the rest of the country was not. (And even in New York, it's unclear if health system capacity was truly overwhelmed.) The U.S., compared to the big countries of Western Europe, did flatten the curve. We experienced two mini-peaks - one in the Spring and one in the Summer - rather than one massive peak in the Spring. Despite the snark now, if the goal of American health authorities in March 2020 was to flatten the curve, then they were successful.
Tim Carney at the Washington Examiner wrote similarly last May:
The data actually suggest, however, that the rest of the United States has succeeded at the assigned task, while New York has failed. That is, the U.S. minus New York has flattened the curve.
We Got Waves 1, 1.5, and 2
Unfortunately, the curve did not stay flat, or flattened. The wave that began to hit the United States in October and crested in January was devastating, and worse than anything that occurred in the U.S. in the spring or summer. And the U.S. cumulative death toll currently stands as one of the highest in the world, having surpassed some of the other large nations of Western Europe. But at no point did the peak death toll in the U.S. match the worst of times in Italy, France, Spain, or the U.K.
There could be a lot of different reasons for this, ranging from weather to population age to density to average household size. And more. It's likely that, outside of a small number of very effective but draconian kinds of lockdowns, spread of COVID is mostly exogenous to the restrictions actually implemented in the U.S. As Geraghty hypothesizes:
This was a really contagious virus, and while governors would like to pretend their executive orders and emergency laws made the biggest difference in lowering transmission rates, it looks like population density and size were among the biggest factors — the more people living closer together, the easier the virus could spread. A year on, Hawaii ranks the lowest of the 50 states in deaths per million; being a far-off, isolated island with minimal interstate travel helped. Vermont, the 49th-most populated state and 30th-most dense one, ranks second lowest in deaths per million. Alaska, the 48th-most populated state and least densely populated one in the union, ranks third lowest in deaths per million residents. Living in places with lots of wide-open spaces offered a lot more than six feet for social distancing.
This makes intuitive sense: the virus is contagious, and density must matter. But it doesn't bear out in the data: there is a weak at best correlation between density and death toll:
There are lots of states that are dense, but with a low death toll. There are plenty of states that are sparsely-populated, but with a high death toll. For every Alaska and Vermont, as Geraghty mentions, there are a Mississippi and North Dakota: low-density states with high death tolls.
But because I was curious, I took a look at another tangent of the density idea: states with high rates of urbanization, as defined by the Census Bureau. Alaska is sparsely-populated, but for the most part the people in Alaska are clustered into the cities that they do have. The same goes for states like Nevada and Utah. The correlation is stronger than strict population density - but still quite weak:
There are a lot of intuitive correlations that don’t pan out. I’ve hypothesized about liberal democracy’s relationship to COVID (didn’t pan out when I looked at the data); it could be weather, or population demographics, or (stated) mask compliance. None of these turn out to be particularly strongly correlated with COVID outcomes.
So while many outcomes may be exogenous to any policy proposals undertaken by governments, we still don't know what those exogenous factors are. And that's the point of why complaining about the "two weeks to flatten the curve" rhetoric: one year in, we still don't know what exogenous factors are causal. We only do know that our health care systems didn’t get overwhelmed, and that lockdowns help to slow spread of the infections.
Lockdowns of March 2021 Are Different From the Lockdowns Of March 2020
The snark about "two weeks to slow the spread" ignores that lockdowns in March of 2021 look completely different than lockdowns in March of 2020. In the early days of the pandemic, not much was known about the virus or why it spread, and everything was shut down. Restaurants, bars and gyms yes, but also public parks, beaches, and nonessential retail. My birthday is in April, and in April of 2020 in Washington, D.C., it was impossible to do anything but celebrate my birthday with dinner at home, without any other visitors. In April of 2021, I'll be able to gather with friends outdoors at a patio bar, indoors at a number of restaurants, or have a picnic, or go to the beach, or a large number of other activities that were restricted last April. We've learned a little bit, and our response has adjusted accordingly. Outdoor activities are low-risk, even when gathering with other people and even with strangers. Indoor gatherings are higher-risk, and indoor gatherings in poorly-ventilated spaces in contact with strangers where wearing masks is inconvenient if not impossible are about as high-risk as it gets.
Restaurants were allowed to open for outdoor seating in D.C. at the end of May 2020, and have never been closed since. Parks and recreation opened around the same time, and the outdoor exercise equipment near where I live opened in June. My read of the data is that it’s pretty silly that these things were closed for as long as they were, but as we learned more about the virus we knew we had little to fear from those activities. None of the outdoor activities have closed or even faced more stringent restrictions since then.2 Indoor gatherings and activities have faced rolling restrictions depending on how bad some of the spread has been, but that makes sense: indoor activities are higher-risk.
The best way to think about the idea of "two weeks to slow the spread" is to think about just how locked-down the world was in March and April of 2020. We're a long way from there. Restaurants and gyms are open. Kids are for the most part going back to school. Every sports league has played their season, and many of them with fans in stands indoors.
The U.S. almost everywhere shut down harshly for a few weeks in March and April 2020 to slow the spread, and it worked. What was mistaken was an understanding of just how resilient the coronavirus would turn out to be. It's not quite seasonal, like the flu: it spread powerfully throughout the summer in some of the hottest places in the U.S. It's not something that will go away with rigorous personal responsibility: it spreads asymptomatically. It's not cleaned away with lysol and bleach: it transmits almost solely in the presence of other people.
The U.S. regime of lockdowns, when compared to outcomes of other major liberal democracy, resulted in outcomes that did not peak as harshly and saw a slower spread throughout the population over time: exactly as the policy goal of "flattening the curve" would have been expected to. The lockdowns did not eliminate the virus. The only major liberal democracies that have successfully suppressed the virus in meaningful ways implemented truly draconian tactics in order to do so.
So while some people may feel betrayed by the "two weeks to flatten the curve" rhetoric as we enter week 53 facing restrictions that still exist on individual behavior and businesses, it's not because the policies didn't work to a certain degree, and it's not because we're still facing the same lockdowns that we were facing in March 2020. We’ve learned a lot in a year, and the U.S. used the time of extremely restrictive lockdowns in March and April 2020 to slow the spread of the virus and adjust the NPI response accordingly. The biggest piece of information that we learned is that the coronavirus has been much harder to completely suppress than anyone predicted.
Perhaps another time we can look at the question of lockdown efficacy but not here. Lockdowns do work to varying degrees to slow the spread of the virus through the population.
We can complain about California's insane policy of closing outdoor activities, but that's a separate issue. Health authorities have been frequently wrong and the tactics they choose frequently stupid.