The case for proactive lockdowns

Imagine, if you can, this time last year. Life was good, or at least normal.

Then history takes a peculiar turn. Your government decides to forge an “elastic nation” to prepare for life in the future. As a first step, it abolishes the school calendar — no more lazy summers or spring breaks. Instead, every Sunday afternoon, it announces whether schools will be open for the coming week. The only guarantee is a total of 39 school weeks per year, just as before, to safeguard children’s education.

Teachers and parents rise to the challenge and politicians are proud of their success. So the government proceeds to the next phase, abolishing the seven day cycle of work days and weekends. No more Saturday lie-ins or Monday blues. Instead, every day at 8pm, it announces whether tomorrow will be a work day. The only guarantee is a total of 21 work days per month, just as before, to protect the country’s economy.

This time there’s some grumbling but businesses and workers learn to adjust. So the government moves to stage three, abolishing fixed working hours. No more 9-to-5. Instead, at the start of every daytime hour, it announces whether work will be allowed. Anyone caught laboring at the wrong time will have to pay a steep fine. The only guarantee is a total of 9 working hours per day, just as before, to maintain employee productivity.

Sounds insane? In the age of Covid-19, this is how we have chosen to live.

You can’t escape R

We now know that, in most Western countries, masks, hand washing and social distancing can slow but not stop the pandemic. Despite these measures, the reproduction rate R remains stubbornly above 1.0. This means that, on average, each infected person infects more than one other, so the numbers grow relentlessly and exponentially. Within a few months, a few daily cases turn into a tsunami, putting health systems in crisis and leaving thousands of dead.

After their first lockdown ended, this scenario played out in Austria, Belgium, Canada, France, Germany, Israel (where I live), Italy, Netherlands, Poland, Portugal, Romania, Spain, Switzerland, UK, many US states and countless other places. So what do our governments do? At somber press conferences, they announce a second lockdown. Schools, offices, stores and restaurants close. Children can’t visit their grandparents. The other life begins.

Mercifully, we also now know that lockdowns work. Within a few weeks, the number of cases will start to fall. Soon after, the death rate will follow and coronavirus wards will empty and close. The restrictions will ease, parents will eagerly send kids to school and restaurants will serve families grown weary of home-cooked pasta. Friends will catch up, colleagues will reconnect, the generations will reunite. It won’t be perfect — still no theaters or concerts — but overall, life will be good again.

Until it’s not. Because we can’t escape R. Time and again.

The virus doesn’t care that we’re coming out of a two-month lockdown. It doesn’t know how few cases remain. Covid-19 is a bag of chemicals which mindlessly does its thing. The moment lockdown ends, R goes right back to where it was before, and the epidemic starts growing again. The countdown to the next lockdown has already begun.

The Covid-19 ladder

If lockdowns are ridiculous, they are also necessary. For as long as R is above 1.0 outside of lockdowns, we inevitably need to be in lockdown some of the time, to stop the numbers spiraling out of control. This will be the case until R reliably stays below one, due to mass vaccination, better testing and contact tracing, or immunity from previous infections.

Here’s another way to look at it. Think of the epidemic as a long ladder leaning against a building, on which we find ourselves stuck.

The lowest rung represents a manageable 1,000 new infections per day. Every other rung represents 5% more daily infections than the one immediately below. So the second rung is 1,050/day and the third is 1,103/day. Higher up, the tenth rung is ~1,600/day and the thirtieth ~4,300/day.

The ladder has eighty rungs in total, with the highest representing a little under 50,000 infections per day. If we step above that, the health system collapses. We fall off and die.

Now, every day, we climb up or down the ladder based on what we choose. On days outside of lockdown, we take one step up — human contact causes the number of new infections to grow by 5%. But on days inside lockdown, we climb down two steps instead, because most people can only be infected by those in their immediate households. (I’m brushing over a number of subtleties here, some of which we’ll discuss later.)

These rules have two immediate implications. First, no matter where we start on the ladder, we cannot spend more than 80 consecutive days outside of lockdown without falling off the top. Second, over the long run, we need to spend at least one day in lockdown for every two days outside. Otherwise, on balance, we’re climbing up more than down, and that way lies trouble.

Nonetheless, within these constraints, we have the freedom to decide exactly when we go up or down. For example, the Minister of Health could roll a dice each week, and we’d lock down if it showed a 5 or a 6. Alternatively, we could shuffle a deck of cards, assign one to each week of the year, and lock down for the hearts and royal diamonds. Either would yield the necessary ratio of one in three.

Putting these weird options aside, what have we been doing in the Western world? The answer, in a nutshell, is “the long up and down”. We keep things open for as long as possible, climbing the ladder steadily as the number of infections grows. Perhaps we close our eyes, hoping the ladder isn’t there, or at least we don’t look down. And yet, soon enough, we find ourselves dangerously near the top. So we panic and lock down indefinitely, staying that way until we’re safely back on the ground.

And then we do it again. And again. There has to be a better way, and there is.

Reactive lockdowns — The long up and down

Short and proactive

Perhaps by now the idea is obvious. Instead of long emergency lockdowns, we need a regular schedule of short proactive ones. Rather than continuing to alternate between delusion and panic, we need to resign ourselves to failure and take back control.

There are many possible ways to do this, but I’ll focus here on a six-week cycle of four weeks open, two weeks shut. During the open weeks, life is mostly normal — schools, offices, restaurants and shops are operating and people can visit each other’s homes. Only the most risky activities are banned, such as large indoor gatherings at nightclubs, concerts and theaters. By contrast, during the closed weeks, all but essential workers retreat back to their homes. Life slows down, so that life can be preserved.

This approach would improve lockdowns in so many ways, it’s hard to know where to start:

  • Predictability. We’ll know what is coming well in advance. Instead of turning us into elastic nations, lockdowns will feel more like pre-planned staycations — hardly a source of anger and stress.
  • Duration. By making each lockdown shorter, we can reduce the emotional cost for the lonely and vulnerable. Children won’t be stuck at home for weeks on end. The end date will always be known and in view.
  • Fairness. At the moment, before locking down fully, countries naturally try to push R below 1.0 using partial measures, such as closing some school years or sectors of the economy. And when emerging from lockdown, they tend to do the same in reverse. But this is a fool’s game. Are gyms more or less important than restaurants? Which school years are dispensable? Any answer creates arbitrary winners and losers, leading to widespread resentment and mistrust. With proactive lockdowns, everything opens and closes together — the pain is fairly shared.
  • Less disease. Let’s say the cycle begins at the bottom of ladder, with 1,000 new infections/day. During the four open weeks, we climb 28 steps, reaching a high point of 3,920 infections/day before turning around. The result is a consistently low burden of sickness and death, averaging ~2,140 infections/day. By contrast, climbing all the way up and down the ladder yields a peak of ~49,500/day and an average of ~12,450/day — twelve and six times worse respectively (see graphs above and below).
  • Empowerment. Instead of reacting to the epidemic, we’ll be back in the driving seat. This will transform the psychology of lockdowns and encourage them to be used creatively. Schools could teach formal subjects for four weeks, then assign art projects and online activities for the two-week break. Families could buy and master a new board game or make a change to their homes. Employers could incentivize workers to learn a new skill, be it Python programming or spoken Mandarin.
  • Easy tuning. Two out of six won’t always fit. During the coldest parts of winter, people spend extra time indoors, so infections might grow from one cycle to the next. Alternatively, improved testing could push us in the other direction. Either way, we can transition to a new pattern — perhaps three weeks in six or two weeks in eight. Most crucially, we provide ample advanced warning of the change. Since the number of infections starts low, we can live with an imperfect cycle or two before moving to the new rhythm.
Proactive lockdowns — Staying near the bottom


All too easy? Let’s explore some of the arguments against:

  • You’re admitting defeat. That’s right, because we are defeated. Every lockdown, whether emergency or scheduled, is a failure. A failure of leadership, a failure of contact tracing, a failure of people to follow the rules. But whatever the underlying reason, it’s time to make the best of this failure rather than wishing it away.
  • Lockdowns won’t be needed once we have faster tests / better contact tracing / a vaccine. I’m not suggesting that we stop pursuing other strategies for combating the virus — quite the opposite. Any success in these areas will manifest as a decline in case numbers from one cycle to the next, allowing the gap between lockdowns to be increased.
  • Why not lock down every third day? While the ladder analogy is helpful, it is overly simplistic. Longer lockdowns drive the reproduction rate R lower, accelerating the speed of descent, because infections take time to develop. This “incubation period” averages 5–6 days but can last for up to fourteen. Extended lockdowns ensure that symptomatic infections are caught and chains of asymptomatic transmission between household members are exhausted. There’s a trade-off between effectiveness and brevity.
  • Why stop on the way down? If we start low on the ladder, I’ve argued that short periodic lockdowns can keep us there. So why not take this to its logical conclusion, and wait to reach 100 or 10 daily infections before we start? The answer is that most Western countries are not hermetically sealed — people coming from elsewhere will introduce new infections at a constant rate which lockdowns cannot prevent.
  • You’ll never reach herd immunity. If we keep infections low, then few people will become immune and the epidemic will never end. (For now, let’s put aside the as-yet-unanswered question regarding long-term immunity.) This criticism can be addressed in two ways. First, the price of herd immunity is significant death of the elderly and sick — an unpopular trade-off to say the least. Second, if the aim is to infect as many people as possible, then periodic lockdowns can help achieve this quickly without overloading the health system. Just start the cycle when infection rates are high rather than low, climbing up and down near the top of the ladder.
  • You’re not an epidemiologist. True, but epidemics happen to be highly mathematical phenomena, and the math is pretty simple. Using a spreadsheet, it takes minutes to simulate a simple SIR (Susceptible — Infectious — Recovered) model of infectious disease. In any event, real epidemiologists have proposed periodic lockdowns in the European Journal of Epidemiology and Science, as has the World Health Organization. Why is this option not discussed more widely?
  • People won’t accept it. Maybe we can only stomach lockdowns when disaster approaches — when we’re teetering at the top of the ladder. If so, proactive lockdowns will be politically infeasible and nobody will comply. This is a strong objection for which I see no magic answer. Perhaps we need to tire of emergency lockdowns before being open to other ideas.

A sustainable strategy

If periodic lockdowns are implemented, I hope it won’t be for long. With improvements in testing and contact tracing, and many vaccines under development, there’s a good chance the pandemic will end by the middle of 2021. We can all look forward to universities, theaters, concert halls and tourist attractions springing back to life.

But even in this cheery scenario, we still need to get through the next half year. Today’s long reactive lockdowns, driven by delusion and panic, are failing our societies. A repeating cycle of short proactive lockdowns would have lower medical, psychological and economic costs. As things improve, the gap between these lockdowns can be increased — when the gap becomes infinite, the epidemic is over.

But what if things turn out less well, and we have to live with Covid-19 for the long term? Perhaps a new strain will emerge annually, faster than we can tweak any vaccine. Or maybe this is just the first of many pandemics, as ecosystem destruction and climate change help other pathogens leap from animals to humans. How will we need to adapt and change?

I would argue that regular scheduled lockdowns offer a sustainable way forwards, because we are deeply familiar with cycles of time. These begin in nature — day and night, summer and winter. Our bodies layer their own rhythms on top — 90 minutes for sleep and 28 days for fertility. And society introduces yet more — work hours and breaks, weekdays and weekends, school term and vacation, sabbatical years and jubilees. Human beings are wired for rhythm. A six-week cycle of lockdowns would fit right in.

And one more thing. Like everything else in Western society, pre-planned lockdowns can be commercialized. Netflix can release family-friendly specials for binge-watching on the sofa. AirBNB hosts can offer socially distanced accommodation with bicycles thrown in. Local restaurants can hawk cheap meals cooked in bulk that are ideal for home delivery.

All of this would be welcome. Because if we’re going to have to live with lockdowns, we needn’t become elastic nations. Let’s try to stay sane instead.


  • Because of the incubation period, there is a delay between changes in lockdown status and our direction on the ladder. This effect was not discussed in the text but is shown in the graphs. It applies symmetrically and doesn’t affect the overall argument.
  • A growth rate of 5% per day was used for illustration. The actual rate will vary across geography and time. Some estimated daily rates for Western countries in early October: Canada 3.6%, France 2.0%, Germany 4.2%, Italy 6.1%, Netherlands 5.7%, Poland 8.2%, Spain -0.8% (under partial lockdown), UK 7.1%, USA 0.9%. These were calculated by dividing the number of cases detected during October 1–14 by the number detected during September 17–30, then taking the 14th root.
  • The suggestion that one day of lockdown reverses two days of growth is also an estimate. It’s difficult to obtain firm data on this, because testing improved drastically during the first wave of global lockdowns, confounding the statistics. From September 18th to October 18th, Israel was the first Western country to enter a second national lockdown with thorough testing in place. During the final week, detected cases dropped by 49%, the equivalent of 9.1% per day with compounding.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store