Coronavirus: a basic playbook
3 open questions, 4 exit scenarios, and we don’t get to choose
We don’t know what will happen next.
We know that the pandemic is really bad and we know that the economic crisis will be really bad. But we don’t know how bad.
We know that the world will be forever changed. But we don’t know how.
This article will not tell you what will happen next—I don’t have the answers—but it offers a simple framework that articulates three of the most pressing scientific open questions with four possible exit trajectories.
The three pressing questions are:
- Does exposure to the virus provide a lasting immunity? We know that that some antibodies persist after recovery, but we don’t know for how long, and we don’t know how much they will protect from reinfection.
- How deadly Covid-19 really is? In a prior article, I explained why both the number of infections and the number of deaths caused by the virus were massively underestimated, resulting in major uncertainty on the actual fatality rate. New data has now brought some clarity on the reality of the death toll, but we’re still in the dark regarding how many people have been infected by the virus.
- Can we deploy containment strategies that are effective AND socially and economically acceptable over the long term? This is a multi-faceted question, with an epidemiology dimension (besides lockdown, do we have real options?), a technology dimension (can we enforce detection, tracing, protection and disinfection at scale?) and very complex social and economic aspects.
Depending on how these questions are answered, the future could resemble one of these four archetypal scenarios:
- The Big Flu. If we discover that we have massively underestimated the number of infected people, it will mean that the disease is more contagious but less deadly that previously thought. If Covid-19 has an Infection Fatality Rate around 0.5%, it may not be just the flu but we can get past this crisis by flattening the curve and reaching herd immunity. This scenario is the best we can hope for.
- Black Mirror. But if Covid-19 kills more than 1% of all infected people, the death toll associated with herd immunity could be judged unacceptable. In this scenario, our societies manage to contain the pandemic while restarting the economy, until a cure or a vaccine is found. How do they achieve this? The script has yet to be written. Optimistic versions involve masks and common sense. Pessimistic versions have a distinctively Orwellian dystopian flavor.
- The Dilemma. This alternate dystopian scenario starts with the same premise (an Infection Fatality Rate above 1%) but diverges from Black Mirror as societies find out that they cannot contain the pandemic without inflicting unacceptable social and economic damage. Between two unacceptable options, they will have to make a choice. Will they sacrifice the old and the sick so that the young and the healthy can have a better life?
- The Winter. If our immune response to the coronavirus is short-lived, or useless because the virus mutates too fast, we’re in big trouble. Our chances of building an effective vaccine might even be jeopardized. It is unclear which societal impact would result from a constantly renewed threat to everyone. The Winter might initially resemble Black Mirror, as both scenarios involve playing cat and mouse with the virus. But in Black Mirror we are the cat and the virus is the mouse, while in The Winter it is the other way around. The Winter will feel long and we will not like it.
At this point, each of the four scenarios remains a possibility.
In the end, we will probably not get to choose: clear cut resolution of the underlying unknowns could abruptly force us into one of the four options.
But there is also a possibility that the answers will not be clear cut, and that the future will be a combination of several of these scenarios, possibly all four of them.
In fact, components of each of the four scenarios are already at play, in some form or another.
Right now, we are probably somewhere between The Big Flu and Black Mirror. This explains why governments are still unsure whether they should flatten the curve and go for herd immunity (The Big Flu) or duck out until we find a vaccine (Black Mirror.) As current estimates for the fatality rate could easily be off-track, the situation is likely to evolve.
If you don’t like these four options, I have included a bonus scenario: The Cure. This one has a happy ending but its timeline is unclear.
1. The Big Flu
In the past week, daily death counts have significantly decreased in several countries, most notably in hard-hit Italy and Spain. Lockdowns are a major part of the explanation. But what if the pandemic is also starting to flame-out in the hardest hit regions, because a significant fraction of the population has already been infected?
Italy has 60 million inhabitants and 20,000 deaths have been attributed to Covid-19. Spain has 47 million inhabitants and 17,000 deaths have been attributed to Covid-19. But how many people have been infected? The official counts are around 160,000 in Italy and 170,000 in Spain, but these countries only test the most serious cases. We do not really know how many mild and asymptomatic cases we are missing.
The Imperial College London epidemiology team currently estimates that 4% of Italians—2.4 million people, 15x more than the official count—and 7% of Spaniards—3.3 million people, 20x more than the official count— have been infected by the virus (the models are updated daily so the figures might be different when you check the links.)
The Big Flu scenario would automatically unfold if we discover that the true contamination rate is 50x or 100x higher than the official counts.
Why is this scenario possible? The Imperial College team estimates infection counts by extrapolating from death counts, using published estimates for the Infection Fatality Rate. But these estimates (somewhere between 0.5% and 1%) could themselves be wrong if many asymptomatic cases were falsely tested as negative. Anecdotal evidence from the Diamond Princess are consistent with a fairly low Infection Fatality Rate, around or below 0.5%.
How will we find out? New serology tests are being deployed that can reveal if someone has been exposed to the virus, even when the virus cannot be directly detected. This will allow a more direct measurement of the infection rate. A first study claims that 14% of the population of one German city is positive. Without large-scale serology testing, we could also find out if observed cases and deaths fall quicker than anticipated by current models (randomized surveys would provide an earlier signal of this.)
What would it look like? If we find out that more than 20% of all Spaniards have already been infected by Covid-19, it would mean that 1/ the virus is much less deadly than previously thought, 2/ it is also much more contagious (explosive contagiosity, not mortality, would be the #1 explanation for our problems), 3/a good fraction of the path to herd immunity has already been accomplished. Not everything would be solved, but there would be light at the end of the tunnel: by flattening the curve, we could reach herd immunity without saturating our healthcare resources. Many more people would die, but not insanely more. In this scenario, we are not being hit by an asteroid: an orderly return to “normal life” can take place on a relatively short timescale.
How we would tell the story to our grandchildren. “What was tricky about this virus was that it spread so fast. We were caught off-guards, with no masks, no tests and no clues. It was such a mess. We basically had to pull the brakes on everything. The irony is that it wasn’t actually that dangerous. It killed more people than a regular flu, but not insanely more. It wasn’t just the flu, for sure, but it wasn’t the end of the world either. It was the Big Flu. Then it vanished. All it left behind was this gigantic economic crisis.”
2. Black Mirror
But if the true number of infected people is no more than 10x higher than the official infection count, then we’re heading into a much more complex situation.
Why is it such a bad news to find out that fewer people have been infected? Because it would mean that the virus has a much higher fatality rate than previously thought.
In Italy, the official Case Fatality Rate (obtained by dividing the official number of Covid-19 deaths by the official number of Covid-19 cases) is above 12%. Even if we miss 9 out of 10 infections, that still puts the true Infection Fatality Rate above 1%. But it is now certain that we are also undercounting fatalities, probably by more than 2x. If we miss out 9 out of 10 infections and 1 out of 2 deaths, that puts the Infection Fatality Rate at 2.4%. Reaching herd immunity requires infecting around 75% of the Italian population, this would kill more than a million people, in Italy alone.
The premise of Black Mirror: a fatality rate too high for the path to herd immunity to be acceptable. This scenario includes no plan to fully re-open in the near future, no flattening the curve. The pandemic must not be slowed, it must be contained. Lockdown becomes the new baseline as we enter an indefinite holding pattern, waiting for a cure or a vaccine.
Why is this scenario possible? It’s not just Italy. Case Fatality Rates are rising almost everywhere and, in seven countries, they have already crossed 10%:
Why are the curves rising? Covid-19 is a fast-spreader but a slow-killer, there is a lag between infection and death. This effect was predicted in my earlier article but I didn’t expect it to be that strong. And in all countries, many deaths caused by Covid-19 are still missing from the official counts, because they aren’t officially diagnosed. France’s rate spiked once it started including cases from nursing homes, not just hospitals. Despite this, analysis of excess deaths in raw mortality records shows that France’s official numbers remain incomplete (to France’s credit, its raw mortality records are publicly available.) Some of the countries with the better-looking fatality rates might be testing a lot more, but most are just brutally understating their death counts with zero transparency.
Which fatality rate would trigger the Black Mirror scenario? This is a tough question and, apparently, a major taboo. Western democracies initially explained their lockdowns within a very clear Flatten the Curve storyline:
Over the past few weeks, the Flatten the Curve theme mostly disappeared from public communication, replaced by allusions to contact-tracing apps and “long-term efforts.” This signals that governments are viewing Black Mirror as a serious option. Yet they are still uncomfortable articulating it clearly to the public.
How will we find out if we need Black Mirror? The smart way: large scale serology testing (to assess the true number of infections) + analysis of raw death records (to assess the true number of deaths) + strong leadership. The hard way: premature reopening without adequate monitoring + panicking.
What would Black Mirror look like? No-one really knows. Technically, it involves keeping the propagation coefficient Rt below 1. Tomas Puyo calls this process the Dance, but after watching this video about life in Nanjing you will probably agree that Black Mirror is a more appropriate name (albeit less palatable):
Most likely, Black Mirror will involve a combination of measures that dynamically evolve as we gain insights on their relative effectiveness.
Countries are in lockdown because they don’t know, or cannot implement, less drastic ways of containing the pandemic: lockdown is a symptom of our ignorance and our ineffectiveness, not a long-term solution.
Some measures are likely to be no-brainers (masks, tests, self-diagnosis, self-quarantine…) Some measures (contact-tracing apps…) will make some people uncomfortable (but they’ll have to live with it.) Some governments will introduce outrightly dictatorial measures (discretionary quarantines, exclusion zones for certain categories of people…)
Black Mirror is about maintaining the effort over multiple quarters, possibly years. So we’ll have to find solutions for all aspects of life. What about school? What about nightlife? What about sport? What about romance?
The harshest measures will not necessarily be the most effective ones, and vice versa. Black Mirror could be horrible, or we could simply get used to it.
3. The Dilemma
The Dilemma is the evil twin of the Black Mirror scenario. It starts with an identical premise: a fatality rate that is too high for herd immunity to be an acceptable outcome. But the true storyline kicks in once attempts to deploy Black Mirror run into problems. Why do they run into problems? Because Black Mirror isn’t just dystopian, it also carries serious execution risk:
- What if the virus is so contagious that we have few options besides full lockdown?
- What if governments fail at deploying advanced containment technology due to social backlash?
- Will the economy truly restart if every step we make feels like going through airport security?
If containing an unacceptably deadly plague comes at a price that is socially and economically unacceptable, which one will we accept? Welcome to the Dilemma.
Why is this scenario possible? 1/ We know very little about Covid-19 and techno-utopian overconfidence is possibly misplaced. Of all the countries in the world, none was better equipped than Singapore to contain the pandemic without resorting to lockdown. Yet its containment strategy failed and Singapore recently announced a one month lockdown. 2/ The coming economic crisis will exacerbate social tensions. The coronavirus disproportionally affects specific groups of people (roughly speaking, the old and the sick) while different groups could bear the brunt of the economic crisis: this will make the tensions worse. 3/ International competition: if some countries opt for The Dilemma while others struggle with Black Mirror, would this create unsustainable imbalances that would force the latter to change course?
What would it look like? Needless to say, The Dilemma is no less dystopian than Black Mirror. In crude terms, the likely outcome will involve sacrificing the old and the obese so that the young and the slim can have a better life. In practical terms, it will superficially resemble The Big Flu: we will reopen and march toward herd immunity, except that for millions of people it will be a march to death. It is unlikely to be explicitly advertised as such. Most likely, it will be presented as the only practical solution, while the human cost will be presented as unavoidable: “Hey, it’s just the flu, and these old people were going to die anyway.” It will be sold to us as The Big Flu.
So how do we know if we’re in The Big Flu or in The Dilemma? It’s all about the fatality rate and how we collectively judge it:
- Marching to herd immunity with a fatality rate under 0.5% will be extremely painful but it can make us feel that we are collectively winning. In the end we might even be proud.
- But there is no way marching to herd immunity with a 2% fatality rate can make us proud. No US president can claim victory if it involves letting 5 million Americans die. Whatever the governments will say, the march will be gloomy, no-one will smile and no-one will cheer. There will be massive dissent. It will break our societies and the recovery will require us to reinvent our core values and institutions.
What about the current estimates for the fatality rate, somewhere between 0.5% and 1%? This is a tricky question and right now governments are struggling with it. For comparison, the baseline yearly mortality rates in Europe and in the US are close to 1%. If the Covid-19 pandemic adds between 0.5% and 1%, this might be judged acceptable, allowing governments to follow the Big Flu storyline. But what if the effective mortality ends up being rejected by societies? On the other hand, if you try Black Mirror for six months and it fails, you won’t be able to return to the Big Flu: everyone will know that you’re in the Dilemma.
Misjudging the fatality rate could break our societies.
Of the four scenarios, the Dilemma is most unsettling one. Black Mirror disturbs us by redefining our day-to-day experience, but the Dilemma confronts us with an unsolvable ethical problem together with an injunction to take sides. If you’re tempted to portray the Dilemma as generous humanism vs heartless utilitarianism, remember this: mass unemployment and starvation also carry unacceptably high fatality rates. The Dilemma isn’t about choosing between people and the economy, it is about choosing between people and people.
This is why I do not believe that the Dilemma is a scenario that modern societies would actively opt for. Spartans would have chosen it in a blink, but we are not Spartans. Some politicians and business leaders are calling for premature reopening; this is driven by ignorance and possibly cynicism, and these calls have zero chances of being heard. The British government initially postured with a reckless herd immunity strategy, but they quickly retreated and initiated a lockdown. If we end up in the Dilemma, it will not happen by choice, it will be a collective failure.
Right now the challenge is to build masks and tests, treat patients, solve the core science questions, experiment with smart containment strategies. Calling for human sacrifice is out of touch.
4. The Winter
Two of the above scenarios, The Big Flu and The Dilemma, involve reaching herd immunity. In Black Mirror, herd immunity isn’t reached, because doing so would cause too many deaths. In the Winter, herd immunity isn’t even an option, because exposure to the coronavirus does not provide a lasting immunity.
Why is this scenario possible? Because we just don’t know if exposure to SARS-CoV-2 (the virus behind Covid-19) provides a lasting immunity. We know that there is an antibody response and that it lasts for several weeks or more, but we still don’t know if it effectively provides immunity, and for how long this immunity would last. With other viruses from the coronavirus family, the immunity is short-lived.
What would it look like? Bad. So bad that it is hard to imagine. On a superficial level, it might resemble Black Mirror: both scenarios involve playing cat and mouse with the virus. But in Black Mirror we are the cat and the virus is the mouse, while in The Winter it is the other way around. And the Winter might be long.
Even if exposure to SARS-CoV-2 provides a lasting immunity, some elements of the Winter will inevitably sprinkle all other scenarios. The virus will mutate and it will come back. Being immunized to the Spring 2020 variant doesn’t necessarily mean that you won’t get infected by the Fall 2020 variant or one of the many Spring 2021 variants. Given the current footprint of the virus, there is zero chance that it will disappear, unless we can eradicate it with vaccines and/or cures.
In general viruses get less lethal with time, as sparing their hosts is a positive selection factor. Our common cold is caused by another coronavirus that could be the remnant of a long-forgotten and insanely deadly epidemic, sparked after one prehistoric hunter-gatherer ate the wrong bat.
But sometimes viruses come back and they’re more virulent. The Spanish Flu pandemic in 1918 consisted of three waves. The second one was the deadliest and for some unknown reasons it was especially lethal to young adults.
Bonus scenario: The Cure
This one is super easy: if we find a vaccine or a cure, the whole thing could just go away.
It could be next month.
It could be next year.
It could be in three years.
It could be tomorrow.
We just don’t know.
