The New York Times as Pollyanna for COVID-19

The New York Times has an interactive model of projected COVID-19 infection and fatalities in the USA titled “How Much Worse the Coronavirus Could Get, in Charts”. It’s a fairly simple model that allows the user to change some options to graphically see the effect of those changes.

It’s pretty scary as it is. But the title of the article is about how much worse it could get, and I don’t think the authors have delivered an actual answer to that question. That would involve figuring out a worst case with some likelihood of happening.

The NYT model has some rather conservative assumptions baked in that one can’t change. Like the mortality rate, which someone there decided should be 1%, though it hasn’t actually been that low anywhere else in the world. Doubling that is still on the lower end of the range of mortality rates that are discussed, and that is simple enough to keep in mind looking at their graphs, but 4% would be closer to the worst case consideration promised in the title. That’s still reasonably easy to guesstimate how it would look in the graph (4x the amplitude of any series about fatalities), but it should have been either the default or an option the user could select.

It also assumes the reinfection rate is insignificant, which it might be, but it might also turn out to be substantial. At least one talking head in the CNN Town Hall last night opined that the reinfection case they were discussing was simply something tied to that individual, not a generic capability of the disease to negate immune responses in general. It would be good if that were the case, and horrific if it turned out that the virus really was ineffectively targeted by immune systems that have dealt with it previously. There’s probably a range of immune effectiveness against reinfection that will turn out to be characterizable for this disease. For most infectious disease, we are able to consider immune protection against reinfection to be nearly complete, and that assumption is one that the NYT model incorporates as well. If reinfection were something that might happen to, say, even 10% of those previously having the disease, the shape of the curves would be altered. I think they would show a skew in the tail, where the number of known cases would decline more slowly due to persons being reinfected. In the limit, the decline would only be due to attrition of the population as the disease’s mortality rate eats away at the remainder of the population capable of being infected. (It seems obvious that graphs should also be done on proportion of population affected to remove the difficulties of interpretation that absolute numbers tend to bring with them.) Based on what has happened elsewhere, it seems obvious that we do not have the nightmare scenario of no residual immunity to the disease. But it is unsettling that there may be some appreciable degree of reinfection that may have to be included in more detailed models. A significant degree of reinfection would also make necessary further changes in behavior to restrict or delay the disease spread, changes unlikely to result in getting back to ‘normal’ quicker so far as the US system of capital and labor is concerned.

There was some discussion of seasonal cycles of the disease, but the assumption seemed to be that we would likely have more effective treatments and then a vaccine, making it less of a concern in the future. Again, that doesn’t seem to approach how much worse it could get. Maybe we don’t get a seasonal breather. Maybe symptomatic treatment remains about as effective as it is now. Maybe a vaccine turns out to be something that is tuned to a particular strain, and like the flu, predicting what is going to be effective for the next round is an issue. Get it wrong, and we get something with all the infectiousness of the flu (or more) and 20 to 40x the mortality rate. Much of the discussion saying “the flu is worse” was simply based on the difference in extent of spread of the diseases. It seems quite possible that COVID-19 will be just as prevalent as influenza after a full year of spread. It is not a given that this disease simply falls off the map after a few months, like several previous outbreaks of somewhat similar diseases. That looks quite a bit worse. And if that is so, we are faced with much broader disruption to the way humans usually socialize and work that could simply be the new normal, at least if we aim to minimize mortality. And that is going to reveal that the economies we’ve built are susceptible to breakage and collapse in various ways. We already have restaurants shuttering as the social distancing measures being promulgated now have an effect. The airline industry is having a severe recession immediately. If humans cannot congregate for manufacturing without getting ill, about the only route out of that is fairly complete automation. And automation of much more will advance at a far greater rate as human labor proves itself to be unreliable to the purpose over time. Do we re-adjust our view of the economy and go to something like a universal basic income to protect the population, or do we opt for rugged individualism that will put many more of us on the streets as the disease continues to take a tool? These questions have been treated as theoretical discussions that can be deferred to some indefinite time in the future, but COVID-19 could be the spur to having to make a decision with immense consequences. All that is premised on not being able to give up certain principles that conservatives hold dear, such as government not being able to accomplish things well. As the risks involved in running a large, susceptible workforce weigh on private business, more will simply leave that sector. Given how thoroughly government services have been peeled off to hand them to the private sector, that means quite a few essential services are likely to end up without private companies willing to actually do those jobs. Which means that it becomes more likely that either government has to get back into actually doing those jobs, or allow the services to be withdrawn without replacement. Mass austerity kills, too.