We’re now at about five weeks since the start of the COVID-19 lockdown in New York. I’ve found myself thinking more about the long-term future. I don’t have real expertise on any of these topics, and I’m hesitant to speculate excessively. However, many of us are finding ourselves needing to make decisions with incomplete information and without the benefit of expert opinions that we can trust. I’m sure there are many people out there who would have useful advice for me, but it’s not readily apparent who they are. I wouldn’t take anything I say below as direct advice: think things through for yourself and draw your own conclusions.

At some point, stay at home orders will be lifted. But, it seems likely that people will continue to behave differently than they did when COVID-19 wasn’t a risk. If they have the option to work from home, they’ll probably continue to do so. Working from home is a skill that can be improved at — people will have had time to improve at it. They will now have comfortable (or at least functional) home offices. They will have routines for buying groceries and cooking. Why risk going into the office? Childcare will continue to be a problem. Those that can afford not to will avoid sending their children to summer camp. The risk of them getting sick or bringing the disease home with them will mostly be viewed as not worthwhile. Together, all of these things will create inertia. They will ensure that things continue largely as they have over the last month. There will certainly be exceptions, like the flood of people trying to get a hair cut. But, there will not be a matching flood of people eating out at restaurants, going on vacation, etc.

As a result, the various factors that are affecting the economy seem unlikely to change much. To be clear, I’m not sure “the economy” in terms of GDP or public company performance matters much to me directly. But, our current economic system depends heavily on using employment to distribute resources. Drastic increases in unmployment lead to people unable to get healthcare, pay rent, or perhaps even afford food. I don’t think I’m able to make specific predictions about the second-order effects of pausing large swathes of the economy. But, on-balance, it seems like decreases in employment, production, and consumption will lead to a reduction in human flourishing.

Some seem to be anticipating a possible “quick” recovery because the underlying fundamentals of the businesses that are struggling haven’t changed. If you’re a restaurant that was popular, nothing that influences that popularity will have changed. Your customers still like your food. So, if you “unpause” the economy, things will go back to how they were quickly. That doesn’t make sense to me. If you sell clothing, or air travel, or hotel rooms, or the experience of eating at a restaurant, demand for your product has decreased, and not just because of a stay at home order. People are less likely to buy from you until they’re no longer scared of getting sick. And, even if you somehow make those fears magically vanish, can you rehire all of the skilled employees you laid off? What about the money you owe your landlord? What if your customers have lost their jobs or are otherwise less economically secure?

So, what does change after stay at home orders end and the initial surge of cases is contained? It’s likely that many businesses that can only exist in-person will try to figure out how to stay open while simultaneously protecting their customers. I’m imagining barber shops only allowing two customers at a time, or restaurants with one-third as many tables. These measures will vary in their effectiveness, and it will be very hard to know which are effective. We’ll probably enter a period of “security theater”, performing some rituals with questionable usefulness just because they’re expected of us.

What questions remain unanswered?

Most thinking about the future involves more asking questions than answers. Here are a bunch of things I have lots of uncertainty about.

Can local/state/federal governments effectively scale up testing, both for active cases and protective antibodies? At first, the answer to this seemed to be “no”. Recently, it’s shifting towards “maybe”. In the most hard-hit areas, testing capacity has increased dramatically. Given the collective benefits, it doesn’t seem like there will be issues around financial access to testing. That might change if the situation requires a prolonged period of extensive and ongoing testing. There is also the chance that test reagents availability might limit large-scale testing.

Can governments combine test availability with active tracing? In my interpretation, “active tracing” means something like: if you test positive, everybody you came into direct contact with over the last n days gets a phone call asking them to come in for a test. This seems more difficult to implement than widespread testing. It involves hiring a large temporary workforce, working across state lines without a comprehensive national ID system, and possibly complex technology-dependent solutions. The difficulties of administering a program like this might be insurmountable for governments in the U.S.

Will testing and contact tracing be enough to keep the rate of new cases low and constant, or will we face repeated spikes of new cases? This seems difficult to predict without more data. If South Korea (seemingly the most effective deployment of this strategy) continues to successfully sustain a low new case rate, that would be a powerful signal that this strategy can work.

Can an effective vaccine for coronavirus be made? There’s evidence that an effective vaccine isn’t a slam dunk. Other coronaviruses mutate quickly and seem to have proven difficult to create vaccines for. One prominent example of this is the common cold. That said, a vaccine would be valuable even if it requires multiple doses, or a yearly booster, or has limited efficacy like the flu vaccine.

How long will it take to develop, test, and deploy a vaccine? Current estimates suggest the earliest date a vaccine will be ready to be mass-produced and deployed would be early 2021. That’s if everything goes according to plan. The deployment phase would depend a lot on how the vaccine is produced, but is likely to be on the order of months or even years.

How long does immunity to covid-19 last? The evidence does suggest that our immune systems produce antibodies specific to SARS-CoV-19, and that they are effective, at least for a few months. It’s possible (maybe likely?) that immunity will decline over time, but still provide some protection. The 1918 flu was deadly due to being caused a new strain of flu virus, but that same strain now circulates in the population without being nearly as deadly.

How aggressive is the federal government willing to be about using its powers to support individuals and businesses? So far, there has been willingness to take actions that I would classify as politically “unexpected”. See: giving out $1200 checks, given the ideology of the party in control of the Senate and the Presidency. The federal government can more or less print as much money as it wants, subject to the risks of doing so. It could pay the rent and unemployment of every person/business affected by coronavirus, if it chooses to. That seems unlikely. So, on a spectrum from that to nothing, what will it do?

What is the risk tolerance of a typical person? How do they respond to the individual risk they face and the collective risk their community faces once the government is no longer making the risk-tolerance decision for them via a lockdown? I’m not sure how to even begin answering this, other than “wait and see”.

What are the hidden, second-order risks that aren’t being considered yet? An example: what happens when everybody is working from home and starts using their air conditioning all day during the summer? I’m guessing cooling a bunch of individual dwellings requires more power than office air conditioning. So, how much extra demand can the power grid support?

How effectively will COVID-19 spread in rural areas? So far, the major outbreaks (in the U.S.) have exclusively been in urban and suburban areas. Will the smaller social networks and lack of density in rural areas be sufficiently protective? Or will those areas be just as badly affected, but on a delayed timeline?