A month ago, there was a ('silly') debate about whether COVID19 is still deadly. We have since seen that COVID19 fatalities up 2-3x in the US from the lows seen by end-June. But where do we go from here; some signs are better for September = THREAD
At the time, many argued that case counts did not matter at all, since fatality rates were headed towards zero (based on aggregate data). But the data has since showed clearly that the aggregate data cannot be used to compute fatality rates
It was not hard to forecast based on state-level hospitalization data and state-level fatality rates (after hospitalization) that fatalities will spike. The disaggregated bottom-up approach has been reliable for a while.
And we can use the same framework, which has worked well for months to predict fatalities, to project into September. There is now some evidence from moderating hospitalization that fatalities can start to moderate within 2-3 weeks (red line in chart) 
Fatalities bottomed around 500/day at end-June (before the second wave started to show up in the, lagged, fatality stats). And now fatalities have been 1000-1500/day for some weeks. But by September, we should be heading lower, to <1000 again.
This trend/improvement is mostly backed in the cake at this point, given the lead/lag between hospitalizations and fatalities. Hence, the question is what happens further ahead, in Sep and Oct.
Many are focused on the school issue. And as with everything in the US, some are 100% for reopening and some are 100% against. But it is probably not that simple.
Looking at the global evidence. We have successful school reopening in Denmark and Norway (for example), and a bad experience in Israel (although it is hard to do statistical identification, when so many variables move at once; school reopening happened along with other steps)
The debate about 'safe' reopening of schools is not settled, but there is a lively debate (see for example @ProfEmilyOster coverage). And the fact that there is a debate may matter by itself.
emilyoster.substack.com/p/guest-post-i…
emilyoster.substack.com/p/guest-post-i…
My big picture take is the following:
Rt is very dependent on complacency. If you are complacent and do not avoid super-spreader events, then Rt can be very high, and we can be on a steep unsustainable path. (The fact that there is a debate => extreme complacency is avoided)
Rt is very dependent on complacency. If you are complacent and do not avoid super-spreader events, then Rt can be very high, and we can be on a steep unsustainable path. (The fact that there is a debate => extreme complacency is avoided)
But if we are vigilant, then a degree of reopening is not that risky. We have seen that in Sweden, and in many European countries. And we may (finally!) be about to see that in the US too.
We have general awareness of the virus. The 2nd wave over the last 2 months has meant that almost all regions in the US are fairly/somewhat careful, at least in terms of avoiding the most risky behavior (the deadly buffet lunch, crowded night clubs etc etc).
With the awareness now in place, with mask mandates from North to South (so to say) the risk of massive third waves around school reopening is probably less. There will surely be more cases, but it will not run amok in the same way as when people are caught truly off guard.
From a markets angle, this means that we are in a regime where case growth will not impact market that much. Probably not in the next 2-3 months, to be a bit more specific.
This is partly because 1) case growth will not be allowed to accelerate as violently again (due to self-policing/more proactive policy 2) markets are pricing in vaccine help within 3-9 months, meaning that the ‘bad period’ is perceived to be short in any case, even if Rt spikes.
The risk is that weather changes, and more indoor activity, from November, will have an impact in Northern states. But it is hard to know what is worse: Being stuck inside when it is cold outside (in the North), or being stuck inside when it is hot (in the South)
In any case, the weather effects will only change materially in some months, so it is a risk for later.
The bottom line. US fatalities have risen 2-3x from the June lows, but are set to moderate into September. And while school-reopening is a risk, it will generally be a be done carefully, as opposed to without any precautions.
Hence, we may finally be at a point where the US may start to look more like Europe, and avoid the super-spreading events that really make the growth explosive and very deadly.
There is a lot of virus in the US still, and it will take a long time to generate a meaningful descent. But there is some hope that the trend can be somewhat better over the next 2-3 months, at which point we will see if changed seasonals create fresh risk.
I will leave it at that. Markets are unlikely to be violently moved by US case counts in the next 2-3 months (as cases /deaths will not explode). But we still have the economic hang-over from COVID. The focus will be on how economic policy will be calibrated to ease the pain. END
