COVID-19 UPDATE: IHME Takeaways. The more "fear" the better the outcome. Masks as good as vaccines. Israel shows markets did not care about cases...

Click HERE to access the FSInsight COVID-19 Daily Chartbook.

STRATEGY: COVID-19 single most important market factor YTD in 2020 and into YE…
Looking at the first 9 months of 2020, the pandemic has been the single most important factor for markets.  The pandemic drove monetary policy, fiscal stimulus and economic and behavioral changes — a quadfecta of impacts.  In fact, while the BIS (Bank of International Settlements) comments that Fed policy accounts for 50% of the recovery of stocks, the Fed move is all due to COVID-19.

https://www.bis.org/publ/qtrpdf/r_qt2009a.htm

And as we look into YE, COVID-19 remains the single biggest influence:

– COVID-19 vaccine/cure will affect 2020 elections
– COVID-19 daily case trends will influence Americans’ behavior
– COVID-19 is behind the fiscal stimulus

So in short, COVID-19 remains a central influence.  And unfortunately, it is completely unpredictable.  Fortunately, daily deaths are trending lower and now 89% off their highs.  The moving average for daily deaths was messed up due to the Labor Day holiday closures but this week will be the true trend.  Read below (Point #3) to see what well-known COVID-19 model forecasters expect for US deaths this Winter — hint, two very different outcomes.

Source: COVID-19 Tracking Project


The latest commentary from JPMorgan US Fixed Income team caught my eye.  They see:

– US rates drifting lower into YE
– US corporate bond spreads tightening (rally)

Both would be good for equities.  Hence, despite the risks developing, stocks have good risk/reward.

Source: JPMorgan





POINT 1: Daily cases 34,797 up +2,132 vs 7D, improving each of the past 5 days
Last week, we commented that the rise in US daily cases seemed to be some type of distortion in trends due to the Labor Day holiday — that is, long weekend and closures mean reported data was less complete last week. And this week, we are seeing a catch up of this data.  So, this week would be a week to see what the proper trend is (baseline). Yesterday’s daily cases were up +2,132 vs 7D ago, but this rise is at a declining rate.  So, the next few days will be worth watching.

Source: COVID-19 Tracking Project

This week will give us a clearer picture…
Again, the daily change vs 7D ago, in our view, is the leading indicator as it is what influences the 7D moving average.  The 7D delta is up, meaning cases are higher vs 7D ago, but the level of increase is slowing.  It was +2,132 yesterday but it was +11,177 last Tuesday.

– and every day since last Tuesday has seen this rise at a smaller pace
– the question is whether this will go negative this week

Source: COVID-19 Tracking and Fundstrat 


These are the 6 states with the highest 7D delta in daily reported cases.  The top of the list is Tennessee.  Tennessee is accounting for ~50% of the rise vs 7D ago, so this state has the biggest impact.

Tennessee daily cases have been rising in the past few weeks as shown below and this is likely heavily influenced by a surge in cases at their universities and colleges.  

Source: COVID-19 Tracking Project

As the NY Times data below shows, there are 59 colleges in TN reporting cases and you can see that the large schools are reporting a surge in cases (the bar chart/heatmap).

Source: NY Times

Source: COVID-19 Tracking and Fundstrat




POINT 2: IHME Takeaways. The more “fear” the better the outcome.  Masks as good as vaccines.  They see a resurgence in US deaths in Winter. 
We hosted a webinar last Friday with Chris Murray, Head Scientist for the IHME.  For a Friday afternoon, this was very well attended with >700 registered and ~500 tuned in.  We had received over 59 questions, which is quite a number, and we were not able to get to all the questions.  If you miss this zoom, the replay is:
___________

The replay can be accessed here –> (click here for video replay and click here for audio replay).
___________

There were many topics discussed, so I want to give a few high-level takeaways:

– IHME sees a US second wave starting in early October, pushing daily deaths to 4,000 (mid-case) and to as high as 8,000 per day
– The peak of US daily deaths in April was 2,200 per day, so IHME sees a 2X-4X surge beyond prior highs
– While several factors drive this, the two primary drivers: (i) seasonality and (ii) Americans will get “careless” as they get comfortable.
– Mitigation comes down to mask compliance, which they see as more effective than US vaccines (see below) and better policy response.
– Things that “contrary” to models: (i) low prevalence in Asia, Africa, Pakistan and Sweden and (ii) fatality rates improving (+vaccine); 

The over-arching message from IHME is that public policy can have a big influence on the outcome of this second wave.  As highlighted above, if mask mandates are universal, the US daily deaths would never rise above 1,000 in the winter.  Currently, mask mandates are not universal.  IHME believes masks reduce risks by 40%.  Murray also noted that this might be more effective than a vaccine.  For vaccines, if 50% of US takes it and it is 60% effective, this is only a 30% reduction.  Thus, masks > vaccine, in this example.

I agree with Murray’s assertion that there is a self-defeating loop in COVID.  As the US cases slow, there is a risk that Americans become less vigilant and this leads to carelessness and we see a renewed spread.  To an extent, this is what we are seeing at colleges and universities. Thus, the more “fearful” Americans are, the better the compliance. 

– But I realize this is easier said than done for America.  Even if there was a national mask mandate, this does not guarantee compliance. 

Why are some nations seeing low prevalence?
One of the most curious charts, to me, is this one. Murray was highlighting case prevalence across the World and he commented that there is unusually low prevalence in: Asia, Southeast Asia, sub-Saharan Africa and Pakistan. 

He did conjecture there could be some other factor at work, such as biological or dietary. 
– He noted that Southeast Asians tend to have a higher exposure to bats and bat viruses and thus inferring immunity (ala t-cells)
– For Africa, he notes there is a lot of work that polio vaccinations could be inferring some immunity.
– Overall, this is why “herd immunity” is not necessarily at 80%-90% but he sees it at more like 40%


– But what really struck me was the exponentially higher prevalence in Latin and Central America.
– Nearly 10X the rest of the World
– And also, that in the US, Blacks and Latinos are disproportionate share of US cases, but this diverges outside the US

Source: IHME


Massachusetts is an example, where Blacks and Latinos are 19% of the total state population but are 44% of total cases in that state.  This is pretty common across many states and the conventional explanation is that Latinos and Blacks are taking the “riskier” jobs and thus, are more vulnerable.  

– the issue with this “broad brush” conclusion is that if Latinos/Blacks are 44% of cases in MA, does this mean they are 44% of the employed in “risky jobs”?  If this was true, public policy could easily fix this.
– I don’t think this fully explains the difference.
– Incidentally, vitamin D is gaining increasing acceptance as a mitigating factor, and Latin Americans and US Latinos and US blacks are known to be vitamin D deficient. 
– Moreover, Nordic nations (including Sweden) eat a lot of salmon, which is a big source of vitamin D and that is a region with low prevalence as well
– I am not a scientist, but this is food for thought

Source: COVID-19 Tracking Project





POINT 3: IHME and Youyang Gu similar model accuracy, completely different views…
Quite a number of clients pointed out that if the IHME forecast is right, wouldn’t this be negative for financial markets?  The answer is not so simple.

Foremost, the future is uncertain, so the key takeaway from every forecast is the process behind the view, but the actual output can be less important.  2020 is a lesson in this.  After all, we saw multiple instances of markets over-reacting to models/forecasts that proved incorrect — even recently, many policy makers were saying US cases would hit 100,000 this Summer.

Israel shows that markets stopped caring about cases…
More importantly, we can look at the case of Israel.  Dr. Murray noted this nation saw a massive massive second wave of cases with recent daily cases averaging 5,000 per day compared to <1,000 in March.  In fact, Israel had to institute new restrictive measures to slow the spread.

– But take a look at the Israel equity market (ISSMYETR Index on BBG) versus MSCI
– Since early June when Israel cases began to explode, ISSMYETR has been outperforming MSCI

Bottom line, the future is uncertain and how financial markets react to data is uncertain.  

Look at the horrific forecast error of Imperial College…
The IHME presentation compared the accuracy of forecasts of different models.  And what stands out is the massive forecast error of Imperial College.  Recall, it is the Imperial College forecast that prompted global policy makers to shut down the economy.  In fact, as one of our clients, John Gordon, passed along to me, the WSJ editorial commented about how policy makers would act differently with their current knowledge. The WSJ editorial referenced a recent study in the Annals of Internal Medicine, which reported that the infection fatality rate for non-institutionalized persons under 40 was 0.01%, way below the 5%-10% of Imperial College.  And the same WSJ edition quoted the German Health minister saying that the government would have applied a less sweeping program of measures with benefit of today’s greater knowledge about how to manage infections. 

IHME and Youyang Gu have similar model accuracy, but completely different views…
The model accuracy of IHME and a young data scientist, Youyang Gu are similar, with about 22% error after 10 weeks.  This is something highlighted by Murray.

But look at these two forecasts today.  They are completely divergent.

– IHME sees a massive rise in deaths in the coming weeks.
– YYG sees US daily deaths falling below 500

Thus, the future is uncertain.

Source: twitter.com