COVID-19 UPDATE: Case trends more than "half-full" and in outright decline. Last time case decline = epicenter rotation

Yesterday, I had a conversation with Chris Murray, IHME Director and Chair of Dept of Health Metric Sciences.  The IHME is the organization that has used big-data models to forecast infections, cases, deaths and hospitalizations.  Their models have been used by several governments, including the US, to inform and govern their COVID-19 policy responses.  

Below is the IHME model for daily infections (infections are infected, cases are “detected infections”) and as shown below, the IHME model suggests that this recent surge in daily infections is likely peaking.  The surge started in early June (which coincides with the BLM protests) but the course correction/mask mandates, etc. are now resulting in an apex in infections.  This is very good news as it coincides with a peak in confirmed daily cases.  In other words, testing is now capturing properly the change in trend.  And this trend change is that the surge in COVID-19 cases is ending (centered in 4 states, FL, CA, AZ, TX, or F-CAT).

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: IHME

But even as daily case surge is peaking, it will take some time for daily deaths to peak.  This is a natural lag and that means the US is still facing more fatalities in the days ahead.  Daily deaths, in other words, will keep rising.  The chart below is the daily figures (cases, deaths, hospitalizations) compared to 7D ago, so it is a measure of trend.

– Daily cases are down for >10 days now.
– Daily deaths are up, but this is a lagged effect and should soon rollover.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation



While we are not trying to minimize the tragedy of deaths, but the solace we can take is that deaths have not risen at the same pace as cases.  So from a combination of better treatment protocols, less hospital overuse and different patient demographics, we are seeing a lower death toll.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project




STRATEGY: If cases are rolling over, epicenter rotation makes sense… but bigger “binary event” is vaccine/cure…
We understand why investors are skeptical of any “trend” in data.  After all, since June, US COVID-19 cases soared (timing coincided with BLM protests) and the resulting rollback of easing, and coupled with many lamenting the “new normal” is nothing like the “old normal” is a reason for investors to stick with bond proxy trades.  And to treat equity markets with a skeptical eye.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation



But the simplified framework below, remains the strategy roadmap for us.  If the virus is weakening (which we believe), we think this necessitates a rotation away from “Growth” and “bond proxies” into “epicenter stocks” (aka cyclicals, Industrials, Financials, Discretionary and Energy).

– also, rotation away from momentum –> value

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation




Since February, this rotation has been visible and if cases are declining, we expect rotation again…
Take a look below.  The relative performance of epicenter stocks (SPHB is proxy) is a function of daily COVID-19 case trends.

– when daily cases rising (Feb to March and June to July), epicenter stocks tank
– when daily cases falling (April to May and mid-July to now), epicenter stocks outperform.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: Bloomberg


In essence, if one believes daily cases are falling, then one should be rotating into epicenter stocks.  Below is the list of epicenter stocks (*) that are also positively viewed by:

– Brian Rauscher, Head of Global Portfolio Strategy
– Rob Sluymer, Head of Technical Strategy
– Ranked DQM by tireless Ken’s quant model

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: Fundstrat


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: Fundstrat


Keep in mind that this rotation is also mismatched from a market-cap perspective.

– 72% of S&P 500 is “secular growth” + “bond proxies”
– 28% are epicenter stocks

So there is essentially a shortage of market cap if we see investors rotate into cyclicals/epicenter

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation





Finally, don’t forget the other “binary” event…
Finally, another binary event remains a cure/vaccine.  US has operation “Warp Speed” and has a target of a vaccine/cure by early 2021 (300mm vaccine doses).  Below are the industries which we see the most positively affected by the vaccine/cure:
– topping the list are travel-related groups like airlines, casinos, cruises, etc.
– these industries have “moats” around their business as long-term substitution unlikely.

It is the travel-related stocks that would arguably be the biggest winners in a vaccine/cure scenario.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: Fundstrat




POINT 1: USA Daily cases are down >5,600 vs 7D ago, slowdown is accelerating…
The downturn in daily USA COVID-19 cases is now accelerating.  The best way to see this is to look at daily cases vs 7D ago.  And as you can see, for the past 10 days, daily cases are down vs 7D ago, but this rate of change is accelerating.

– Tuesday, the 7D delta was -4,780
– Yesterday, the 7D delta is -5,648

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project


Daily COVID-19 cases came in at 63,627, which is up +4,908 vs 1D ago, but this is typical of the mid-week surge.  We feel it is now pretty conclusive, the post-BLM surge (starting in early June) is now over.  The key questions are now:

– how quickly will cases fall, if this is similar to late-May, we should see daily cases accelerate to downside
– will NY tristate see a relapse/resurgence? this time from teenage parties?

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project


After our comment earlier this week about the teenager parties in NJ/CT leading to superspreader outbreaks, several of you sent us similar stories in other states, towns and cities.  So, this is something we should be mindful of. There are cohorts of those who have not been exposed, and these events lead to equivalent of superspreader events and could eventually lead to hitting some vulnerable Americans.


The NY Post, this week, had a tragic story of a Texas couple who hosted a family event.  And unfortunately, this resulted in 14 members of the family getting infected and 1 death and 2 on life support.  These tragedies will happen.  And the lessons from these incidences is not to close the economy, in my opinion.  Rather, it is a reminder that COVID-19 is a mysterious and very contagious disease.

And it is unpredictably extremely dangerous to some Americans.  Because we do not know who this can be lethal, we all need to exercise diligence.


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


https://nypost.com/2020/07/28/coronavirus-strikes-14-family-members-kills-1-after-texas-party/?utm_campaign=iphone_nyp


The trend in hospitalizations has shown improvements as well.  Notice below the level of gross hospitalizations (33 states) and net hospitalized (17) is down.  There are now net discharges in those 17 states.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation



Daily fatalities are surging but this is a lagged function.  We expect daily deaths to rise for some time but this relationship between deaths and daily cases is not as severe as it was in March/April (see above chart)

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation




6 states with sizable 1D increase in cases…
California         8,755 vs 6,000 (1D) +2,755
Texas               9,042 vs 8,342            +700
Louisiana         1,735 vs 1,121            +614
Michigan             996 vs    669            +327
Illinois              1,393 vs 1,076            +317
Iowa                    500 vs    223           +277
Total 6 states                                   +4,990

6 states with sizable 1D declines
Georgia              3,271 vs 4,209 (1D) -938
Tennessee          1,778 vs 2,555         -777
Pennsylvania         834 vs 1,120         -286
Colorado                482 vs    749        -267
Oklahoma              848 vs 1,089         -241
Nevada                  870 vs 1,105         -235
Total 6 states                                    -2,744


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project





POINT 2: F-CAT, virus in “retreat” as R0 < 1.0 in each FL, CA, AZ, TX…
The good news is that F-CAT looks like it is definitely past the peak and now positivity rates are declining and even the calculated R0 is below 1.0 in each of the F-CAT states.  The reason a peak in these 4 states matter, is that these 4 states are really the last of a group of states seeing widespread COVID-19 outbreaks (>60% of USA) and many of the other 40% were seeing concurrent breakouts.


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project


R0 is <1.0 in each of F-CAT states…
For the R0, we use the calculation from the website rt.live and a R0<1.0 means the virus is in decline.

– as highlighted below, AZ and TX have among the lowest R0 in the US, thus, further evidence of a decline in cases.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: RT.live


But a simpler proxy for R0 is the 7D delta in detected cases.  We have highlighted the 7D delta for each for the F-CAT states below and as you can see, the trend in daily cases (vs 7D ago) has definitely turned negative in each of those states.

– the only tenuous start, potentially, is CA.  

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project


And we come back to these 2 charts below.  Hopefully, F-CAT follows the path of NY tristate, vs VIMMP (VA, IL, MI, MD and PA) and that once cases peak, we see a crushing decline.  AND no relapse.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project

As we have written in past commentaries, one reason for the relapse in VIMMP is that case prevalence was simply not high enough (~herd immunity?).  And as such, these states are vulnerable to a resurgence/relapse.  This is a question that will matter a lot more as we get into Winter.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: COVID-19 Tracking Project


The good news is F-CAT states are following the path of NY tristate, so far…
The good news is that the states are following the path of NY tristate. And if this is the case, we could start to see some serious declines in detected cases in the next few weeks.  As shown below, we could see cases down -75% within 3 weeks, which is by early August (F-CAT peaked >10 days ago).

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: Fundstrat


The daily cases per 1mm for each of the F-CAT states is shown below, and with the outcomes for NY tristate + MA + RI are comparative curves.  As you can see, the F-CAT states look to be following this pattern pretty closely.


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: State health websites

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: State health websites

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: State health websites


COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: State health websites

POINT 3: “new normal” probably Federal + State/Local employment replace “leisure”
There is going to be quite a lot of “stranded” capital in this post-COVID world.  There are multiple reasons for this, among them:

– changes in corporate behavior (shrink supply chains)
– changes in consumer leisure time allocated (away from crowds)
– changes where consumers choose to live and work (de-urbanization and work from home).

These changes are potentially likely to take place, even if there is a cure/vaccine within the next 12 months.  Whether this is simply the scars of the fear from the pandemic, or whether the pandemic is just the “catalyst,” we think some of these changes will be structural (see below).

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation



But the consequence of these changes is also stranded capital.  That is, capital that is spent and cannot be recovered.  This will result in realized losses and also losses in employment.  The future is uncertain, and we are simply outlining those thoughts above as a framework to think about the future.

But the impact on employment is probably the more immediate policy question.  Using data from the BLS (Bureau of Labor Statistics), below is the breakdown of industries and their contribution to changes in labor employment since 2000.  3 industries actually shrank by a total of nearly 5 million:

– Information                                 -657k
– Non-durables manufacturing -1.6mm
– Durables manufacturing        -2.8mm

But this was more than offset by growth in other industries.  In fact, as the pie chart below shows, the biggest contributors to employment gains since 2000:

– Education + Healthcare     +9.4 million
– Professional + Biz svcs     +5.0 million
– Leisure + Hospitality         +5.0 million

Of these, the greatest risk to employment is the leisure and hospitality sector.  Restaurants already have a high failure rate and low returns.  And the imposition of sanitization protocols and social distance and other mitigation is raising the cost of business.  So, the economics will simply be more challenging and thus, reduce the future employment (likely).

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: BLS


The category that created the most jobs in the past 20 years is ‘food services and drinking places (bars)’ which added 4 million jobs (of the 5 million in leisure).  This is the category that is likely to see diminished prospects for growth in the future.  Even if there is a vaccine.  The new safety protocols simply impair the economics.  Unless, of course, prices are raised.  But there is likely a level of price elasticity in those services (demand falters).

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: BLS


But of course, this is simply creative destruction.  The US economy is designed to see capital lost and capital re-deployed.  So, while many will prognosticate this to be a long-term driver of malaise (possible), this pandemic is also a catalyst for business creation.  For instance, many of those whole are displaced by lost jobs in leisure may find new employment in:

– Business sanitization
– Delivery
– Work from home services
– etc.

And lastly, we also think the employment levels of both Federal and State/Local governments likely to expand in the next decade.   The chart shows the Federal and State/Local share of employment since 1970.

– Total employment in the US is about 152 million
– Federal govt share of employment fell from 3.8% to 2.0%, or down -1.8%
– State/local share from 15.5% to 13.5%, or 2.0%

If employment share recovers to the 1973 levels, the increase in employment would be:

– Federal govt    +2.8 million
– State/Local      +3.0 million

Thus, as a last resort, we might simply see an expansion of those working for the public sector.  And it is certainly better to “hire” these Americans than to pay them unemployment insurance.

COVID-19 UPDATE: Case trends more than half-full and in outright decline. Last time case decline = epicenter rotation


Source: BLS

(*) Please note that the stocks rated OW on this list meet the requirements of our investment theme as of the publication date. We do not monitor this list day by day. A stock taken off this list means it no longer meets our investment criteria, but not necessarily that it is neutral rated or should be sold. Please consult your financial advisor to discuss your risk tolerance and other factors that characterize your unique investment profile.

More from the author

Disclosures (show)

Stay up to date with the latest articles and business updates. Subscribe to our newsletter

Articles Read 1/2

🎁 Unlock 1 extra article by joining our Community!

Stay up to date with the latest articles. You’ll even get special recommendations weekly.

Already have an account? Sign In

Want to receive Regular Market Updates to your Inbox?

I am your default error :)