COVID-19 UPDATE: CA backlog pollute data for the next few weeks but case trends still down... 9 reasons the coming "epicenter" rally could dwarf the May-June rally of 30%
Click HERE to access the FSInsight COVID-19 Daily Chartbook.
We finally have power and internet back at my house yesterday evening. I live in an area hit hard by Tropical Storm Isaias and I have been without high-speed internet since last Wednesday. And our head of data science, tireless Ken, only got his power back on Monday morning. Both of us have been doing a lot of improvising and I wrote our commentaries since Thursday (last week) from the parking lot of Dunkin Donuts (which has GREAT INTERNET SPEEDS!)
Biden has picked Kamala Harris as his VP. Tom Block, our head of Policy Strategy, has been calling for this nomination for some time, so great call, Tom Block. But he also notes that the VP rarely makes a difference. In fact, at JPMorgan, he told me that people can never remember the VP of the losing ticket. Who was John Kerry’s VP???
Given the widespread prevalence of COVID-19 in the US, the best case scenario is for cases to peak now, fall 90% by September, and then not resurge in the Fall (back to school and flu season). Of course, the best case COVID-19 scenario would have been for this never to happen. But for COVID-19 to fade in this manner, one of three things need to happen:
– cure or vaccine developed
– communities reach herd immunity at lower levels than 60%-70% prevalence;
– mitigation measures reduce community spread;
We got some vaccine news today from Russia which approved the world’s first vaccine. But this is met with some concerns as this was approved before all tests were completed. But there are so many vaccines/cures in the pipeline, that the odds of a successful cure is extremely high — I don’t want to say 100%, because nothing is ever 100%, but it is high.
https://www.nytimes.com/2020/08/11/world/europe/russia-coronavirus-vaccine-approval.html
But the potential for herd immunity reached at a much lower threshold, say 10%-20% vs 60%-70% conventionally, is nearly heresy for many experts. But growing research around T-cell immunity is supporting this possibility. This is discussed more fully below. Our adaptive immune system T-cells can recognize types of infections and neutralize them without specific antibodies. And multiple studies show that as many as 50% of the population has T-cells that recognize COVID-19, explaining potentially why many are asymptomatic and why some may not catch the disease. Thus, if case prevalence reaches 10%-20% and add in the 50%, you get close to herd immunity.
We are not healthcare experts, so take our commentary with a grain of salt. But in favor of this T-cell assertions is the collapse in cases in NY tristate + MA + RI, where case prevalence was >15% (serology) and Sweden (20% estimated prevalence) — in all cases, cases are falling and remain low.
We were asked by a client today why figures reported by various groups never sync up — COVID Tracking Project vs Johns Hopkins vs CDC vs World-o-meter, etc. We primarily use COVID Tracking Project, but our data science team, led by tireless Ken, actually extracts this same information from the state healthcare websites as a redundant audit. And this is why our figures might differ. But we have already found that the state data, because time of publication may not line up, leads to differences in the other sites. In other words, tireless Ken assures us our reported data is arguably the cleanest.
But Ken also points out the messiness of the data itself, due to reporting lags, variances in methodologies and even the mixing of serology and PCR results. Plus, multiple states, like FL acknowledge individuals are counted as infected multiple times — ala, a person testing positive 25 times is 25 cases. Not 1 case.
California backlog going to pollute the reported data…
As many are aware, California discovered several serious reporting errors with its COVID-19 tests, including a huge backlog of 300,000 tests that disappeared. For a state that is one of the cores of the US technology juggernaut, this is embarrassing.
– Today, CA is including this backlog to reported cases and it is going to cause a massive rise from the state.
– Today, the state reported 12,500 cases, a rise of +7,974 cases vs 7D ago (4,526)
Does anyone believe CA has a sudden massive new outbreak on their hands? I certainly don’t.
Source: Fox News
Expect us to start looking at case trends ex-CA for the next few weeks…
The state is going to be working its way through these missing tests, so this means the restatements will come out in batches over time. This is going to pollute the data for a while. So, we will be looking at data ex-CA and with-CA for the next few weeks. That way, you get the complete information but also understand the distortion.
STRATEGY: 9 reasons why epicenter rally could dwarf the May-June surge of 30%…
I spent most of my career and my earliest career as a wireless services analyst, so my personal experience is modeling hyper-growth industries and using cross markets to understand the impacts on equities (hence, why I always say stocks are the junior piece of the capital structure). So I am a believer of the superior returns of owning growth companies that capture value.
But at the same time, cycles are cycles and as we pointed out yesterday, the next decade 2020-2030 is probably going to belong to cyclicals and value.
Because of the short-lived epicenter rally in May-June, many investors are also looking at this potential epicenter rally with a skeptical eye. And this is perfectly understandable — after all, fool me once…
US Cyclicals, aka Epicenter, account for 62% of EPS growth in 2021, and are 24% of market cap…
Based on consensus, the epicenter is accounting for most of the EPS growth in 2021. And is only 24% of market cap.
9 reasons this rally would have legs…
But here is some reasons why an epicenter rally, which if it does ensue, could be far larger than the epicenter rally:
1. positive trend in cases far more convincing –> in April/May, while tristate was falling, rest of US surging
2. economy past the bottom and is now expanding –> ISM >50
3. bond proxies vs epicenter divergence far greater now –> since June, bond proxies/growth have exploded upwards –> mean reversion
4. US companies drastically lowered cost structure, so greater operating leverage
5. USD is rolling over –> rocket fuel for Value and Cyclicals
6. US corporate bonds have rallied further and GOOGL is borrowing $ at virtually same rate as US Govt
7. $5 trillion of cash on the sidelines
8. everyone expects this cyclical rally to last only 10 days
9. Americans are tired of “doom” all day and every day –> Americans are ready to get half-full
The Russia vaccine news today has sparked another bout of rotation FOMO. But we have to keep this in mind.
– If a vaccine/cure is developed, demand recovers nearly 100%
– costs are lowered, so realized profits far greater
Many clients have asked for a list of the “beaten” down stocks in the above categories. Once again, thanks to the efforts of our summer intern, CJ Woodberry, below is a list of Russell 3000 stocks within these categories (*).
* Please note that these stocks are solely compiled based on if their businesses have exposures in these 5 industries (Airlines, Cruise Lines, Casinos, Hotels/Resorts, and Theme Parks). Please consult your financial advisor to discuss your risk tolerance and other factors that characterize your unique investment profile.
Source: Fundstrat
POINT 1: For the next few weeks, we will look at results w/CA and ex/CA
As we noted above, daily cases in the US are going to be distorted by CA backlog. This is 300,000 tests were never recorded and that need to be adjusted and depending on the positivity rate, it could be tens of thousands of new cases. But these figures do not give a full picture and proper picture of the current trend.
– to adjust for this, when we look at daily case delta, we will show the total USA and total USA ex-CA.
Source: COVID-19 Tracking Project
As we have commented for several weeks now, we believe the best way to look at case trends is to look at 7D delta. This is a leading indicator of the 7D moving average (if <0, it is going down). And with the CA backlog adjustment, the 7D delta is positive.
– the lower chart is ex-CA and you can see here, the 7D delta is negative = good thing
Source: COVID-19 Tracking Project
Before someone gets alarmed and thinks we are “gaming” the data by ex-CA, look at positivity rates. These are pretty muted and well below 10%. And in fact, also look like it has been trending down.
Source: COVID-19 Tracking Project
Below are the 6 states with the largest 7D delta and topping this list is CA. GA is higher, but this state has seen lumpy reporting.
– 4 of 4 epicenter states, FL, CA, AZ, TX, or F-CAT, have seen a rise today. But the rise is modest vs surges previously.
Below are the 6 states with sizable 7D delta declines. And these states have been the ones many cites as having surges. So, this is good.
Source: COVID-19 Tracking Project
Trends in daily deaths have been good as well. The 7D delta is not surging. So again, flat is good.
POINT 2: Los Angeles cases drifting lower — supporting the notion of backlog distortion…
The 4 epicenter states of this outbreak are FL, CA, AZ, TX, or F-CAT and the nucleus are the 4 cities at the center of the breakout. Below is the updated chart, normalized by looking at daily cases per 1mm residents:
– Los Angeles is seeing cases drift lower
– So despite, the optical surge in CA, Los Angeles is well past its peak
– Look at Phoenix, wow… cases are utterly collapsing
Source: Johns Hopkins
Look at AZ county-level data. 77% of the state is seeing cases 75% off their peak. This is an utter collapse of the spread. If this is correct, the state should be commended for this spectacular result.
Source: Johns Hopkins
For the US overall, the percentage of the US with cases 75% off their peak is starting to ever slowly creep up. This is now 27%, which is the best level in about 6 weeks.
– this figure peaked at 40% in late June
– so 40% is the key level, and if the US can surpass this, the virus is under control.
Source: Johns Hopkins
POINT 3: T-cell immunity might be as high as 50%
Over the past few months, a number of clients have shared with us many interesting studies. And we have received several research pieces showing that our T-cells might play a bigger role in contributing to herd immunity and/or “infection break point.” Our immune system relies on both antibodies (produced by B-cells) as well as our adaptive immune system (T-cells) and the focus on research and commentary has been primarily on antibody-based immunity. But there is growing research suggesting that T-cell immunity is providing protection for individuals, and thus, limiting spread to a community. In addition, T-cell immunity can last for decades. So this type of immunity is less potentially less fleeting.
https://medium.com/microbial-instincts/if-antibodies-fall-short-of-covid-19-immunity-call-the-t-cells-abeb5cae6832
I am not a healthcare expert so I am sharing a diagram of how the body uses both T-cells and B-cells to ride itself of infections. For those with greater science backgrounds, this is probably more informative.
https://twitter.com/EricTopol/status/1278400526716211200
And this graphic just points to the idea that if we have innate immunity (T-cell based), our bodies basically deal with a virus before we need to call in the antibodies and the B-cells.
https://twitter.com/pbleic/status/1278357445023109121
One of the most recent commentaries on this T-cell immunity comes from James Todaro, MD. He is citing some research suggesting that once a community has 10-20% infection rate, some type of herd immunity is reached. The link to his thread is below. This is below the 60%-70% commonly cited, but his assertion (not proven) is that T-cell immunity is 50%, so the overall effect at 10%-20% is equivalent to herd immunity.
https://twitter.com/JamesTodaroMD/status/1292873236716433416
The research he is citing is a study published in Nature (see below) and this looked at the T cell immunity and the presence of T-cells in recovered and non-infected persons. The study found that 37 people tested had SAR-COV-2 specific T-cells but had no history of infection with either SARS or COVID-19. Thus, their bodies were previously exposed to some other coronavirus and this granted them some type of immunity to COVID-19.
https://www.nature.com/articles/s41586-020-2550-z
And apparently, if someone has these T-cells, it is possible that one develops a milder form of the infection. Todaro suggests this might be why there are a high number of young people asymptomatic.
https://twitter.com/JamesTodaroMD/status/1292873244236820481?s=20
And he cites Sweden as an example of how the infection break point is lower than many assume. Sweden never locked down like the rest of the World and yet, cases are diminishing rapidly there. But the serology tests suggest only 20% prevalence of COVID-19. So this level of prevalence is akin to some type of herd immunity.
https://twitter.com/JamesTodaroMD/status/1292873251853684742?s=20
This takes us back to this chart we have been publishing recently. It looks at what % of the counties in NY tristate and F-CAT have 12,500 cases per 1mm residents (1.25% confirmed rate and if we assume 10X infection rate, 12.5% prevalence). And notice, how this flattens out — meaning that at this level of prevalence, COVID-19 starts to slow.
Source: Fundstrat
This last point he makes is noteworthy, not because it is true. I believe most healthcare companies and executives want to improve lives, but this assertion just highlights the complex incentives. It is the same reason many experts dismiss steroids or HCQ as treatment regimens, because it could also not be that profitable.
https://twitter.com/JamesTodaroMD/status/1292873255540359169
