The virus in mid-July

1. The 7-day average daily death rate was about 430 a couple of weeks ago, but it is up to about 830 now.

2. Holman Jenkins in the WSJ writes,

The Denver Post interviewed a couple in their 70s who were risking the virus to resume weekend junkets to Las Vegas. Their only concession to Covid-19 was a plan to quarantine for 10 days before seeing their grandkids again.

Wow. Not what I would do at all. Imagine if we lived in a country where either I could impose my lifestyle on them. Or they could impose their lifestyle on me. Unfortunately, we live in a country where so-called “public health experts” are being granted such power.

3. I think that as an individual I could calibrate virus risk more reasonably if I had two pieces of information. The goal would be to enable me to have a better sense of my risk of getting a severe case, which we might define as causing me problems for more than 30 days; alternatively, we could define a severe case as requiring more than 48 hours of hospitalization.

One of the numbers that I want is what I call a “personal safety factor.” The scale would go from zero to one hundred. It would be based on a statistical model that combines age with polygenic analysis. My thinking is that the polygenic statistical analysis would pick up the likelihood of known risk factors (such as a tendency toward obesity) as well as other factors that are not currently known. In order to know my personal safety factor, I would have to have a DNA test. A personal safety factor of 90 means that I am in the top 10 percent, so that I am relatively unlikely to get a severe case. A personal safety factor of 10 means that I am in the bottom 10 percent, which means I really want to try to avoid getting the virus. If it turns out that I have a low personal safety factor, then I want to be in the front of the line to get a vaccine. Otherwise, I probably would be happy to wait.

The other number that I want is a “community safety factor.” This would estimate the probability that if I am in the same room with 50 people at least one of them will have the virus. If that probability is more than 10 percent and my personal safety factor is less than 90, then I am not ready to try a live dance session. Somebody else might be more risk tolerant.

4. Victor Chernozhukov and others write,

Our counterfactual experiments suggest that nationally mandating face masks for employees on April 1st could have reduced the growth rate of cases and deaths by more than 10 percentage points in late April, and could have led to as much as 17 to 55 percent less deaths nationally by the end of May, which roughly translates into 17 to 55 thousand saved lives. Our estimates imply that removing non-essential business closures (while maintaining school closures, restrictions on movie theaters and restaurants) could have led to -20 to 60 percent more cases and deaths by the end of May. We also find that, without stay-at-home orders, cases would have been larger by 25 to 170 percent

I would note that the economic cost of people wearing masks is considerably less than the cost of staying at home.

16 thoughts on “The virus in mid-July

  1. I think there is a lot of bullshit in all these estimates.

    The second wave has clearly been caused by the BLM protests. As many people note, they “wore masks” or at least many did. Wearing masks doesn’t make a packed mass protest safe. It just means that the droplets stay a little closer to you, it doesn’t get rid of them. Masks make sense for activities that *you can’t avoid*, which is exactly how they are used in Asia during flu season (like on mass transit).

    To the extent “wearing a mask” makes people feel invincible enough to engage in risky behaviors, it’s a completely counter productive notion. As I say to my parents, if you are in a situation were you need to wear a mask, you probably shouldn’t be in the situation if it can be helped.

    Beyond that, I find this whole notion that forced closure of nearly all of society had a benefit. I’ve yet to see any data that keeping playgrounds closed helped. In general, I suspect that the restriction of lots of safe (often outdoor) activities has increased risk (by funneling more and more people into the few things they are allowed to do).

    I’ve flip flopped on this compared to March. Back in March I was told that surfaces were potentially a big deal. If so, then yes the whole world was a threat. Now that we know that isn’t true, I think it’s time to drop all this nonsense.

  2. I think you missed one piece of data to inform your actions, possible morbidities from having covid. It appears that people who are asymptomatic and otherwise healthy can get neurological damage (predisposing them towards strokes), lung scarring, and heart damage. We don’t know the prevalence. I think that is one of the things that is missing in the conversation of Covid. Right now it is still being framed as you get it and one of three things happens: nothing, you go to the hospital and survive, you go to the hospital and die.

  3. What even is the goal at this point? Are we still flattening the curve on a ‘throttled’ path to herd immunity, or are we trying to push R0 below 1, or are we trying to keep new infections low-ish (how low? why that number?) to buy more time for vulnerable people until the vaccine comes out, or what? Is there a website where people can read what that is?

    These goals are not reconcilable: they yield entirely different answers in terms of both the optimal public health and economic policies, and they also lead to completely different interpretations of current events, “going as expected according to our strategy” on the one hand, or “disastrous” on the other. If enough ICU capacity is available to make sure that very few people who could be saved aren’t dying because of lack of availability of care, then should we be freaking out, or not?

    The new conventional wisdom is that there was a conspiracy among top leaders and public health officials in the US to discourage mask hoarding by lying to the public and telling people that “masks don’t work” – even knowing that the fact of this lying would become widely known in mere months and torch whatever credibility capital anyone had left. This was apparently judged to be the minimally obnoxious method available of sufficient power to ensure that there would be enough of the best masks for health care providers, researchers, and other top-priority workers in the short term, until production ramped up and sufficient PPE supply chains were re-established.

    The alternative could have been to level with the American public and tell them that the government was confiscating and commandeering all PPE with its eminent domain and emergency authorities, and that any other kind of sale or export was unauthorized *until such time* as production and supply could catch up, in which case masks would be rationed out in a fair manner with some preference for the most vulnerable populations. But in the mean-time, citizens were highly encouraged to use improvised facial coverings such as scarves or bandanas or cold-weather gear to help slow the spread of the infection, and that quarantines of all foreign travelers would be enforced strictly. It was not to be.

    The torching of that credibility capital has been disastrous. It was probably well-deserved anyway in terms of basic competence, but a few months of even misplaced trust would have gone a long way.

    When no one believes that expert authority is both (1) Honest, and (2) Competent, then they are left to fall back on their own instincts and impressions and judgments, and in the nature of human psychology that means (A) A tendency to go to extremes of either insouciance or paranoia, and (B) convergence to the consensus of one’s political coalition, which in polarized contexts in which one party is trying to make electoral hay and not let a crisis go to waste, means whatever the opposite of what the other other party is doing.

    • Credibility in our government experts is now gone. The diehards are still clinging to the illusion that their side has maintained some credibility. In reality, the response to COVID-19 has been a massive failure on both sides. Masks don’t work, then they work. Large gatherings are discouraged, then they aren’t (then they are again). Borders should be closed, then they shouldn’t be, and I guess we are back now to where they should be. Businesses are closed, then opened, then closed again. Law is enforced, then it isn’t.

      There is an alternate reality where we could have seen leadership and not political polarization. But it ain’t gonna happen.

    • The first and most obvious goal at this point is to not allow local healthcare systems to be overrun. When that happens unnecessary deaths from both covid and non-covid causes will skyrocket.

      Beyond that there are a series of difficult trade offs available where desirable goals conflict with one another in ways that make outcomes difficult to predict. You don’t need to invoke a conspiracy to understand why many mistakes will be made in good faith at a point where many of the most important facts about a novel virus are unknown. There was no question early on that there was a critical shortage of PPE for healthcare professionals. There was a question at that point whether or not the public would benefit from mask wearing. As better data came in, scientists and healthcare professionals corrected their guidance appropriately. Politicians much less so.

      The virus does’t care about your politics. We are running any number of inadvertent natural experiments that will ultimately reveal much about the virus we do not currently know. Who has been giving advise in good faith will be readily revealed by who is most willing to correct their inevitable mistakes in a timely way.

    • The collective failure of public health authorities to have consistent and honest messaging was the biggest failure of the response to COVID.

  4. Arnold Kling, what is your position on reopening K-12 schools? And particularly, K-5? Can it be done safely with face mask rules and temperature checks? Or is it not safe to reopen K-12/K-5 schools with the recent spike in cases?

    People cite school reopenings in Europe as evidence in favor and school reopenings Israel as evidence against.

    Also, there is evidence that the risk to children is lower than the normal flu, and that children very rarely infect adults, making that a low risk as well.

    This has become a left/right issue: the right favors reopening schools. the left opposes. This issue doesn’t seem to be inherently partisan.

    • “Can it be done safely” is a meaningless question. It all depends on what you mean by “safely”. If it means “there will be no transmission (zero, zilch, nada) in schools”, then no, it can’t be done safely.

      If “some” transmission is okay, you have to define how much. But lots of people think talking that way is mean and inhuman.

      The way lots of people talk now, “safely” seems to mean “zero transmission” or “experts will say it’s safe and I won’t ask exactly what they mean by that”.

  5. Here are two spreadsheet sets I created from the COVID-19 Tracking Project and have been updating every day.

    United States COVID Data
    Individual States COVID Data

    In the top link, in the tabs at the bottom, you can find different graphs for the 7 Day moving averages of new tests, new cases, and new deaths, plus the daily deltas of all three. Also included are the cumulative positive test rate and the 7 day moving average of the daily positive rate (7 Day New Cases/7 Day New Tests).

    The bottom link has every state and D.C. in its own separate spreadsheet- tabs at the bottom. Included in the the individual sheet is the 7 day average of new cases and new deaths for quick visual of the trends in each state.

  6. I think the experience of the entire world is that reducing R nought to less than 1 will probably not extinguish the virus until pretty much everyone it can infect has been infected- masks, no masks, shutdowns, no shutdowns. Also, I think one of the things that probably isn’t every well understood by anyone (and I didn’t fully understand this at the start) is that these respiratory viruses never disappear. They simply evolve to stay alive in the population. In other words, even with a vaccine, updated every single year, COVID will continue to survive in the population essentially forever going forward. Most likely, survival pressure will reduce its lethality over time to something more akin to other beta coronaviruses.

    In short, other than flattening the outbreak to no overwhelm the hospitals, I can’t see a single benefit to the shutdowns or face masks, and if there are actual benefits, then why not make face masks mandatory from now on- there are always new influenza viruses to avoid every single year- and then shut down the economy and the schools every time there is a measurable outbreak of influenza? Basically, we have gone insane the last 5 months.

    • I don’t think there is much pressure for the virus to become less virulent. Killing 70 year olds to spread more easily among 20 year olds is good trade off. And even then the older infected victims seem to be able to spread the virus as well before they become really sick and go into the isolation ward. That is just as good as a regular cold virus that makes a person sick and infectious for a few days before being cleared from the body.

  7. Tyler Cowen linked to a tweet (below) that supposedly showed CDC data with respect to when deaths have been recorded and when they in turn report them. The chart shows that from July 7 there has been a large reporting of deaths that occurred in April, May and June so that there hasn’t been a rise in weekly deaths – just a rise in reported weekly deaths.

    I’m not sure that is correct but have been tracking cases and deaths since the pandemic hit and noticed that the daily deaths shot up for Tue, Wed, Thu (which normally are the highest) from 300, 700, 600 before the Fourth of July holiday
    down from 900, 800, 600 the previous week to 1,100, 900, 900 and then most recently 900, 1,000, 800.

    It doesn’t make sense for those three days in mid-week to very slowly decrease over many weeks and then shoot up by 80% in just a week.

    Marginal Revolution #1
    https://marginalrevolution.com/marginalrevolution/2020/07/sunday-assorted-links-272.html

    1. … … “Here is a claim that the increase is a reporting spike, due to previously unreported past deaths. “54% of deaths reported in the week of July 11 ( if her data is correct) actually occurred over 2 months earlier.” At the present time I am unable to confirm the actual distribution of numbers one way or the other — opinions and leads welcome!”

    • As I pointed out in that comment thread, what you need is the granular data from the states themselves. As one can see at the very top of Lamb’s table, the deaths seem to be reported out to the public on a timely basis, but it takes the CDC several weeks to take that data and update their own table of deaths for a given week. I did total the very top and the CDC reported deaths from the beginning to the end, and for a while in May, the CDC total ran almost 30,000 deaths behind the public reports, but then narrowed each week after that to less than 10,000 by July 11th.

      So, the question is- in the state’s public reports, captured by the COVID Tracking project, how many of each day’s reports today are deaths that occurred prior to July? I don’t know the answer, and I wish I did. You can only get a glimpse of this when a particular state reports a big batch of new deaths that are way above the trend, like NJ did in the last week of June and again in early July- otherwise the process is opaque.

  8. Are the authorities in some states really over-reacting?

    For example, Los Angeles county had 22 Covid deaths last week. It has a population of almost 10 million.

    Isn’t that a tiny, tiny event to justify another stay-at-home order?

    I do not share the perfectionism of public health officials. Of course I can be wrong though.

  9. Missing a risk variable, medical response. Being three months sickly is enough time for the medical industry to possible come up with an inhaler made of artificial antibodies or some other antigen to the covid. Three months for some kind of guestimate is a good number if I read the news on this correctly. Three variable, relative rate of sickness, relative rate of infection, and relative rate of better medical products. In equilibrium we are breathing our inhalers according to our activities.

Comments are closed.