General update, May 10

Posted on by

1. In Connecticut,

Between April 22 and April 29, the state’s death total rose from 1,544 to 2,089, or 545 new deaths, according to data released by the state Department of Public Health. In that same seven-day period nursing home deaths rose from 768 to 1,249, meaning 481 among the 545 new deaths — about 88% — were nursing home patients.

We have a friend who lives in the highest-end elder-care facility in the St. Louis area. She is truly locked down. The front door is guarded, and the residents are not allowed out. They can outside into the back courtyard, but they cannot leave the facility. They can receive grocery deliveries. Meals are not eaten in the cafeteria, but instead are placed in front of the door, wrapped in plastic.

The nation’s governors are fighting on the beaches, they’re fighting in the fields, . . .and the real battle is in the nursing homes. Not our finest hour.

2. I recommend this Heying-Weinstein podcast from yesterday. I listened at 1.25x speed, because he talks slowly for my taste. From about minute 30 to minute 47 they discuss the issue of whether the virus likely originated in nature, came from a lab intentionally, or came from a lab by accident. I think they would put their money on “came from a lab by accident,” and their arguments make sense to me.

On another topic, they are not “openers.” Of course, they raise concerns about deaths from the virus, as do all “closers.” My responses are two fold. One is that it is not fair to assume that everyone will immediately revert to gathering in closed spaces and traveling as much as they were before. The other is that with so many of the deaths in nursing homes, I am not convinced that the marginal saving of life from locking down the rest of us is all that high.

One argument that they make as evolutionary biologists is that the more that we expose humans to the virus, the more chances we give the virus to evolve in ways that make it more adept at infecting people.

My problem with that argument is that it seems to me that a lockdown only stops the virus from adapting if we eradicate the virus. But the lockdowns we have are not going to eradicate the virus. Those of us who shelter in place need to eat, which means that there are people out there who need to move around to get food to us.

It seems to me that these two people for whom I have great respect have allowed themselves to fall into the “moving goal posts” fallacy. The lockdown that was originally sold as delaying the spread of the virus is now being defended as if it could eradicate the virus.

As a thought experiment, a total worldwide lockdown that is militarily enforced might be sufficient to eradicate the virus. But the partial, flatten-the-curve lockdown does nothing in the long run to deprive the virus of opportunities to explore mutation space.

3. Clare Malone and Kyle Bourassa write,

Almost uniformly across these states, people started staying home beginning on March 14. The percentage of people staying home rose rapidly over the following nine days and tended to plateau by March 23.

The Cuebiq data suggests that behavioral changes were largely driven by people making a voluntary choice to stay home rather than being forced to do so by a state-sanctioned stay-at-home order. One need only look at the behavior of residents in North Carolina and their neighbors in South Carolina: While North Carolina issued a stay-at-home order eight days before South Carolina, a stabilized number of people in both states started staying at home about a week before North Carolina’s order.

Pointer from Tyler Cowen. In mid-March, people were making decisions to restrict activity independently of what political leaders were telling them to do. I don’t think that has changed.

11 thoughts on “General update, May 10”

  1. In mid-March, people were making decisions to restrict activity independently of what political leaders were telling them to do. I don’t think that has changed.

    That cuts both ways. Why should the claim be, “People were locking themselves up early, so they’ll keep locking themselves up after,” instead of “People were locking themselves up early, and they’ll likewise also be early in opening themselves up.”?

    If they were locking themselves down prior to politicians ordering them to do so, then, so long as they think they can get away with it, why wouldn’t we also expect them to open themselves up, prior to when politicians officially allow it? That is, the claim, “official action is a lagging indicator” seems to me more plausible than “official action doesn’t matter,” a king of “Hansonian Law Enforcement” theory.

    I think that’s exactly what we’re seeing right now, and I’d bet those cell-phone location-tracking studies would show it. I’ve been to a bunch of big-box stores this weekend, many are packed with long lines and only 50% of people in masks.

    The beginning will not be the same as the end.

    Back in March, people didn’t know what to expect. They were encouraged to be terrified, and they were. Just because any particular government in the US had not yet imposed a lockdown, other countries were doing do, the stock market was tanking, and the news was harping on the topic 24/7. Scary! And going into lockdown, it doesn’t seem like it will be so bad, or last so long. It was cold winter, and people figured they were stuck inside anyway.

    Two months later, we know a lot more about who is at risk and who is not. We see things getting better or past the peak in a lot of places, especially internationally. The stock market is back up, and high status people are frequently talking about reopening. It’s spring and beautiful and people are itching to enjoy the out of doors and related gatherings.

    The point is that at the beginning, the government is closing the barn door after the horse has bolted. But at the end, the government is the dam barely holding back the flood.

  2. I think they would put their money on “came from a lab by accident,” and their arguments make sense to me.
    —
    I read a report, unverified, that some researchers took pets home.

  4. it seems to me that a lockdown only stops the virus from adapting if we eradicate the virus

    Getting infected can be viewed somewhat like bitcoin mining; there is a very small chance for each infectee of becoming host to a significant mutant strain of the virus*. The more people get infected, the more likely it is that our vaccines will be obsolete before they’re even invented. There’s a big difference between a virus that’s adapting too fast for medicine to target and one that isn’t, and limiting the number of infectees could easily be the difference.

    *and a 100% chance of maladaptive mutant strains, but they will be quickly outcompeted by the base strain and can be ignored.

  5. If you rank order the present case fatality rates (total deaths/total cases), the states with the lowest overall fatality rates are pretty much all the ones with low levels of cases in nursing home and long-term care facilities. The only outlier in the ranking is New York state which reports cases in such facilities that is essentially the same number as the deaths reported- in other words, their data is a reporting error in total cases. In short, if every state had done as well as Wisconsin has done with nursing homes, there would be half as many deaths attributed to COVID.

  6. My parents recently left a retirement community before this began, and it is as you described.

    One big problem is that even if you “lock these places down”…the work staff are all high transmission risk people IMO. Generally poor, living in dense environments, and taking public transport to work. The gap between what my parents were paying and what they were paying the staff was pretty jarring.

    Lastly, its difficult to get your (large) deposit back from these places in the best of times, but far more difficult right now.

  9. I have been using a 7 day death reproduction rate and Arnold’s assumptions of 9 days from infection to death and a death rate of 2% to estimate the number of cumulative cases for the last month or so. I have added an additional assumption of 14 days from infection to recovery. This assumption allows me to estimate the current level of “Active” cases (infections). A graph of these data can be found Here.

    In addition I have graphed a 7 day moving average of daily deaths with a 7 day moving average of the estimated infections above. That graph can be found Here. This graph indicates John Cochrane’s hypothesis that increases in deaths signal caution which cause s self isolation and eventually fewer active cases. Fewer active cases leads to fewer deaths which in turn signal less caution and more infections and eventually more deaths. The data and my assumptions suggest the second wave has already started.

  10. Thanks for a good set of think pieces, Arnold.

    On thinking about your “bitter regrets”, and possibly how to advise you so as to become more influential, I think you would be better off being more explicitly prescriptive.

    I’ve now seen a couple recent links to this piece:
    http://www.rationaloptimist.com/blog/we-know-everything-and-nothing/

    I think your “desired experiments” were being offered too soon – sort of like Apple’s Newton (pers. asst.). I suggest you revise your suggestions slightly and offer them again, explicitly as what Trump & Bill Gates should be funding. And try to generate some publicity about them, so they actually get done (which will happen sooner or later) AND that you get some credit, which takes work by you to push it and become identified as a public intellectual pushing it.

    Sort of like Miles Kimball for negative interest rates.

Comments are closed.