Previously: Covid 10/8: October Surprise
Last week the President got Covid-19, so the post was all about details and timelines. This week, the President has tested negative multiple times and my town closed its playground due to mask non-compliance, so let’s take a step back. A summary of the story so far, if you will, so that we can imagine where it will likely go from here.
A deadly pandemic falls upon the land. Hundreds of thousands or even millions may die. Huge percentages of all businesses may close, a large fraction of jobs may be lost, perhaps permanently. Social isolation threatens to radically decrease quality of life well beyond the loss of material prosperity. Tens of millions cannot pay their bills and face eviction. Impossible choices must be made.
Once all this is clear, the people divide into two camps, mostly along existing partisan lines.
One group claims their goal is to prevent economic disaster while saving lives. In practice, they act as if their goal was to have everyone ignore the pandemic, actively flouting all precautions and discouraging any countermeasures that might help.
The other group claims their goal is to save lives while preventing economic disaster. In practice, they act as if their goal was to destroy as much economic and social value as possible in the name of the pandemic as a Sacrifice to the Gods, and to pile maximum blame upon those who do not go along with this plan, while doing their best to slow down or block solutions that might solve the pandemic without sufficiently destroying economic or social value.
Is that unfair? Look around you. Think about all you’ve read, seen and heard since February. In this space and elsewhere. Decide for yourself.
How could this be? Systems that reward loyalty to the ingroup reward symbolic destructive action, while being suspicious of and therefore punishing constructive action that might be motivated by things other than loyalty to the ingroup. Morality has been reversed. Read my sequence on immoral mazes and my posts on simulacra levels for further explanation.
Thus a continuing dialectic that solves neither problem. That which favors death of things symbolically associated with the outgroup wins out. That which favors life has loyalty to life, so kill it. To stop the horrible outgroup, one must support the ingroup that much more. The dialectic continues.
The pandemic has enough physical-world, simulacra-level-1 impact on people to steer most ordinary people’s individual physical actions towards what seems to them like useful ones that preserve economic and social value while minimizing health risks. And it manages to impose some amount of similar restrictions on the collective and rhetorical actions.
That doesn’t mean equivalence between sides, let alone equivalence of individuals. But until the basic dynamics are understood, one can’t reasonably predict what will happen next.
I predict that next we will run the numbers.
I’m changing the order of the numbers from here on out, to avoid starting with the lagging indicators.
Positive Test Counts
|Aug 20-Aug 26||67545||66540||132322||18707|
|Aug 7-Sep 2||55000||75401||127414||21056|
|Sep 3-Sep 9||47273||72439||106408||21926|
|Sep 10-Sep 16||45050||75264||115812||23755|
|Sep 17-Sep 23||54025||85381||127732||23342|
|Sep 24-Sep 30||55496||92932||106300||27214|
|Oct 1-Oct 7||56742||97243||110170||34042|
|Oct 8-Oct 14||68284||125744||117995||38918|
That is a huge jump in the Midwest, which has the most cases of any region for the first time, and the Northeast and West counts also are clearly headed in the wrong direction. The Northeast jump is small week over week, but it affirms that last week’s jump was not a fluke. This is a lot of new cases.
Before jumping to conclusions, of course, we must look at the positive test percentages as well, to ensure we aren’t mostly doing more testing. We definitely are doing somewhat more testing.
|Date||USA tests||Positive %||NY tests||Positive %||Cumulative Positives|
|Aug 6-Aug 12||5,121,011||7.3%||506,524||0.9%||1.58%|
|Aug 13-Aug 19||5,293,536||6.2%||548,421||0.8%||1.68%|
|Aug 20-Aug 26||4,785,056||6.0%||553,369||0.7%||1.77%|
|Aug 27-Sep 2||5,042,113||5.5%||611,721||0.8%||1.85%|
|Sep 3-Sep 9||4,850,253||5.3%||552,624||0.9%||1.93%|
|Sep 10-Sep 16||4,632,005||5.8%||559,463||0.9%||2.01%|
|Sep 17-Sep 23||5,719,327||5.2%||610,802||0.9%||2.10%|
|Sep 24-Sep 30||5,857,097||5.1%||618,378||1.1%||2.19%|
|Oct 1-Oct 7||6,025,633||5.2%||763,935||1.3%||2.29%|
|Oct 8-Oct 14||6,324,833||5.8%||850,223||1.1%||2.40%|
Despite the additional testing, and New York doing an ever-increasing share of that testing, positive test percentages clearly took a large step in the wrong direction, including a 9.8% positive rate on 10/14 outside of New York. That happening yesterday is of course scary, but it is also a single day of data that is the majority of the increase in positive percentage, and that could still be a data artifact. Time as always will tell.
The news in New York itself is very good. Swift action was taken to attempt to contain the flare ups within hot zones, and for now that seems to be working.
Now let’s separate all of this into regions.
Positive Test Percentages
|8/13 to 8/20||2.06%||5.62%||9.41%||6.47%|
|8/20 to 8/26||1.86%||5.78%||9.93%||5.88%|
|8/27 to 9/2||1.87%||6.37%||9.38%||4.78%|
|9/3 to 9/9||1.97%||6.02%||8.48%||4.13%|
|9/10 to 9/16||2.41%||5.99%||11.35%||4.49%|
|9/17 to 9/23||2.20%||5.96%||7.13%||4.11%|
|9/24 to 9/30||2.60%||6.17%||6.18%||4.27%|
|10/1 to 10/7||2.61%||6.05%||6.74%||4.23%|
|10/8 to 10/14||2.57%||8.14%||7.09%||4.75%|
The Northeast’s push to increase testing makes it look stable here, but this chart confirms that the alarming Midwest number is real and the slide in the West is real as well. The South also seems headed in the wrong direction. The Midwest has the highest positive percentage on tests, not merely the most infections.
If you do live in the Midwest, check your particular state. Some like Illinois are mostly holding together, whereas others like Indiana most certainly are not. This could be the time to hardcore hunker down for a while, or it might be mostly business as usual.
|Aug 13-Aug 19||1576||850||4264||422|
|Aug 20-Aug 26||1503||745||3876||375|
|Aug 27-Sep 2||1245||759||3631||334|
|Sep 3-Sep 9||1141||771||2717||329|
|Sep 10-Sep 16||1159||954||3199||373|
|Sep 17-Sep 23||1016||893||2695||399|
|Sep 24-Sep 30||934||990||2619||360|
|Oct 1-Oct 7||797||1103||2308||400|
|Oct 8-Oct 14||782||1217||2366||436|
The trend in the Midwest continues and is even more clear than last week, with the same true to a lesser degree in the Northeast. The deaths are starting to show up. The West and South, meanwhile, repeated their numbers from last week, which is disappointing but neither terribly scary nor all that surprising. One can of course say that lagging indicators are still lagging, and the deaths are indeed coming Real Soon Now.
Probably a moderate increase in deaths is coming in the medium term. The whole “additional 200k deaths by year’s end” is almost certainly nowhere close to happening, as there are only 11 remaining weeks in 2020 (doesn’t that sound lovely?) and zero indication we will hit anything like that pace. That of course is point-scoring rather than of practical use.
The practical implication is that the personal risk of dying may or may not get worse, or get worse in your region. It almost certainly will continue to get worse in the Midwest. There are reports that states such as Wisconsin have already run out of hospital beds, and things have at least a few weeks of getting worse before they can plausibly get substantially better.
Thoughts on Various Causes
A key worry is that Trump’s messaging has shifted. He’s claiming the pandemic is effectively over, and that the virus is no big deal. Don’t let it control your lives and all that, including a number of stronger comments than that.
It is too early to tell if this messaging is having much impact. Depending on when you count from, it has been at most a week since this started in earnest, so any changes in behavior would only now be beginning to show up in the numbers. That’s another reason yesterday’s number in isolation seems scary.
As noted last week, if this has a big impact, we should expect things to get much worse in red areas compared to the situation in blue areas. So by next week’s report, it should be clear whether this is scary or not.
Schools continue to be an obvious potential factor, as compounding infection rates take their tool over several weeks and problems spread out beyond the schools themselves. I have a section on this later on for colleges.
Cold weather has always been a fear, and would explain the Midwest starting to have a hard time, but objectively things have not been all that cold yet. Midwesterners don’t scare off this easily, as anyone who listened to A Prairie Home Companion growing up knows well. Wait until December and check back in. I don’t think it’s that obvious even which direction this ends up going on net, given the incentive to stay home entirely.
What this latest increase matches most, to me when I look at it, is basic fatigue over the whole thing. It’s been many months, the virus is steadily showing itself to be less deadly over time, and people can only put their lives on hold for so long. Gradually, they’re learning to live with the risks. This doesn’t have to be complicated.
CDC is newly warning that an increasing percentage of infections are coming from small private gatherings, rather than from the public sphere. Their operative model is that compliance in public has actually mostly been pretty good, but in private people are meeting their friends indoors without masks. This mirrors what I have seen from several people I know, and makes sense. The precautions in public are something you get used to, and we can use social pressure to enforce them, and everyone has reason to worry about strangers. But wearing masks around your friends, in private, feels unnecessary, is much more annoying, and can feel like an insult. Over time, it becomes increasingly painful not to see one’s friends, and not to be able to relax around them. So I do think they’re probably right about this, and we should be reasonably worried about Thanksgiving and Christmas.
FiveThirtyEight looks, finds that voting in primaries didn’t have substantial impact on the pandemic. Given the physical dynamics involved this is entirely unsurprising. Don’t be afraid to vote in person.
One mistake people may be making, and that you too might be making as it is a very easy thing to overlook, that in turn is causing things to spiral a bit out of control, is failing to adjust their personal risk for the herd immunity level. What this means is, if (as I presume) roughly 15% of Americans are currently immune due to previous infection, and they (again, as I’d guess) even before immunity engaged in general in behaviors that were far more risky than average and lived in riskier places, then that is cutting new infections by substantially more than 15%. Thus, if we know how many people are infected, this increases your effective risk by the inverse of that factor, which will also be well above 15% – if we know 50,000 people a day get Covid-19, then those infections are all coming from the pool of vulnerable people. My guess is that this effect is at least a 30% or so increase in real risk for those not yet immune. Adjust your risk estimates and risk tolerance accordingly!
Playtime Is Over
What made me finally write that long-overdue introduction? Let me tell you a little story.
Warwick has a wonderful gigantic playground. For months, we would often take our children there around 8am. It was reliably deserted, and we were able to give them a great experience in complete safety. Slowly between 9 and 10, a few other families would show up, and things would slowly become more crowded. Since we are going above and beyond in terms of safety, we made sure to always get there early and leave before we’d have to be around anyone else. But the times we checked later in the day, things were never tightly packed or actively worrisome – we just preferred, given the option, to play it safe.
One day there was a sign warning that there had been – gasp – mask violations. And warning that if these mask violations did not stop, they would have to close the playground. We were annoyed, and yes kids sometimes didn’t wear masks at the playground largely because they obviously didn’t need to, but did not think much about it.
Then last week, we show up and the playground is closed. They actually did it. Now they’re talking about maybe reopening it, with a guard outside to enforce mask compliance and also check Warwick IDs, which we’d have to go get since we never bothered, given how much of a pain getting them is under pandemic conditions. So we’d have to prove to authorities, who would then be watching over us enforcing needless mask requirements on fully isolated three year olds while doubtless themselves being a huge percentage of our actual risk, that we were following all the official rules and were properly authorized to go there.
Meanwhile, we had no choice but to travel half an hour to another playground, which we then had to crowd into, taking far more risk than we would have in Warwick while imposing extra risk on others. As other families also doubtless did. This might well considerably accelerate the timeline of our return to New York City, where the playgrounds are an order of magnitude more crowded.
Children and parents suffer, basic childhood experiences are damaged or lost, Covid-19 risk goes up, but Very Serious People get to feel like they punished the wicked who sinned by failing to offer their proper sacrifices sufficiently reliably when called upon to do so. Meanwhile, the town permits indoor dining.
Yeah, I’m pretty pissed off about the whole thing. Thanks for noticing.
San Francisco, it seems, has kept its playgrounds closed this whole time, because they don’t know how to show ads on them or sustain a subscription model. But they’re proud to announce they’re getting bold and are ready to reopen them next week, maybe because now it’s cold enough that they hope enough kids won’t want to go. Requiring two year olds to wear masks can help with that, as has the massive well-deserved and long-overdue exodus out of the area.
Maybe Lockdowns Have Negative Consequences
This week’s most surprising news was that the W.H.O. came out with a statement condemning lockdowns as unethical. The killer quote:
Dr. David Nabarro from the WHO appealed to world leaders yesterday, telling them to stop “using lockdowns as your primary control method” of the coronavirus.
He also claimed that the only thing lockdowns achieved was poverty – with no mention of the potential lives saved.
“Lockdowns just have one consequence that you must never ever belittle, and that is making poor people an awful lot poorer,” he said.
File under: Things that could have been brought our attention yesterday. Or, you know, MONTHS AGO.
Also emblematic of this style of thinking, where anything but the official prescribed sacred values and the official prescribed actions to promote them over other things is unethical and makes you horrible.
Still, thanks for sharing, W.H.O. We appreciate your belated honesty and willingness to find a frame that allows you to notice that destroying people’s lives and livelihoods might have negative consequences and that this might be something bad instead of a symbol of our righteousness.
In case you thought there was hope for them yet, of course, for balance, they also condemned herd immunity strategies as unethical.
So if it isn’t ethical to allow the virus to spread, nor is it ethical to lock down your population to stop it, then it’s… underpants gnomes time?
A Time and a Place For Everything, and It’s Called College
This chart from the United Kingdom is certainly quite suggestive:
So you’re saying there might, just might, be a reasonably simple and direct explanation for a good portion of what’s happening? No way. Couldn’t be.
Another hint, and a very good sign if we ever decide to do challenge trials and are worried we wouldn’t get volunteers, would be this story about a student seeking deliberate infection in order to get boosted payments for their plasma. Seems like this would actually be net likely to save lives, if proper isolation measures were taken after infection.
Whether or not lower schools are an important vector, it seems clear that colleges are a major vector. It’s not clear this should worry us or even necessarily be a bad thing. It’s a reasonably efficient way to get herd immunity, and hopefully predicts further declines in death rates.
Certainly I can’t fault the students. If permanent damage is sufficiently infrequent, it makes sense that college students would opt for a college education at the cost of a to-them mostly harmless infection, and if we acted sensibly we could mostly contain such problems (including keeping older professors remote) slash the infections would burn out within a short time anyway if the above graph is any indication. We ask our young to sacrifice four years of their lives while going into debt for some combination of a zero-sum signaling game, networking and learning. Now they should give up the networking and much of the learning, while still going into debt and losing those years?
Quick Note on Election Odds
Market prices on the election are increasingly nonsensical. Election is not over and anything could happen, but I have stopped primarily doing more arbitrage and am now primarily betting on Biden despite my suspicion that I could get an even crazier price two weeks from now. That is all.
In Other News
McConnell admits he’s wisely been avoiding the White House for quite some time out of concern for his own health and that of the senate.
Firms are so terrified of raising prices when demand is high that we are looking at a shortage of appliances including refrigerators.
Another trial says HCQ doesn’t work. Old news and I wouldn’t bother if I didn’t still occasionally see European sources claiming everybody knows it works.
Nick Saban, coach of the Alabama Crimson Tide, and his wife, along with Alabama’s athletic director, have tested positive for Covid-19. So far as we know they are both doing fine while isolating, and presumably helping as best they can for the Big Game this weekend against Georgia, which you should watch if you enjoy some football. So have Power 5 head coaches Mike Norvel and Les Miles, plus Colby Carthel at Steven F. Austin. That’s a reasonably high rate among coaches, who are the people most likely to both get regularly tested and to reveal their positive test to the public. That seems like a useful data point to keep an eye on.
The NFL is struggling with positive tests on various teams, but managing to muddle through via rescheduling. I have no doubt that the NFL is doing less than they should have to keep players tested and protected during the season, but it also seems like cancelling the preseason to protect players was a large error. Injuries this year seem dramatically up due to lack of such practice, and pose a much bigger health risk to the players than Covid-19. Seems to fit the general pattern.
In Other News: Stopping Insufficiently Painful Solutions Because Be Afraid
We are definitely not telling anyone that Previous infection with other types of coronaviruses may lessen severity of COVID-19.
Immunity Still Ending Real Soon Now, Now More Than Ever, Say Experts, Because Be Afraid
Trump says that because he got infected and recovered, he is now immune to Covid-19. He further claims that because he has had the virus, there is an anti-Trump conspiracy to deny that people who have recovered are immune.
This is interesting for several reasons. It shows that Trump is so unaware of ‘health expert’ claims about Covid-19 that he did not realize that the Very Serious People were making this ludicrous claim that immunity will end Real Soon Now ever since day one, long before he got infected. It shows that Trump’s mindset is such that any claim that is bad for him must be an anti-Trump conspiracy, or (more precisely) that his systems output that he would benefit from accusing such statements of being part of an anti-Trump conspiracy. In this case, the claim of non-immunity is so ludicrous that he actually can’t fathom that the claim was being made the whole time and has nothing to do with him.
To be fair to Trump, it does have something to do with him, because with him in charge there’s more incentive to say things that make the situation look worse and that scare people, but that would be the way things would lean anyway. It’s hard to tell how much this changes the effect size.
Even without him, we’d probably have even the ones trying to get this right having statements like you find here, posted on October 14:
For months now, people have been watching closely to see if it’s possible to get re-infected with the coronavirus. It’s taken a while for the signal-to-noise to get better, but by now there’s no doubt that the answer is yes, it’s possible. We’ve just had the first of these in the US, a man in Nevada who was infected twice six weeks apart, with the second round being worse than the first. And in the Netherlands, the first fatality from a reinfection has been reported. All this sounds immediately like bad news, but I’m going to break out the same advice I was handing out yesterday: don’t panic.
Why not? Because from everything we can see, re-infection is a very rare event. The confirmed examples worldwide could possibly be counted on your fingers (depending on whose count you believe) out of at least 38 million total cases.
All of that sounds like great news! We are now in month nine and the Netherlands reported the first fatality from a reinfection? We have one confirmed case in the United States? Oh no, it’s possible.
It’s also possible that I will be struck by lightning and then elected president in 2024 on the Libertarian Immortalist ticket using the classic slogan “An End To Death and Taxes.”
There’s also a new study out that shows that not only does immunity last a minimum of seven months, concrete physical evidence of immunity in an individual patient also reliably lasts seven months or more. There is broad agreement that immunity only counts if it can provide antibodies that are measured on tests on an individual level, you see, and ‘almost no one getting reinfected ever’ does not count as evidence.
The bottom line is very very simple. If you have previously had Covid-19, the correct play for everyone is to act as if you cannot be reinfected, until such time as there is a very strong reason to believe you’ve somehow been reinfected, until such time as we see several orders of magnitude more confirmed reinfections. Period.
There seems to be a general rise in conspiracy theory in general these days, on all sides. Given how difficult it is to fathom what is really going on, and to understand that what is really happening is not actually a conspiracy at all but something arguably far worse, I am highly sympathetic, and would encourage my readers to be sympathetic as well.
Low Death Rate May Also Be Ending Real Soon Now, Because Be Afraid
Washington Post says: Covid-19 mortality rates are falling worldwide, but scientists worry that may be temporary. I see actual zero reason, other than a pure desire to promote Fear, Uncertainty and Doubt, to design the headline on such a story in order to scare people. Which in my model is exactly what is going on, in order to be seen as maintaining people’s compliance with countermeasures, and helping to justify further economic destruction in the name of containment.
It would be highly irresponsible, you see, to give people the amazingly great news that the disease is no longer as deadly, and thus that people might want to make different decisions or set different policies based on this change.
Rapid Antigen Tests Still Coming Real Soon Now
Reuters reports that Germany, Italy, Canada, the United States and others are purchasing lots of rapid antigen tests for only a few dollars each. Which is great news. Alas, based on how these things work, I expect them not to be deployed for some time at best, and likely used inefficiently, since they offer help without sufficient sacrifice and directly threaten the status of medical professionals.
AstraZeneca Trial Remains Partially Halted, Because Be Afraid?
Tyler Cowen raises an interesting question. Vaccine trial by AstraZeneca gets halted due to illness in individual patients. Another trial of the same vaccine in Britian has multiple adverse effects, but continues. “Public health experts” praise the halting in America for its responsibility, despite it in expectation causing more deaths and also massive economic damage via delays in vaccine availability. Yet they do not criticize the British for doing the obvious right thing and allowing their trial to continue. Of course, it’s the same vaccine, so if it’s possible that trials should halt (which they definitely shouldn’t) then either both trials or neither trial should halt. So, as he says, given the disparate reaction, solve for the rational Bayesian equilibrium.
Tyler offers seven potential explanations, most of them implausible at best. I see the most plausible of his offerings as his #3, but I think it’s off a bit. As a Public Health Expert and thus Very Serious Person, they can never criticize someone acting ‘responsibility’ by going against their own economic interests by not taking action.
What I don’t buy is that they think that it is fine for the British to keep going. This is a category error. There is no opinion on what is fine or not fine, or any model of physical consequences at all. When a corporation halts its trial over something trivial, that is ‘responsibility’ and ‘safety’ and very in-group aligned. The more it destroys value, the more confident we can be that it represents and signals loyalty. Thus, we praise it to the extent we can, and avoid any criticism.
What is trickier is why they aren’t criticizing the resumption in Britain, or the resumption in Brazil. I see a few explanations. The ‘free rider’ theory where non-Americans don’t really count so who cares has some merit, but I doubt it is primary. One element is that the Very Serious People score points for themselves and reinforce their power via praising the original halting decision, which would be threatened if they then criticized the resumption. In particular, if they publicize the resumption, they’re calling their preferred rules into question on multiple levels and pointing out how insane they are, so they’d prefer to avoid that. They’d also be criticizing the Very Serious People in charge of such matters on health boards in Britain, and they’re committed to maintaining such people’s reputation as part of the general support structure. They can’t go after AstraZeneca without going after those who agreed to let the trial resume.
Also, pointing out the partial resumption would potentially cause the United States to allow things to resume. Trump in particular, if he were to notice, would potentially apply pressure towards this end. Which would lead to American-based data being gathered, which as we all know is the data America must trust most. And we can’t have that, now, can we?
And again, none of this has anything to do with the actual value or cost or consequences of any potential actions, or any actual moral or ethical considerations, let alone any Bayesian calculations or expected values.
Monoclonal Antibodies Seem to Dramatically Lower Death Rate, So Be Afraid Of That Too Somehow
Trump getting treatment, as we all know, is terrible and cause for alarm – it might have worked, and it also might not have worked, and either way that would have been terrible and we must clutch at the pearls of our medical ethics. Plus, as Regeneron’s CEO says, Trump’s recovery is the weakest evidence that their monoclonal antibodies work. His failure to recover would have been the weakest evidence against them working. Of course, as Bayesians, we recognize that a natural N=1 experiment, even with a lot of noise to its boolean result, still has a non-trivial likelihood ratio, but far from enough to go from ‘probably doesn’t work’ to ‘probably does.’
The thing is that the other evidence is much stronger, and it increasingly seems to all point in the same direction. This is for real. Great, right?
Wrong! It’s a logistical nightmare! For starters, we’d have to actually use medical facilities to properly administer this medicine. Then comes the real issues. We have a way to cure some patients but not all patients. How are we going to ration this? Who decides? What if affluent or influential people get the inside track, and the whole thing is – gasp – unequal? We have patients who can benefit from the drug but can’t afford to pay money for it, what happens to them, huh?
I wonder to what extent such things are “this thing is now associated with Trump so we need to ensure it is viewed as maximally awful,” to what extent such this is “if someone has two cows then what you have to do is kill both cows because there are three people so that now we all don’t have a cow equally, man” and to what extent it is noticing that there might be a solution that doesn’t involve sacrifices and thus this solution must be stopped. Certainly all three play a role.
The solution to “we have created a limited supply of something valuable that helps people and there is now lots of demand for it from people with varying willingness/ability to pay for it, oh no this is an impossible problem whatever shall we do” is left as an exercise to the reader.
Deaths of Despair
On Twitter, Mason points out that in San Francisco, the so-called deaths of despair are both up 60% year over year and dwarf Covid-19 deaths four to one. There are probably more additional deaths of despair in San Francisco since Covid-19 came around, than there are Covid-19 deaths. That doesn’t automatically mean mistakes have been made, but it helps underscore the terrible human cost of our continuing to stall for time long after such strategies stopped making any sense. And also their playgrounds have been closed this whole time, so I’m pretty sure mistakes have been made around those parts regarding lockdown severity.
Don’t despair. Yes, it can feel like this pandemic is never going to end. But that’s probably not true, and probably not true sooner rather than later. Immunity and testing continue to increase, our treatments continue to improve, and vaccines are probably on their way on a timescale of months. Despite the best efforts of both camps, it would greatly surprise me if we are not past the halfway point.
Aaron Gertler is helping to crowdfund a prize for this series of posts (not eligible for the EA forum prize since it doesn't get crossposted there). You can PM him for details.
It would be a lot easier if Aaron could just post a link here to the fund, instead of everyone having to PM separately.
There's no link per se. Arose from discussion on the EA forum
Zvi, I really appreciate your posts on COVID.
One of the things I notice in your style is your recurrent meme of "sacrifices to the gods." This seems like a classic example of the "dark arts." It's using shame and uncharitable argumentation to tear down your opponents' argument and advance your own. It's unusual to see in writing here.
On the other hand, when confronting Very Serious People, who aren't interested in listening to a contrary argument that might damage their status, perhaps this is necessary. "You're not listening to me, so I'm not going to make your (stupid) argument for you: I'm just going to call you out." After all, I agree with you in your assessment of these arguments. They're pretty stupid. And it's a relief to hear them described as such.
Sometime I'd love to hear your thoughts on the tension here - if you see it as tension at all. Thanks for your continued work.
The model of "sacrifices to the gods" is a very specific gears-level model of how decisions get made. It's a relatively straightforward application of Hanson's theory of costly signaling, and as such has a lot of content to it that can be proven or disproven.
I think it's really important to distinguish arguments that are tribal and act on an associative level, which I think are generally bad, from arguments that operate on a gears-level and make concrete predictions, even if those predictions are uncharitable. The first one does often seem bad and to deteriorate discussion, but the second one seems quite important and usually moves discussion forward. It seems to me that Zvi's use of "sacrifices to the gods" falls into the second camp here.
You are right.
For others, here's what I talked myself into while trying to argue with habryka/Zvi in writing this comment:
Sacrifices to the gods are an ineffective or harmful waste of resources. People make sacrifices because they expect that their sacrifice will earn them status points with their in-group.
Why does a sacrifice need to be portrayed as an effective action? Because that's the whole point of the sacrifice. You have a problem. I have a solution: sacrifice. And some malarkey to explain why my sacrifice will solve the problem.
Sacrificers tend to prevent others from taking effective actions. Why can't sacrifices to the gods and effective actions peacefully coexist? Because that would prove my malarkey was false, the sacrifices I called for were in fact harmful and unnecessary and self-serving, and I will lose tremendous status.
Why does sacrifice and malarkey so often win out, when we have such a vested interest in effective action? Because malarkey can optimize for simplicity, and give the consolation of in-group status for people willing to participate. The truth can only optimize for the truth, which may be complicated. Sometimes, the necessary work and attendant rewards only achievable by elites or an unpopular subgroup.
Why does sacrifice have to be costly? This operates under the theory that we use sacrifice as a heuristic to tell who's doing the most to help out our side - or as a self-signaling mechanism to remind ourselves about how loyal we are to our own team.
This model predicts that in any given complex crisis, simple, stupid, symbolic sacrifices will be heavily represented in speech and action. While this is an uncharitable explanation for behavior we disagree with, it's still kinder than the alternative (conflict theory) and more epistemically responsible than questioning our own judgment just because some alternative authority sez so. When a behavior appears to be easily explained as a stupid symbolic sacrifice, our prior in a complex novel circumstance should be that that's exactly what it is.
Not sure how this applies to face masks -- they are not literally costly, but the fact that you wear them on your face should make them a great tribal signal regardless.
What about not wearing them? It’s hazardous to your health and that of others. We can clearly see that not wearing masks at large social gatherings is functioning as a universally acknowledged tribal signal. It’s not the masks, but the lack of masks, that is a sacrifice to the gods.
Although what’s interesting is that when one side sacrifices to the gods, the other side gets a “free” signal. If one side rejects masks, the other side gets to adopt masks - which are already good for them - as a symbol of loyalty, which just makes the mask wearing even better.
Does this mean that the mask-wearing side needs to find alternative sacrifices? Perhaps by refusing to see friends even when they would be able to safely? Or does it mean they get a freebie?
I appreciate these continued reports. They save me from having to choose between wasting time on noise and not keeping tabs on things at all.
Same! Zvi you're my only source of information on COVID and it's wonderful to not have to follow the news.
It's not a good idea to rely too heavily on a single source, even a good one. Every source will have some errors, and ideally you want to follow multiple sources whose errors are relatively uncorrelated.
Upvoted for the important consideration, but your own brain is a source of errors for which it's hard to decorrelate, so is it really worse (or worse enough to justify the additional costs of the alternative) to just trust Zvi instead of your own judgement/integration of diverse sources?
ETA: Oh, I do read the comments here so that helps to catch Zvi's errors, if any.
Note that my own blog also has an active comments section that has some good people in it (and some not so good of course), if you want to watch for errors without checking other sources directly.
I do think that if there's something that impacts your decisions a lot, you should do your own investigations too!
There are less cynical ways to view countermeasures that go too far. I'd compare it, especially early on, to many of us developing mild OCD because of how terrifying things were - compliance was also very high early on.
A lot of the response in Europe/UK has not looked like this, or like your opposite side but it still hasn't been very good.
The UK/Europe response been more like an inefficient, clumsy attempt to strike a 'balance' between mitigation and saving the economy, while showing no understanding of how to make good tradeoffs - e.g opening the universities while banning small gatherings. It looks more like an attempt to do all the 'good' things at once for the economy and health and get the reputational/mood affiliation benefits from both. E.g. in the UK in summer we half-funded the tracing and isolation infrastructure, ignored that compliance was low and gave subsides to people eating out at pubs and restaurants after suppressing the virus hard and at great cost, and now might be employing incredibly costly lockdown measures again when we could have fully squashed with a bit of extra effort in the summer when numbers were almost zero - and that's the same story as most of Europe.
That's more a failure to understand/respond to opportunity costs than either of the failures you describe, though it has aspects of both. It doesn't look like they were acting with the goal of getting people to adhere to the costliest measures possible, though - witness the reluctance to reimpose restrictions now.
This is the part that I like to emphasise, and the reason that we're still bound for a better outcome than most March predictions implied is because of a decent level of public awareness of risk imposing a brake on the very worst outcomes - the Morituri Nolumus Mori. Many of us didn't properly anticipate how much physical reality would end up hemming in our actions, as I explained in that post.
This is also worth emphasising. In general, though not in the examples you mention from e.g. California, going too hard works better than going too soft because there just is no pure 'let it rip' option - there's a choice between coordinated and uncoordinated suppression. It looks like voluntary behaviour has (in Europe and the US) mattered relatively more than expected. Countries that relied on voluntary behaviour change like Sweden didn't have the feared uncontrolled spread but also didn't do that well - they ended up with a policy of effective ‘voluntary suppression’ with a slightly different tradeoff – economic damage slightly less than others, activity reduction slower and more chaotic, more deaths. This was essentially a collective choice by the Swedish people despite their government.
The initial estimates said that 40-50% infected is a reasonable lower bound for when weak mitigation plus partial herd immunity would end the pandemic naturally. I think that's still true. So, it would all have been 'worth it', in pure death terms, if significantly fewer than that many people end up catching coronavirus before much better treatments or vaccines end the epidemic by other means. Last time I checked that's still likely.
These are mostly deaths due to fentanyl. When fentanyl displaces heroin in a region, it usually causes this kind of spike in deaths. (I don't know if there's an uptick in fentanyl in SF over the last few years, but such an uptick has happened in various places in the US). SF already had serious drug/homelessness problems. Why think this has anything to do with the specifics of SF's Covid response?
It also seems odd to criticize SF. Their Covid track record looks superb compared to major US or cities of Western Europe (save for Germany). Lots of businesses will be forced to close, but that's also true in places that have had more permissive rules.
I think it's worth noting that the woman that died from a confirmed reinfection was 89 years old and suffering from a rare from of Leukemia.
For future reference, there is a website tracking COVID-19 reinfections. We are up to 24 confirmed reinfections.
By the way, your series of posts on the subject inspired me to write my own, focused more on those risks that those of us who recovered from COVID-19 face.
Thank you for the inspiration.
Is the CFR going down because the virus is less deadly in general, or because the virus is infecting younger people? I haven't been able to find a decent study on this.
In the UK(https://www.gov.uk/government/publications/demographic-data-for-coronavirus-testing-england-28-may-to-26-august/demographic-data-for-coronavirus-covid-19-testing-england-28-may-to-26-august#age-and-gender) the median age of people testing positive has gone down from 52 to 31:
Annex table 7: median age of people newly tested and newly testing positive for COVID-19 under pillars 1 and 2, England
I think death rates are still down despite younger demographics, but how much? Is it still a 10%+ risk for a 60+ person with 1 comorbidity? What about if ICU's are overflowing and people end up in temporary care units? Romania, where I'm from, has a very poor healthcare system and seems to be heading into an ICU overflow situation sadly, but not sure how concerned I should be for my parents and other older relatives(they're all taking it quite seriously of course).
(epistemic status: assuming good faith)
... “test, trace, treat and isolate”?
Except it's not ethical to test at any practical cost, privacy rules mean you can't trace, we all know the rules about treat, and we already know what we think about isolate!
As someone who has not been following this threads and only read the last two, can I throw in some criticism? Do stop me if I am violating norms here.
The style seems to assume that the reader has read many previous threads and agree with you on every aspect of covid. I'm not sure I disagree with you on anything, but I would like some motivations. Combining this with a lot of sarcasm actually makes it hard to understand what positions you are arguing for sometimes.
On the topic itself, I am hopeful that the public discussion will get less partisan after the election. Unfortunately, a virus can spread far in three weeks.
Just go back and skim a couple of them. You wouldn't start a book in the middle and then criticize the author for being hard to follow