“America is used to blaming individuals for systemic problems. Let’s try to avoid that this time.”

I like this news article by Aviva Shen:

In normal times, policing has been America’s primary response to a host of societal ills that cannot be solved by punishment. Homelessness, mental illness, violence, racism, poverty, and toxic masculinity are all fed through the criminal justice system, rather than getting addressed in any meaningful way, never mind resolved. Now we expect police officers to contain this virus for us, even as they too become infected. Even though trying to arrest our way out of this just puts everyone in more danger. . . .

The problem with the way the government has failed to contain this crisis is that now, stopping the coronavirus has shifted to be a matter of personal responsibility. . . .

When the virus first entered the U.S., the people who could have done something to contain it did nothing. They didn’t even bother to track it. Meanwhile, South Korea and Taiwan immediately instituted widespread testing, created “quarantine hotels,” and deployed their robust social safety nets to ensure everyone could have basic needs met while staying indoors. Those countries have seen far lower infection rates than the U.S. . . .

Letting fear guide our responses won’t lead to measures that make anyone safer. And when a policy response is designed around the worst of the worst—the exceptions—we get nonsensical and brutal systems. . . .

This reminds me of some things we’ve been talking about lately about the tension between individual and group decision making.

To return to Shen’s article, another thing we’ve been thinking about a lot is how wrong many of us have been in our expectations of the coronavirus epidemic. I read this comprehensive news article from a few days ago by Yasmeen Abutaleb, Josh Dawsey, Ellen Nakashima and Greg Miller about how unprepared the country has been: “From the Oval Office to the CDC, political and institutional failures cascaded through the system and opportunities to mitigate the pandemic were lost”:

It did not have to happen this way. Though not perfectly prepared, the United States had more expertise, resources, plans and epidemiological experience than dozens of countries that ultimately fared far better in fending off the virus. . . . Warnings were sounded, including at the highest levels of government, but the president was deaf to them until the enemy had already struck.

The Trump administration received its first formal notification of the outbreak of the coronavirus in China on Jan. 3. Within days, U.S. spy agencies were signaling the seriousness of the threat to Trump by including a warning about the coronavirus — the first of many — in the President’s Daily Brief. And yet, it took 70 days from that initial notification for Trump to treat the coronavirus not as a distant threat or harmless flu strain well under control, but as a lethal force that had outflanked America’s defenses and was poised to kill tens of thousands of citizens. That more-than-two-month stretch now stands as critical time that was squandered.

But it was not just the president. As of January, even through late Februrary, I had no sense of the severity of the problem. Sure, I wasn’t as bad as this guy, but that’s a pretty low standard.

Why were I and others so wrong? I think because of a natural and unavoidable division of labor, which is that that most of the time we think about our own individual decisions, with the assumption that larger organizations will work out group and societal decisions. In January and February, there wasn’t much for any of us to do as individuals, and even the individual decisions we could make—for example, wearing masks or buying extra food for our pantry—were not going to have much direct effect on us and weren’t particularly urgent in any case. But, as we’ve learned, these decisions, in addition to societal decisions such as not to institute mass testing for the virus, were urgent and consequential at the national and global levels.
Nassim Taleb and Joseph Norman put it well:

We’ve been trained and coached and nudged:
(a) to think of larger social outcomes in terms of the aggregation of individual decisions, and
(b) when thinking about social decisions, to have a very narrow palette of options, pretty much limited to policing, wars, and throwing money at a problem.
Recall the newspaper headlines during those couple of weeks in March when it became clear that coronavirus would be a major problem in this country: lots of talk about the stock market, followed by lots of discussion of how many hundreds of billions of dollars would be thrown at the problem. I’m not saying that massive government spending is a bad idea here, just that it was bit of a simplification to focus on the How much? question, rather than on the What’s being done? question.

Again, my point here is not to criticize others. I’ve pretty much been going through the same sequence of thoughts as a lot of other people on this one. I’m a statistician, but it’s not like in January I was screaming that we need a billion test kits. There were some Cassandras back then, but I wasn’t one of them. The social science in this post of mine is largely introspective: What were the mental roadblocks getting in the way of clear thinking for me? I think part of it was that my implicitly individualistic perspective. Hence my appreciation for the above-linked article by Shen’s.

P.S. I made the mistake of looking at the comment thread on that Slate article. Oof! Seeing that sort of thing makes me realize how lucky we are to have the commenters we have here.

P.P.S. Speaking of Taleb, I just saw this great line:

Do not conflate doing science and being nice and polite to scientists.

122 thoughts on ““America is used to blaming individuals for systemic problems. Let’s try to avoid that this time.”

  1. “I made the mistake of looking at the comment thread on that Slate article. Oof! Seeing that sort of thing makes me realize how lucky we are to have the commenters we have here.”

    Looking at online comments is both horrifying and educational. Almost all of it is just random sludge. So what do we learn?

    1. The Enlightenment ideal of rational discourse is not the way the world actually works.
    2. A large part of public debate is just various tribes yelling insults at each other.
    3. Therefore, tribalism is bad, and more tribalism is badder.

    So why is this comment section better? My guess is the topics covered naturally restrict commenters to mathematically and scientifically literate people, so there is a culture of objectivity, a set of rules on how debate should proceed, and a set of tools (logic and math) capable of convincing other commenters.

    • The pessimistic interpretation is that this blog is sufficiently niche that it is frequented and commented on only by a single tribe. And/or the subject matter is one that does not potentiate “tribal” divisions among its readers.

      • Seth:

        We do get disagreements here, and people are not always polite. One thing about that Slate comment thread, and I’ve also noticed it on Marginal Revolution comment threads, is that the comments often start super-aggressive and then just go on from there. So, not even the hope of learning anything from the discussion (except for learning that certain commenters have these particular views).

  2. I had a similar reaction as the response seemed hugely disproportionate to the information being communicated at the time.

    It wasn’t until I merged some state level mortality rate data to the state level covidtracking.com data that it hit home for me. What I saw for NY was this:

    a. Estimated mortality Rate for 2018 per 100k Pop = 626.7. This puts the daily avg at 1.7/100k.
    b. Covid-19 New Deaths on April 1st were 391. Or ~ 2.0/100k

    I reasoned that if the mortality rate per 100k encoded in part the level of resources in existence to deal with a typical rate of deaths that the doubling had to be a huge strain on resources which could in turn affect outcomes. Recently, it has reached 3 times the rate and other states have started to make that climb. That said, I suspect that MSA is probably a better level of aggregation for that comparison rather than state level.

  3. Government tracking is not tenable due to the Constitution, which isn’t a problem in South Korea and Taiwan. Here’s the ACLU’s comment on the 4/10 Apple/Google joint venture on contact tracing.

    “We will remain vigilant moving forward to make sure any contact tracing app remains voluntary and decentralized, and used only for public health purposes and only for the duration of this pandemic.”

    https://www.aclu.org/press-releases/aclu-comment-applegoogle-covid-19-contact-tracing-effort

    • Effective contact tracing doesn’t require that everyone’s data be collected in a centralized database and merged with every other ‘flippin’ dataset scraped from the internet.

      I wouldn’t knowingly give my health information to Apple or Google and no one in the US should be forced to do that either.

      There’s a suitable alternative: deploy an app that can be used independently by county health departments, with the data’s use being restricted to COVID tracking, not transferable to an other entity, and destroyed in 2 years.

      There’s no need to sell out to Creepy Tech to manage public health.

  4. Contact tracing and privacy law. My summary, current privacy law is too doctrinaire to handle urgent public health crises like this.
    https://reason.com/2020/04/02/covid19-exposes-the-shallowness-of-our-privacy-theories/

    Also, ACLU’s statement on 4/10 Apple and Google joint venture on contact tracing.
    “We will remain vigilant moving forward to make sure any contact tracing app remains voluntary and decentralized, and used only for public health purposes and only for the duration of this pandemic.”
    https://www.aclu.org/press-releases/aclu-comment-applegoogle-covid-19-contact-tracing-effort

  5. Aviva Shen said:

    “Now we expect police officers to contain this virus for us, even as they too become infected. Even though trying to arrest our way out of this just puts everyone in more danger. . . .”

    Say what? Why should I believe anything this person says?

    • Terry:

      I think she’s saying it puts people in more danger because taking people to the police station or putting them in jail is itself an effective way of spreading the disease. And people cycle in and out of jail, spreading the disease back into the general population.

      • Who is this “we” who “expect police officers to contain this virus for us” so we can “arrest our way out of this”? Such a thought never entered my mind. How many public officials have called for “police officers to contain this virus for us” so we can “arrest our way out of this”? Out of 330 million people, how many have been jailed to “contain this virus”? Is it more than ten? Are arrests anything but a tiny, trivial part of the whole nation’s effort?

        The statement is just ludicrous on its face.

        • Terry:

          Apparently a lot of people are calling the cops on their neighbors. Nobody called the cops on me, but they did come by the park and tell me and some others to stop playing basketball. The cops weren’t mean about it, but it seemed like a waste of time. Had they put us in jail, I think that would’ve been a mistake from a public-health perspective, beyond everyone else.

          I agree that whatever the cops have been doing on this, it’s been minor, and I didn’t read Shen’s article as claiming otherwise. My take on it was that she was using the extreme cases (people calling the cops on their neighbors, cops putting people in jail for socializing) to make the larger point that it’s a mistake to think of the virus response as just a collection of individual actions.

          Also, the number of people being put in jail is not zero, and apparently the infection rates in jails are high. So, even if this represents a small fraction of the population, it’s worth writing about.

        • That’s pretty civilized. Out here ( India) the cops have just resorted to indiscriminate lashing out with batons at anyone spotted in a park or playground.

          I don’t even blame the cops.

          Often I think people ( in general) are not greatful enough about the previlage of living in USA.

        • +1

          Thanks, Rahul, yes, we are very lucky. I think we’ve done well here despite a lot of confusion because so many people not only cooperate but take their own initiative to do things that need to be done.

  6. You’re being way too hard on yourself, Andrew. Your reaction was too tempered because you didn’t have access to trusted, accurate, specialized knowledge.

    You could have been given that knowledge–not about what would ultimately transpire, but certainly about what you should prepare for and how. Instead, obstacles to knowledge were systematically placed in your way by the administration: Trump pushed propaganda and disinformation hard, while federal experts were reprimanded or sidelined for disagreeing. You couldn’t rely on establishment Republicans to raise honest concerns, or the press to give more weight to facts than talking points, because both have been conditioned by the never-ending sequence of leadership crises over the last three years to believe these are just political skirmishes with no real consequences. (Remember how worried everyone was about the North Korea tweets? Seems like eons ago…) But Trump and his allies have had long enough to overcome the institutional resilience that shielded us before, by defunding programs, replacing honest advisors, firing career officials, and corrupting or destroying faith in entire agencies.

    In other words, you were confused because several individuals created systematic problems.

    • Anon:

      It’s fine that you were alert earlier than I was. As noted above, I was not really paying attention. I could tell that Sunstein was being a buffoon when he was telling people not to worry, but I wasn’t in touch with the facts, myself. And, for all Sunstein’s flaws, the do-the-opposite-of-what-Cass-Sunstein-suggests strategy probably isn’t such a good plan either! But the main point of my post above is that individual stocking up and individual precautions, while probably helpful on net, is still no substitute for social action. And, when it comes to social action, there were lots of debates inside the government, as we’ve recently learned, for example from the news articles cited in the above post.

      • Well the funny thing is once the hysteria started we had enough info to see the virus was not that bad…

        From the washington post article:

        the urgent messages of public health experts.

        Who were the public health experts who recommended everyone start preparing in mid January? I didn’t see anything, so I guess it was all conveniently top secret. This is why I don’t pay attention to the news, they just quote and refer to anonymous sources and no one his held responsible when they are wrong.

        And when I know about a topic, they are almost always not even close to correct about it. So I am going to assume this didnt even happen. Or it happened in a way very different than what is implied, like a few people sent an email with “Urgent” in the header to someone but they also do that every year for the flu, etc.

        • Pretty sure that was an urgent but secret message to top brass, ie Trump, maybe some governors, etc. They usually withhold this stuff from the public in the early stages until a decision is made about how the response will go… Trump just didn’t bother responding at all, so public health never got the go ahead to say anything to the public. Remember when the white house announced everything had to be cleared through Pence?

        • I guess it depends on your baseline… Pretty sure Anoneuoid in the early days like Jan was looking at Wuhan and thinking this was going to be maybe a 10% case fatality rate across the board, spreads airborn in the wind, dead people in the streets, panic/riots sort of thing.

      • By the way I didn’t think you considered yourself a medical expert. And the useful info available at the time was not something that would be analyzed via statistics or epidemiology. It was just seeing what was going on in Wuhan via what leaked out onto youtube and seeing the western media minimize the problem.

        • Anon:

          No, I’m not a medical expert. That’s my point! There were experts who had a sense of what was going on, and then so many of the rest of us babbling away in ignorance. I find it an interesting data point that I was so clueless. The only thing separating me from Cass Sunstein was that I knew enough not to be overconfident. But not-being-overconfidence, that only takes us so far.

        • There were experts who had a sense of what was going on

          Who were these experts though? From what I saw they were all saying something along the lines of “there is no evidence this is a big deal”.

        • Very interesting to read these emails. The OCR is so bad I’m not going to copy/paste but search for “chatter”:
          https://int.nyt.com/data/documenthelper/6879-2020-covid-19-red-dawn-rising/66f590d5cd41e11bea0f/optimized/full.pdf

          It says the WHO and CDC seems to be downplaying this and hes is “certainly no public health expert” but he is very concerned. The NYT presented him as if he was a public health expert.

          They weren’t people who had a horn to the public though.

          This doesn’t exist anymore. Look at what Cameron Kyle-Sidell has managed to do.

        • Also, those emails were from Jan 28th. By mid-Jan I was buying solar power stuff and in late Jan ordered 5 kilos of vitamin C. By Feb 3rd 50 tons (4 grams/resident) of vitamin C were being shipped into Wuhan with the recommendation that every healthcare worker take 2 grams/day: https://mobile.twitter.com/DSM/status/1224262885729349633

          I’m sure if you go ask 10 public health experts today at least 8 will still tell you using vitamin C for covid19 is a myth.

        • by a horn to the public I mean one that won’t get them fired from their federal govt job for insubordination.

          And doctor had to step down from the ICU he ran because he refused to treat any more patients with early, aggressive ventilation. There are always consequences for not going with the groupthink.

        • I’m still not clear on the evidence on Vit C. You seem to be pretty convinced, but I haven’t seen any actual one-way-or-another on it. Hard to get data if no-one will take it seriously :-\

          My current thinking on it is that it probably doesn’t hurt, that it’s known to be a good thing to have lots of anti-oxidants if you are having high inflammation, but that I’m not convinced there’s as big an effect as you seem to be convinced of.

          That’s neither here nor there though. The key is, lots and lots of people KNEW this was coming in mid Jan, it took til Mar 20 for NY to give a stay at home order. That’s more or less 60 days, during which it got ~ 1 Million times worse in the US. If we had put that stay at home order in place Feb 20, country wide we’d have like 10k cases probably, and we’d have had 2 months to roll out dramatic testing and control measures like S. Korea.

        • I’m still not clear on the evidence on Vit C. You seem to be pretty convinced, but I haven’t seen any actual one-way-or-another on it. Hard to get data if no-one will take it seriously :-\

          It is deficient in every severe illness and especially viral illnesses. It is a travesty that not a single COVID-19 patient has had their serum ascorbate levels measured.

          But the key is to give it *before* severe illness because even though it would help with wound healing, etc via collagen synthesis the much bigger role is preventing oxidative damage in the first place.

          My current thinking on it is that it probably doesn’t hurt, that it’s known to be a good thing to have lots of anti-oxidants if you are having high inflammation, but that I’m not convinced there’s as big an effect as you seem to be convinced of.

          Yes, it is cheap and safe and the couple dozen incidents of possible serious side effects reported in the last 50 years can be basically rounded down to zero when compared to any other medical intervention.

          That’s neither here nor there though. The key is, lots and lots of people KNEW this was coming in mid Jan, it took til Mar 20 for NY to give a stay at home order. That’s more or less 60 days, during which it got ~ 1 Million times worse in the US. If we had put that stay at home order in place Feb 20, country wide we’d have like 10k cases probably, and we’d have had 2 months to roll out dramatic testing and control measures like S. Korea.

          Stay at home was not necessary, only closing the borders and properly screening/quarantining the people who came into the country. Look at what Taiwan did. Of course even when something like that was done too late in an inadequate fashion it was criticized by the media:

          At this point, sharply curtailing air travel to and from China is more of an emotional or political reaction, said Dr. Michael T. Osterholm, an epidemiologist and director of the Center for Infectious Disease Research and Policy at the University of Minnesota.

          https://www.nytimes.com/2020/01/31/business/china-travel-coronavirus.html

          Do not get your info from these people…

  7. Largely, this won’t be sorted out for a while.

    In Canada, largely I believe the experts were listened to but there were a number of blind sides of the experts.

    S**t happens.

    Right now, how to best persevere and then how to do better next time…

    • Yeah I kindof agree with this. I guess retrospectives are always good, but it seems like there’s still plenty to come medical-wise, public policy-wise, etc.

      > how to best persevere and then how to do better next time

      In the whole individual vs. system thing, the New York budget bothers me. Like it’s simultaneously saying there’s a middle class tax cut but also saying the state might need to cut $10 billion in spending: https://www.governor.ny.gov/news/governor-cuomo-announces-highlights-fy-2021-budget .

      At the same time, I got an emergency message (maybe a week or two ago) asking for volunteers (I think it was for this: https://www1.nyc.gov/site/helpnownyc/give-help/volunteer.page). Someone sent out an e-mail to Columbia postdocs looking for volunteers as well.

      So the state needs money to handle the situation (Cuomo has said we’re broke), but the state government has decided (of its own accord) to not try to raise money? And the solution is instead to ask for volunteers? It doesn’t add up.

  8. “policing has been America’s primary response to a host of societal ills that cannot be solved by punishment. Homelessness, mental illness, violence, racism, poverty, and toxic masculinity are all fed through the criminal justice system”

    I don’t think neither “Policing has been America’s primary respone to …” nor “cannot be solved by punishment” are true.

    • The article is such a hodge-podge of wild rhetoric that it is hard to see the logic. In such a case, it often helps to just assume the author wanted to arrive at a certain endpoint and threw together whatever seemed most expedient to that end.

      We know with high confidence that the author wanted to reach an Orange Man Bad conclusion. I speculate that the author thought the article’s contribution was to head off the competing claim that individual misconduct was partly to blame. It is there in the title “America is used to blaming individuals for systemic problems.”

      I further speculate that the resulting article was too boring and unconvincing, so the author wrapped the whole thing in a layer of Social Justice rhetoric, tossing into the pot whatever came to hand: “homelessness, mental illness, violence, racism, poverty, and toxic masculinity” along with random anecdotes about policing, “mass incarceration”, and what not. When you start cranking the Social Justice handle, any old mish mosh of claims works just fine. Everything is one big gooey ball of buzz words, and you can count on a Social Justice audience to instinctively thrill to the warm, virtuous soup of tropes.

      • Terry:

        I don’t see this at all! I think there are real issues of people not having a good set of tools for handling problems at a societal, rather than an individual level. And one example of this is throwing people in jail as a solution to coordination problems. We realize this in some settings: when people park their cars illegally, we don’t throw them in jail, we mail them a ticket and maybe tow their car. Homelessness, mental illness, violence, racism, poverty, toxic masculinity, mass incarceration: these are real things with real consequences. This is not to say you should agree with everything in Shen’s article, but I don’t think it’s just a “gooey ball of buzz words” or a “warm, virtuous soup of tropes.”

        Thinking about this from a social science perspective, I think there’s a real problem that we so often either focus on individual actions. And when we think of social actions, they’re so often on the crude level of, Pass a law, Call the cops, Send in the troops, Throw government money at the problem.

        • Andrew, when author confidently asserts the criminal justice system is our primary tool for dealing with mental illness and violence can’t be solved with punishment, it sure looks the kind of bloviating Terry described. It’s hard to take such an author seriously.

        • Um, jails and prisons have become our de facto tool for treating mental illness in the US. That’s a given that you can verify yourself just by googling it. Start with that hotbed of SJWs at the US Department of Justice.

          And violence isn’t solved by punishment as the effects of the violence still exist. If you think punishment deters future violence, ok, but there’s limited support for that because deterrence and deterrability are not the same thing. One has to be able to do the cost/benefit analysis in order be deterred and there’s great variation in deterrability.

        • Some people with mental illness wind up in jail, but the vast, vast, majority of people with mental illness do other things for it (look it up).

          Locking more people up for crimes has a dramatic effect on the amount of violent crime, which seems like a “solution” to anyone who’s a potential victim of violent crime.

          But at a more basic level, suggesting punishment doesn’t work on violence is a red-flag to expect pure sophistry in the rest of the authors article.

        • This isn’t so true if you restrict to severe mental illness. Let’s say schizophrenia, bipolar, and psychopathic personality disorder.

          There are lots of people with either major or minor depression, anxiety disorders, Narcissism etc who don’t wind up in jail. But most of the people who experience hallucinations or mania or have zero empathy for others and would kill someone’s grandmother to get a better place in line at the sandwich store eventually wind up in the criminal justice system, and it doesn’t do anything to treat the underlying cause nor is it a deterrent to future issues particularly.

          When someone “sees the aliens controlling your body” and “puts their hands around your neck, and squeezes to drive the aliens out and release you from their power” it doesn’t go over well with the police, but when the police come, it doesn’t change their hallucination either.

        • The author’s claim that we are trying to arrest our way out of this is ludicrous, and is is the author’s central claim.

          Your own story about being reprimanded for playing basketball contradicts this. You weren’t thrown in jail, you were gently admonished, exactly what you say you want when you say we should ticket illegally parked cars rather than jailing people.

          Further, saying we should blame the system and not individuals is a false dilemma. There are probably some people who should bear some blame. Probably not many, but some. To date, there has been little blaming of individuals. It might be worth a gentle caution for the future, but there is little evidence that will happen on any significant scale.

          The rest is extraneous verbiage. Dragging in other alleged injustices about a laundry list of other things doesn’t change the falsity of the central argument. The proposition that A is a moral outrage is not proven by showing that B, C, and D are moral outrages.

    • Thanks, I’d seen that US data so theres no reason to think the same thing wouldn’t show up in europe. I’ll have to look at that closer and see if it shows what he says though.

      There was also a study where they tested an entire nursing home in washington and found 7/53 (13.2%) negatives were smokers vs 1/7 (4.4%) positives. Taken alone that doesn’t mean much but it fits with the pattern.

      https://www.cdc.gov/mmwr/volumes/69/wr/mm6913e1.htm

    • From eyeballing it I get 105/705 (15%) of deaths were smokers with very few unknowns and 400/10k (4%) total cases were smokers with 8% unknowns. For general prevalence of smoking in France 20-30% seems to be typically cited.

      How long is the medical research community going to continue this farce? Because word is spreading anyway and people are going to start smoking if they don’t figure out why this is happening.

      Like read this. I guess without looking at the data they have already concluded we need stricter laws against smoker:

      https://blogs.bmj.com/tc/2020/03/24/covid-19-and-smoking-the-elephant-in-the-room/

  9. I might be one of those grousing commenters for that Slate article, although since the comments there are terribly designed it’s hard to say. If we’re going to have rules restricting what people can do, they’re going to be enforced by police and an individual person’s risk of infection in lockup has to be contrasted with the total amount of activity deterred by police. Even the claim that “Americans […] call the police too much” is undermined BY THE VERY ARTICLE IT LINKS TO, pointing out that many groups don’t call enough.

    As for why the comments there are so bad, I think one point of comparison might be youtube. People go there for the videos, not the comments, and most people comment without reading other comments. The authors of Slate articles generally don’t read their own comments, so people don’t write comments there with the expectation of providing useful feedback. Instead it’s more for firing off snark because something has irritated you.

  10. Interesting possibilities as this pre-print suggests in Spain, covid19 was circulating earlier than they knew being missed as just the flu –

    Excess cases of influenza suggest an earlier start to the coronavirus epidemic in Spain than official figures tell us: an analysis of primary care electronic medical records from over 6 million people from Catalonia ttps://www.medrxiv.org/content/10.1101/2020.04.09.20056259v1.full.pdf

    • This has also been a theory in Santa Clara California… but Stanford university went back and analyzed samples from student health centers and etc, and did NOT find the virus. It’s also the case that if it were circulating earlier, we should have seen an exponential growth, and corresponding deaths etc. I don’t think we have that pattern.

      https://news.cgtn.com/news/2020-04-13/Coronavirus-in-California-circulates-months-earlier-than-anyone-knew-PECgMGS2cw/index.html

      In the end I think this theory is debunked because it predicts things that didn’t happen (a major peak in elderly deaths much earlier than now and detectable virus in archived samples)

      I suppose it could be true in spain, but not in CA.

      • In the end I think this theory is debunked because it predicts things that didn’t happen (a major peak in elderly deaths much earlier than now and detectable virus in archived samples)

        I disagree. I’ve been guessing for awhile that the 80+% asymptomatic/mild cases may not have antibodies that are detectable for more than a few months. In NYC right now 13% of pregnant women are testing positive (for active infection) with almost 90% asymptomatic: https://www.nejm.org/doi/full/10.1056/NEJMc2009316

        • That just confirms what I’m saying.

          If COVID19 circulated in NY say starting Dec 20, then doubling every 3 days as it seems it does, by Mar 20 we’d have 1.4 billion people infected. Obviously there aren’t 1.4 billion people in NY, and logistic growth would have kicked in, but we’d have WAY WAY more than 13% of pregnant women positive, well over half the population of NY would have had it. Also, there’s no way that people who got it earlier were dramatically more likely to be asymptomatic, so that all the death and destruction just happened to kick in around April 1-15

          The model that makes sense is it started spreading uncontrolled around second week of Feb or something like that.

        • No, the testing increases for sure, but the doubling every 3 days during rampant spread is pretty much constant across multiple countries regardless of the testing rate, and in particular it’s visible in the severe cases as well (which is a small but one assumes relatively constant fraction of the identified cases), and it’s got an *exponential* shape whereas the number of tests that can be run asymptotes to a constant linear rate. The doubling every ~ 3 days represents the spread.

        • the doubling every 3 days during rampant spread is pretty much constant across multiple countries regardless of the testing rate

          I’ve been plotting the number of cases vs tests for awhile and it has been very close to constant (given the changes in testing criteria). The deaths are all also closely tracking testing rate. Do you have data showing poor correlation between testing rate and doubling time?

        • What’s needed is to look at the *short time* asymptotic behavior. I’ve got a second comment that has some links, it’ll take a little time to get through the spam filter. It shows the cases and the deaths through time for all kinds of countries.

          Asymptotically for *large* time (say 40-infinity days) the cases can’t grow faster than the testing, so plotting a constant slope line on a logarithmic curve will show a bending logarithmic kind of curve… but for *short* times, say from the time the 10th case is detected to 20 or 30 days later, the testing can outpace the growth and you expect to see a relatively straight line on the log/linear plot, and the slope of the curve in that region gives you a decent estimate of the spread doubling time, aggregating that slope across multiple countries gives a decent estimate, and it’s a doubling between 2 and 4 days or so.

          So, the phenomenon you’re looking at is later in the dynamics, and yes it would fool people. But early in the dynamics the curve is quite linear like and works well as an estimate.

        • Here’s a wide range of countries highlighted on the ourworldindata.org graph for cases.

          https://ourworldindata.org/grapher/covid-confirmed-cases-since-100th-case?country=AUT+AUS+DEU+CHE+USA+GBR+BRA+LVA+NLD+BEL+ISL

          Look at the dynamics in the first 15 days since the 100th case (x axis between 0 and 15). They all cluster around the “doubling every 3 days” line… This is in the range of 100 to 5000 cases or so. During that period of time in these countries were running vastly more tests per day than the positive cases found, and had yet to put in place any social distancing. This means the growth is a good estimate of uncontrolled spread

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=AUT

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=DEU

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=FRA

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=ESP

          Iceland is a good example:

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=ISL

          Their tests ramped up VERY quickly between Mar11 and 21 but the confirmed cases didn’t do the same thing.

          https://ourworldindata.org/grapher/covid-tests-cases-deaths?country=BRA

        • https://ourworldindata.org/grapher/covid-confirmed-cases-since-100th-case

          You can see the same approximate dynamics for Spain, France, Italy, Germany, US, Brazil, Netherlands, Austria…

          The same patterns occur in deaths for multiple countries.

          https://ourworldindata.org/grapher/covid-confirmed-deaths-since-5th-death

          They can’t all just be increasing their testing exponentially at the same rate… That’s not a model of testing that makes sense, whereas it IS a model of viral spread that makes sense.

        • They can’t all just be increasing their testing exponentially at the same rate…

          Thats an assumption. There was no reason the number of tests in the US needed to rise in a curve that mimicked an epidemic curve during March, but it did.

        • It’s actually not an assumption, you can see the cumulative testing curves in some new comments above, there are a wide range of shapes, some of which include some pretty rapid “step” like ramp-ups, or various temporary plateaus and soforth which don’t show up in the positive cases. So, it really does spread about doubling every 2-4 days in uncontrolled first-world city environments.

        • there are a wide range of shapes, some of which include some pretty rapid “step” like ramp-ups, or various temporary plateaus and soforth which don’t show up in the positive cases.

          For whatever reason getting the number of negative results is harder than the positive for some areas so they are not exactly comparing apples to apples:

          Other states provide some or none of these numbers on an ongoing basis. Some crucial states in this outbreak, notably California, Washington, and New York, have not been regularly reporting their total number of people tested. For these, we have to use other reporting tools: directly asking state officials, watching news conferences, gleaning information from trusted news sources, and whatever else it takes to present reliable numbers.

          https://covidtracking.com/about-data

          There is no reason to think this data is anywhere close to perfect.

        • that’s a point in time, today, across multiple countries.

          What I’m talking about is d(log(cases(t)))/dt for small t, where t is measured in days since the Nth case for some small N, like 100.

          The behavior of the curve of log(cases(t)) for large t is different for many reasons.

        • Here’s something you can do, take 6-10 first-world developed countries that didn’t slam on the brakes right away (so not south korea, taiwan, singapore etc), other than that, whichever ones you like.

          Find the point in time where they had 100 official cases… take the next 20 days of official case data…

          using the natural logarithm, plot log(cases) vs t = days since 100 cases

          now fit lines to the log(cases) vs t and get all the slopes.

          Now calculate the number log(2)/slope[i] and plot the histogram of those numbers.

          I haven’t done it, but I guarantee it will be a cluster of values around 3 or so.

          This is basically what I’m talking about as the estimate of the unmitigated spread rate.

        • that’s a point in time, today, across multiple countries

          Actually drag on the blue thing below it and it will show each country over time on the same chart. I like this idea.

          I haven’t done it, but I guarantee it will be a cluster of values around 3 or so.

          And what do you expect if you plug the testing data in there instead?

          Also, Iceland’s tested a lot of people without symptoms. That is a good example of why some countries will be outliers: https://www.nejm.org/doi/full/10.1056/NEJMoa2006100?query=featured_coronavirus

        • If i plug the testing data (total tests conducted) in i expect not a very good fit to a line, and whatever the slopes are they will vary much more across countries. Please keep the t values the same, based on the 100TH confirmed case of you do this. If you have the data and some time I’d be interested to see the results, ourworldindata let’s you grab their data right from a link on the graphs…

        • Daniel,

          Not a nation but I plugged in the cumulative cases numbers for my state. The first day with >100 cumulative cases was March 20. So I regressed log(cases) on day number start with March 21 being Day 1 and going through April 9 being Day 20.

          LN(2)/slope was 4.7, does that mean my state accumulated cases more slowly than most nations by your reckoning?

        • Brent, you don’t say which state, and I don’t know what kind of population densities you have (it should be places with big cities to see this kind of growth) but by Mar 20 everyone knew what was up, CA had put in their stay at home order, WA had had it in place for a while… WHO had declared it a pandemic Mar 11, people were reading the news and already doing their own social distancing. So, yes I do believe your region had slower growth, because it probably wasn’t fully uncontrolled growth.

          That being said, 4.7 is within the edge of the range I’m talking about. Something like 2 to 4 for most places, tailing off into the range of say 6-7 for lower density regions. Technically if you put in good lockdowns the doubling time goes to infinity, to see what’s up in heavy mitigation you want to plot a histogram of 1/time instead of time

  11. As usual, Andrew is brave enough to say the unfashionable. Holed up at home, I’ve been pulling my hair out hearing all the hindsight claiming it was perfectly clear from the start this was going to be an unprecendented pandemic. I think there are several dimensions worth discussing:
    a) it’s hard to unsee the numbers that came out since January but in January, without the data we now know, we had very little data to make a judgment. (see also point d below)
    b) like all statistical analyses, it’s always easier to retrace your path after you got to the end. I wrote about this here – anticipating the holier-than-thou reaction we are now witnessing
    c) I’m surprised there isn’t more discussion of heterogeneous decision agents. There appears to be a presumption of a grand scientific consensus on risk when in fact every individual has a risk tolerance parameter. This is well accepted in financial investing for example. It applies to health risk as well. The variance in response is due not just to the uncertainty in the risk estimate but also variance in risk tolerance.
    d) There is a law that needs a name: in a group decision with uncertainty, there is always at least one hero if the decision were to be proven wrong. This is because in a group decision, it is guaranteed that at least one person raised objection to the group decision (and subsequently went along with it). Thus, the ex-post correctness of the decision is not by itself evidence that the whistleblower/hero was smarter than the group. The existence of a hero just reiterates that the facts used to make the original decision were not all consistent i.e. there was uncertainty. On the other hand, if the group decision were correct, the media do not hyperventilate over the false alarm raised by the would-have-been hero.
    e) The whole situation relating to the cases and fatalities curves is a real-life lesson that data compose of not just the numbers but also the meta-data like the definitions, the context, the missing values, etc. The media’s coverage of these curves as a horse race is a tragedy. I’m sure the excuse is that’s the best data we’ve got. The media is not ready for the argument that a statistically adjusted dataset may paint a more realistic picture than the raw counts.
    f) Just in case people are not aware, the press finally realized that even the count of fatalities is wildly inaccurate. The British looked at all-cause mortality for the last few weeks, there are many excess deaths, and the gap is not explained fully by Covid-19. (D. Spiegelhalter had a tweet about this yesterday.) The Covid-19 in this count includes doctor opinions in the cases that were not hospital certified. This also confirms that in many countries including UK and I think also the US, the death counts have excluded people dying in care homes who so far are not eligible for testing.
    g) As always, testing is the elephant in the room. It’s not just the number of tests, it’s also who is being tested. And also, different decision units have different testing policies, and these policies do change over time, may even flip flop, so analyzing the testing data is very tricky business. This is before you had to add politics to the context, since fewer tests mean fewer cases and fatalities linked to Covid-19, a different “flattening” of the curve for PR. (This is most suspicious in the case of Japan wrt the Olympics.)

    • > in January, without the data we now know, we had very little data to make a judgment

      This was published (barely but) in January in a major jornal: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30260-9/fulltext

      Findings

      In our baseline scenario, we estimated that the basic reproductive number for 2019-nCoV was 2·68 (95% CrI 2·47–2·86) and that 75 815 individuals (95% CrI 37 304–130 330) have been infected in Wuhan as of Jan 25, 2020. The epidemic doubling time was 6·4 days (95% CrI 5·8–7·1). We estimated that in the baseline scenario, Chongqing, Beijing, Shanghai, Guangzhou, and Shenzhen had imported 461 (95% CrI 227–805), 113 (57–193), 98 (49–168), 111 (56–191), and 80 (40–139) infections from Wuhan, respectively. If the transmissibility of 2019-nCoV were similar everywhere domestically and over time, we inferred that epidemics are already growing exponentially in multiple major cities of China with a lag time behind the Wuhan outbreak of about 1–2 weeks.

      Interpretation

      Given that 2019-nCoV is no longer contained within Wuhan, other major Chinese cities are probably sustaining localised outbreaks. Large cities overseas with close transport links to China could also become outbreak epicentres, unless substantial public health interventions at both the population and personal levels are implemented immediately. Independent self-sustaining outbreaks in major cities globally could become inevitable because of substantial exportation of presymptomatic cases and in the absence of large-scale public health interventions. Preparedness plans and mitigation interventions should be readied for quick deployment globally.

      • Carlos: First, thanks for bringing the science here. Second, what you’re proving is that it is possible to interpret the existing data (from late January) to advocate for the what-is-now-obvious answer. Even in these researchers’ description, they used “probably”, “could become inevitable”, etc. This interpretation does not preclude other plausible scenarios, until more data came in.

        • I agree that making predictions is difficult, especially about the future.

          But “we had very little data to make a judgment” is not the same as “things are not certain until they have already happened (and even then we could argue if the data is complete and correct)”.

          At what point did you (or will you) have enough data to make a judgement?

          The judgment made by the authors of that paper, based on the data they had in January, was clear “Preparedness plans and mitigation interventions should be readied for quick deployment globally.”

    • Kaiser, I don’t think your characterization is accurate. Public health officials knew what was up by early Feb. Politics controlled this decision, not knowledge.

      My wife just received an email from some of her summer-course co-organizers. In mid Feb, I was encouraging her to start planning for coronavirus to cancel the summer course they organize.

      In *early* Jan the only people who would have known what was going on were in China, but you could guess that things might be not so good. By late Jan people in public health had a pretty good idea that things were headed south:

      Here’s the NYT data dump of email records from some whistleblower:

      https://int.nyt.com/data/documenthelper/6879-2020-covid-19-red-dawn-rising/66f590d5cd41e11bea0f/optimized/full.pdf

      • Public health officials knew what was up by early Feb. Politics controlled this decision, not knowledge.

        Your link is an email chain starting Jan 28th from a guy who explicitly says he is “certainly no public health expert” and “you guys made fun of me for screaming to close the schools”. So I think it shows the opposite of what you are saying.

        Ironically his solution wouldn’t have done much since the kids in school rarely get infected anyway.

      • DL: You just proved my point though. I agree politicians went above the consensus public health opinion. But a statement like “public health officials knew…” suggests 100% of all public health officials agree, which can’t be true. Later, you use “people in public health had a pretty good idea…” which is more to my liking. It expresses the low level of confidence they had in the projection. As more data show up, the confidence increases. It is only after we saw the script that we now over-estimate the confidence of the earlier analysis.

        I find it hard to believe that anyone can predict a rare event with almost 100% confidence at the front end of the crisis – on Jan 31, there were 6 confirmed cases in the U.S. Think about the counterfactual. What if this pandemic ended up more like SARS, which was well contained within 6 months without a need to shut everything down?

        I am not trying to justify what happened. How to avoid hindsight bias is a concern of mine for a long time when it comes to analyzing data. LIke I wrote in that linked blog post, at the end of any data analysis, I wished I could have skipped all the unnecessary dead ends and wasted time; why didn’t I just do the right thing from day 1? That’s the analogy I’m making.

        • Sure, we can’t predict what would happen, but that doesn’t mean we can’t uniformly make essentially the same decision between two options.

          The decision theory is choose the option that has the least expected cost (or most expected benefit). We can all agree that the right tail of how bad this would become was not well understood… But anyone who put essentially zero plausibility on this becoming a pandemic with a CFR at least 1% was a fool.

          Given the rapid climb in cost as number of cases and CFR climb, almost regardless of whatever you would have put in for your estimate of how bad the pandemic would be — led to the same decision: close everything as early as possible, wait for the initial peak to die off, keeping it as small as possible, and in the mean time buy equipment and supplies and start planning for mechanisms of responding in the long term.

          You can argue about “deaths could be this high” or “deaths could be that high” but there was no plausible argument for anything other than “act quickly to mitigate”.

    • “Holed up at home, I’ve been pulling my hair out hearing all the hindsight claiming it was perfectly clear from the start this was going to be an unprecendented pandemic. ”

      I agree. There were warnings but it wasn’t patently obvious. I think a person could argue, though, that it was significant enough that public health officials in the US could have moved much quicker on laying the groundwork in case it turned out to be serious.

      But I see a larger problem here:

      The response to this situation illustrates the increasing smugness and dogmatism of the “expert elite” in the US. We’ve lost our practical edge. Arguments from authority and doctrine are rising and arguments from careful consideration of all available information are falling. The discussion over mask use illustrates this perfectly. We don’t need RCTs to make this decision. On the other hand, of course, outside of this emergency situation, RCTs are still critical for most things in medicine.

      At the same time, it also illustrates the value of expertise. The mistrust of expertise by leadership was very destructive. While it wasn’t patently obvious that this would become a pandemic, it was clear that it could become one and *most* experts understood that and advised accordingly, but political leadership didn’t take any action.

      • Agree with much of what you said.
        I wouldn’t characterize the dogmatism as “smugness” though. There is an over-confidence in model outputs. Many epi models are highly structural, and it’s not always clear whether the outputs are conditioned mostly by the data or by the structural assumptions. It is a form of our professional disease.

  12. If i plug the testing data (total tests conducted) in i expect not a very good fit to a line, and whatever the slopes are they will vary much more across countries. Please keep the t values the same, based on the 100TH confirmed case of you do this. If you have the data and some time I’d be interested to see the results, ourworldindata let’s you grab their data right from a link on the graphs…

    Here you go.

    1) Downloaded the data from here: https://ourworldindata.org/grapher/covid-19-total-confirmed-cases-vs-total-tests-conducted?time=..
    2) Subset to keep the countries you linked above,
    3) Kept the ones with at least 4 testing datapoints and did linear interpolation on the test data

    Plots: https://i.ibb.co/g4FDymH/danielcovid.png
    Final data: https://pastebin.com/0QMGdjHh
    Code: https://pastebin.com/V1HNyCwR

    I didn’t follow the purpose of the log(2), etc by the way but you can see the results of running that on cases vs tests are very similar.

    • The log(2) is to put it on a “doubling” basis rather using “e” as the base of the log transform. Or at least that’s what I assume it’s meant to do.

      A somewhat odd thing has happened with testing in my state. The ratio of of “positive tests” to “all tests” started at about 10.8% for the first couple days the health department posted the stats then quickly settled down to a very narrow range of 10.5-10.6% of tests being positive after each day’s updates. Now after nearly a month the overall positive test percentage is pretty much locked down at 10.5%.

      I guess if they haven’t (much) changed the criterion for getting tested and the nature of the disease hasn’t changed it stands to reason the proportion of positive tests will be a constant.

    • Hey thanks. yes 2^(t/n) = exp(log(2)*t/n) so the slope you’re calculating is more or less log(2)/n so if you do log(2)/slope you’d get n which is the doubling time in days.

      I’d say that my intuition was confirmed, the histogram of the cases is that 70% were between 2 and 5. Whereas for the tests it was substantially more variation, with values ranging from 2 to 16 and 70% between 3 and 8.

      So the test quantity doesn’t fit a tight range of constant exponential growth rates through time across countries, but the positive cases do fit that narrow range, with the modal value for growth rate in cases at doubling every 3 days.

      • Whereas for the tests it was substantially more variation, with values ranging from 2 to 16 and 70% between 3 and 8.

        I wouldn’t consider this substantial. This is very messy data. I would consider it the same for all practical purposes.

        • The cases data is substantially more peaked about every 3 days than the testing data. Instead of a histogram try plotting the kernel density estimates.

          you can try plotting the doubling time of the cases vs doubling time of the testing, see how strongly correlated they are, of course there should be some correlation, in fact if there are a lot of cases growing, there will be pressure to test more, so the causality can go both ways.

          I just think when the cases concentrate around doubling every 3 days regardless of the testing being spread over a much wider range… it’s an indicator that there’s an underlying process which has an inherent timescale.

        • I just don’t think you are appreciating how disparate the different data collection methods can be from country to country. And we have also seen much more problem with collecting negative results than positive results in the US, since positives are seen as more important.

        • The disparateness of it all just strengthens my point, if something consistently grows exponentially at doubling every ~ 3+-1 days even though you measure it in all kinds of different ways… it suggests that there’s a real underlying phenomenon.

        • And you can say the same for the testing. There is some mechanism causing it to be rolled out at about the same exponentially increasing rate everywhere.

        • Probably because a lot of testing is prompted by symptomatic patients arriving at hospitals, and by retesting of recovered patients. There’s a causal relationship from cases to tests under a lot of regimens.

        • Exactly what Zhou said. Most places to get a test you have to be a person with symptoms or a known contact, so as the spread grows so does the testing, particularly during the early phases. later testing ramp up may be caused by desire to screen people en mass or to help reopen certain businesses or whatever.

          I stand by this estimation method of the doubling time as about as good as the data warrants. doubling every 3 days in city regions without mitigation.

  13. We have to be clear what the counterfactual is. It isn’t to shut down the economy earlier. The counterfactual was to conduct aggressive testing, with contact tracing and isolation, to contain the spread of the virus. When there were only a handful of cases in January, it would never have been possible to win the lockdown argument against business interests. But I don’t think public health experts advocated broad-based lockdowns from the get go. There appears to be numerous reasons why the U.S. failed to test, not all are political. Even now. That 15-minute test was announced weeks ago; what happened to it?

    btw, there are public health experts (heard on Asian news) who now believe that the testing, contact tracing and isolation strategy that worked for SARS – even if properly implemented – would still be less effective for this novel coronavirus because there is so much transmission by people with no symptoms.

    • All direct flights between Taiwan and Wuhan City in central China have been canceled, starting Thursday, as Chinese officials have imposed a lockdown on the city as part of their efforts to prevent the spread of a deadly new coronavirus, according to Taiwan’s transportation ministry.
      […]
      On Wednesday, the emergency committee of the World Health Organization (WHO) met in Geneva to discuss the outbreak but did not declare the virus a “public health emergency of international concern,” which would have required a coordinated global response.

      https://focustaiwan.tw/society/202001230007

      The outbreak is categorized as level 3, she said, adding that the government has established a “central epidemic command center” and a reporting system.

      Government agencies have also formed response teams, Tsai said, adding that epidemic prevention measures at the nation’s airports are “very strict.”

      Twenty-six people have been tested locally for the 2019-nCoV infection, with one case having been confirmed, she said.

      The patient is a woman who arrived at Taiwan Taoyuan International Airport on a direct flight from Wuhan, Tsai said.

      The woman has since been treated and no longer has a fever, she said.

      Other passengers and crew members on the same flight would be monitored for 14 days, Tsai said, adding that those with signs of infection would be quarantined, diagnosed and treated immediately.

      http://www.taipeitimes.com/News/front/archives/2020/01/23/2003729719

      CDC officials said the United States will be more strict about health screenings of airplane passengers arriving from Wuhan.
      The patient, who is not being named, is in isolation at Providence Regional Medical Center in Everett, Washington. He is in his 30s and lives in Snohomish County, Washington, just north of Seattle. He had recently returned from Wuhan.

      He arrived at Seattle-Tacoma International Airport on January 15, before any health screenings for the Wuhan coronavirus began at US airports. He sought medical care on January 19. The CDC and Washington state are now tracing the people he was in contact with to see if he might have spread the disease to someone else.
      “We believe the risk to the public is low,” said John Wiesman, secretary of health for the state of Washington.

      […]

      The CDC raised its travel notice for Wuhan, China, from level 1 to level 2 of three possible levels, according to its website. As of Tuesday afternoon, the agency advised travelers to “practice enhanced precautions.” The highest level, “Warning – Level 3,” advises travelers to “avoid nonessential travel.”

      […]

      “This isn’t anywhere near in the same category as measles or the flu,” Dr. Martin Cetron, director of CDC’s division of global migration and quarantine, told CNN.

      https://www.cnn.com/2020/01/21/health/wuhan-coronavirus-first-us-case-cdc-bn/index.html

      So we see the difference from Taiwan was not contact tracing, it was the public health experts downplaying the risk and weak travel restrictions.

      And when stronger travel restrictions were attempted, the public health experts were calling it “emotional or political”:

      At this point, sharply curtailing air travel to and from China is more of an emotional or political reaction, said Dr. Michael T. Osterholm, an epidemiologist and director of the Center for Infectious Disease Research and Policy at the University of Minnesota.

      https://www.nytimes.com/2020/01/31/business/china-travel-coronavirus.html

      These public health experts are worthless…

    • Lockdown in Jan, no, but before the end of Feb or the first week of March, yes. I advocated to our local school board to close schools as early as Feb 26 or so, I pulled my kids from school and advocated for closing them starting Mar 4. They actually closed Mar 13.

      We can estimate that closing everything 10 days earlier in california would have saved an additional 500 people or something, but doing it in New York would have saved about 9000 and a huge amount of medical resources. Remember, the alternative was NEVER “don’t close everything” the alternatives were always *on what day do we close everything*?

      the “don’t do anything” option results in 100M to a billion people dead worldwide, so it was always just when.

      Every 10 days of delay in metropolitan areas multiplied the death toll by 10.

      • Those are made-up numbers, Daniel. There has never been a virus in recorded history with a world-wide impact comparable to a billion people dead.

        I thought this blog was all about respecting the limits of what we know, what we don’t know, what is not knowable with the available data.

        • Under Wikipedia for the Black Death, it was estimated to have killed between 30 and 60% of the population of europe. These days there’s no isolation between different regions, so what’s 1/2 of 7 billion people?

          It’s not unprecedented. More to the point this kind of massive die off has been seen in OTHER animals than humans MULTIPLE times.

          Flu in 1918 killed on the order of 100M people at a time when the global population was ~ 1B, so what’s 10% of 7B? There were small countries that lost *** 25% *** of their population to 1918 flu.

          none of this is unprecedented.

        • Mass die offs are actually pretty darn common.

          https://www.wired.com/story/the-macabre-science-of-animal-mass-die-offs/

          https://en.wikipedia.org/wiki/Mass_mortality_event

          The thing is that you would never get 1B dead, because people would never just go about their daily lives as if everything were normal even though people around them were dropping like flies…

          That doesn’t mean it isn’t the logical consequence of “business as usual”. It just emphasizes that business as usual was never on the table. Mass closures were inevitable. If not by govt, then by individuals refusing to leave their houses no matter how much coaxing.

      • We can’t even agree on the alternative :)
        South Korea and Taiwan and Singapore and Hong Kong, at least in the first wave, did not have to close down everything. This is credited to their early testing, contact tracing, finding clusters, and isolating the positives. I’ve been following HK pretty closely. Even now, with the threat of the second wave, they have closed selectively certain establishments – but for example, people are still able to go to restaurants in groups of 4 or fewer.
        In the early stage, the alternative is to try to contain and localize it so that you don’t have to shut everything down.

        • Hong Kong is an island nation whose population isn’t even as big as my county (LA). Also there are several other counties nearby that exceed HK’s population. SoCal has 24M people overall, around 4 times the size of HK and that’s geographically a tiny corner of the US.

          They are able to totally control all traffic in and out of their country because it comes either by boat or by plane. On the other hand, the US can’t even slow down the rate at which drugs cross our Mexican border with an entire agency in charge of doing basically just that. It is in fact illegal at the highest level (Constitutional) to limit travel between the states.

          No, the US had one option with one parameter: close everything starting on day X.

          Later, when sufficient resources were in place to do control, we could re-open doing something sneaky, but the ramp up time to doing that was always too long to prevent closure.

        • It may be easy to cut off traffic from the rest of China, and there is a Hong Kong Island, but the bulk of Hong Kong is not an Island.

        • Thanks for this, you’re right. I was under the impression there was a river that made them separated, but I see looking at a map it’s not the entire border.

          Still, it’s a border that’s pretty darn easy to control compared to the border between say Orange County and Los Angeles county for which there is NO political capacity to prevent people to cross that border.

  14. I was living in Beijing at the time of the outbreak and could see people react, and worked with people who were officials during the SARS outbreak. With this background, when I returned to Europe, I “knew” governments were being foolish, and took the epidemiological modelling much more seriously than I would have without this personal experience. I didn’t possess any better perception, just greater “salience.”

Leave a Reply

Your email address will not be published. Required fields are marked *