Sunday, March 15, 2020

The Costs Of Overreaction V. Underreaction

Steve Sailer has a good post this morning that's worth considering carefully in light of the news coming out of Europe--and what we can glean of the news in the US: Is Boris Johnson's "Controlled Burn" Policy Wise?

The blog itself is long, but what I'll quote comes from Dylan Morris, “PhD candidate in Ecology and Evolutionary Biology interested in mathbio, popgen, and viruses, among other things,” Princeton, NJ.

Morris first summarizes the UK "strategy" (and he's more respectful than to use "" like I just did):

There was a tendentious fight about “herd immunity”, but since then the Govt’s position has been clarified to be as follows. Given: 
a) restrictive control measures carry social and economic cost 
b) behavioral research suggesting people will only comply for a short time… 
The most restrictive and effective control measures should not be imposed immediately. They should be used during a period of rapid epidemic expansion, thus maximizing the number of cases averted. This is more efficient than playing your ace on the first hand.

If you've been following what I've been writing, the problem with this is that when an epidemic is building toward exponential--i.e., for practical purposes, uncontrolled--increase, growth always appears to be slow ... until it's not.

So Morris goes on to explain his reservations about the UK "strategy":

I see two problems with this approach: 
1) Asymmetry 
It is far easier to back off from too strong an initial response than to salvage matters after an insufficient one. 
As measures are imposed, the UK can monitor case count growth. If numbers of new cases start to decline, controls can be relaxed. 
But if the UK’s initial response undershoots what is needed, it may become too late for even an ultra-aggressive China style response to stop the British healthcare system from being overwhelmed. 
Why? Exponential growth magnifies modeling errors. If the Govt’s model estimates for (a) current cases (b) control measure effectiveness or (c) time to implementation are off, the epidemic could reach overwhelming size before the Govt has a chance to course correct. 
In short, overreactions are costly, but underreactions are costly too. And everything we know suggests underreactions are just as plausible as overreactions in situations like these, less reversible, and *far, far more costly* 
2) Aggressive measures pay greatest dividends early

Morris proceeds to provide epidemiological theory with lots of statistical modeling concepts, but he concludes:

So there’s in fact no reason to hold your Ace in the hole. Your most powerful control measures should come out early, and then you can gradually relax them as the controlled burn goes on. This is more or less what the democracies of East Asia have achieved. 
This avoids the asymmetry problems of trying to time things for maximum efficiency, but it also provides a clear solution to problems of compliance and frustration with control measures: people get the worst over with immediately, and things just get more relaxed from there. 
In short, I see both frightening risk and false economy in the UK’s strategy as I understand it, and I would urge them to consider why Taiwan and South Korea see things differently.

President Trump appears to be coming to that conclusion and it also appears to be the consensus of the American people: Let's get this thing under control NOW.


  1. So, the President should be like Hoboken, NJ in which Mayor Ravinder Bhalla and have mandatory curfews.

    Now, exactly how will the curfew be enforced?

    First by local/county/ state police. After that military.

    There have been reports rumblings from Democrats to declare martial law and, not surprisingly, assailing Trump over martial law even though he has not indicated in any way that is his plan.

    What extreme are we really willing to go to right now and what justification is there for it, exponential potential or not?

    1. Martial Law? Nation wide? I'm going to need a powerful lot of convincing before I go down that path.

      Not saying you're advocating it. But this is the first time I've seen it mentioned.

    2. Michigan's governor refused to rule it out. All fine and good.

      BUT ...

      Michigan has a whopping 33 cases with no deaths. All she could have said was that there was no reason for such drastic measures, but she did not.

  2. Prevention and cure may already be here. See

    But, there's the one in a million or so that will suffer from quinine toxicity, which is usually fatal.

    1. Yeah, I've been reading about that for a few days. I wonder whether there's a test for the toxicity? After all the years of using it for malaria you'd think there might be.

    2. I can only access the abstract, so I would want to know what exactly is meant by efficacy here- does it prevent 90% of the deaths vs placebo, 50%, 10%, 5%? Having worked in the pharmaceutical field as a career, I know professionals have different views of "efficacy" vs what a layman might expect.

    3. Paul Sperry

      BREAKING: S. Korea significantly reducing lethality of coronavirus by prescribing a chloroquine diphosphate salt + zinc treatment combo to block COVID-19 viral enzyme @ 500 mg per day of chloroquine + zinc for 10 days

      10:28 PM · Mar 11, 2020

  3. To be clear, I am all for taking precautions. Restricting/banning air travel from infected countries, the push to be in less crowds, etc are good things. Wiping out TP, hand sanitizer, canned goods, meat, etc bad. Putting in mandatory curfews real bad.

    1. "Wiping out TP...."

      Nice phrasing.

  4. If you do a nationwide lockdown, it needs to knock the virus back completely. To do this requires, probably, at least a month, and it requires wide obedience of the population who do non-essential work. Also, at this point, you need the rest of the world to do it, too, else you will just be repeating this next Winter, and the Winter after that until everyone is exposed and has immunity.

  5. Remdesivir may also help.

    Reference: by the blogger Michael K., who is retired surgeon.

  6. To underpin what I am talking about ... it is now just a small step to martial law in numerous states and nationwide ...

    "The governors of Florida, Iowa, Louisiana, New York, Rhode Island and Washington have all activated their National Guard to support operations to stem the spread of coronavirus."

    OK. What will they be doing and why now? Yes, the Chinese Wuhan Virus is a new strain and highly contagious, that's a fact. Exponential growth of infected and deaths appears to be a fact especially when observing other countries.

    Yet, the National Guard? While this is affecting millions, it is not directly by infection.

    So, exactly what will state military be doing?

    From the GP article sourcing the Military Times, the states' National Guard will be ...

    "... fully involved at the local, state, and federal level in the planning and execution of the nation’s response to COVID-19,”

    According to the Military Times the National Guard will be "training personnel on COVID-19 response, identifying and preparing National Guard facilities for use as isolation housing, and compiling state medical supply inventories"

    Further, they will help "provide assistance to the states that include logistical support, disinfection/cleaning, activate/conduct transportation of medical personnel, call center support, and meal delivery.”

    That's all well and good.

    But, and this is a big but ...

    The National Guard is normally called in to quell riots and help in national disasters that directly after hundreds of thousands or millions of people. From protests to hurricanes to tornadoes, they, along with state guards, are called in.

    Are we really there yet?

    Take Louisiana, 2 deaths and 91 cases of infection.

    Florida, 4 dead, 100 or more cases.

    On and on it goes.

    Is this something to be concerned and worry about? Heck, yes, but this.

    If this is lessons learned from Obama's mishandling of the swine flue, OK, fine, but there is no indication from anyone, local, state, or federal that is using that for this response.

    Sorry, my cynical predisposition has all ears on alert.

    Again, it is good to take precautions, but I fully believe we are in a panic driven mainly by political reasoning than anything else.

  7. I am getting a bit more more positive on the US response to CoronaVirus. Trump does not play for status quo. Trump seems to have broken the log jam on testing, so the US should be in good shape soon.

    I am surprised at the speed of a CoronaVirus Vaccine being produced. One in Israeli may finish clinical trials in 3 months. Another one in the US is starting clinical trials next month.

    The Roche Testing mentioned in another blog comment is very exciting.

    Darpa (US Defense Advanced Research Projects Agency) is working on a short term therapy that would give time for a vaccine to be produced:

    The evidence for the treatments mentioned above are also very exciting.

    And warmer weather may help a lot. The study they linked to was discussed in the comments here a week ago.

  8. Lastly, no one Democrat or Republican is calling for martial law, contrary to my initial reply. There are flirtatiouns and a not ruling it out, but no calls. For that I am sorry.

  9. As bad as it's been in Italy for the last week or more ...

    They just had their biggest one day increase.