Henderson-Wolfers non-debate

David Henderson debates Justin Wolfers on the lockdown. I stopped watching when Wolfers made the shocking assertion that without continuing the lockdown a million lives would be lost, and with the lockdown only 60,000 lives would be lost.

One way or another, that claim ends the debate. If you believe it, then you cannot argue with Wolfers, because you have to agree that it makes sense to continue the lockdown. And if you don’t believe it, then you can’t argue with him, either, because it is so outlandish.

The original purpose of the lockdown was to “flatten the curve.” That meant that, relative to a no-lockdown baseline, we would trade a lower rate of illness now for a higher rate of illness later. The question about a lift-the-lockdown scenario becomes: how many people are out there who as a result will get the disease in May or June who could have been cured with adequate treatment and who cannot get adequate treatment in the near term but who could get adequate treatment in July or August. Wolfers implies that number is close to one million. That seems to require implausibly high estimates of the cure rate as well as the likely difference in resource availability.

My estimate of the number of lives saved by maintaining the lockdown is close to zero. Probably a closer would say that my estimate is outlandish and ends the debate. Wolfers and I disagree about a hypothetical, and only if we run the experiment will we find out who was right.

Honestly, I think that what has happened to many people, I hope not including Wolfers, is that the lockdown has morphed into a miracle cure for the disease itself. Once that becomes your mindset, it becomes impossible for an “opener” to argue with a “closer.”

49 thoughts on “Henderson-Wolfers non-debate

  1. In addition to flattening the curve, isn’t the rationale for lockdown to buy time to study treatments? I would MUCH prefer to have the disease a year from now than today.

    • The death rate is low enough where you still need strict safety standards for a vaccine to be sure that it is safe enough compared to the alternative. So I expect vaccine development to be a slow as normal. I don’t think we can afford a lockdown that long so I wonder if Wolfers considered that ending a lockdown might be inevitable. So we would get all the deaths and all the costs of trying to lockdown.

      • It need not be a “cure” to be immensely helpful. For example, doctors have already scaled back use of ventilators, are putting patients in the prone position, experimenting with various anti-vitals, etc …and that’s after only a month or so. In a year I’m pretty sure our treatment will be more advanced even without a vaccine.

        I can’t say if that’s worth the “cost” because that’s subjective. But I can say that I will not resume normal life even if we say “lockdown over.”

  2. The closers seem to assume that the appropriate response to Covid-19 is to eradicate it using the same techniques that were used against the smallpox virus – rigorous contact tracing, universal vaccination. May be tough to implement against an unusually severe cold bug.

    • Taiwan did it. Only 55 (that is, 2 in a million) got it locally, and 31 on the Panshi, a naval ship. Despite 30% of the population being older than 50, there have been no new deaths or cases caught on the island, in two weeks.

      Victory. The downside of eradicating the virus (actually making sure it never had a chance to get going in the first place), is that no one is immune, and they can’t let up on everything until there’s a vaccine or the pandemic is over everywhere else.

      But wait, if you can believe it, it gets even better. Because they did it without:

      1. Treatments,
      2. Vaccines (they are contributing to crash programs to develop and test one)
      3. Very Low Error tests (they had the same noisy tests everybody else did, though lots of them)
      4. Lockdown Socialism (neither Lockdown, nor Socialism, no draconian measures, schools and most businesses open, they didn’t break the bank, no big bailouts, no crazy money printing, etc.)
      5. Authoritarianism (they are a functioning democracy with all the usual political churn and turmoil).

      They did it with:

      1. Strict border and travel controls
      2. Quarantine for all suspected individuals, with frequent follow-up.
      3. Rigorous, digital panopticon-based, track and trace
      4. Masks (and the ability to ramp up mask production 100x in weeks)
      5. The world’s best epidemic control institution, and a population that trusts and complies with its pronouncements.

      The Northeast Asians have done pretty well in all this (Japan is a question mark) but Taiwan leads the pack, which raises the question of whether there is something special about them too, in addition to the above list. People say, “Oh, they went through SARS,” – but all the rest did too.

      I think there is, which is that (A) They really, really dislike and distrust China – it’s the issue their politics revolves around – and (B) Because of China’s effective pressures to isolate them diplomatically, they can’t get the time of day from most international institutions or efforts at cooperation and information sharing.

      That means that Taiwan did not have much political trouble completely cutting off travel to China early even though this had major economic consequences with a lot of their big businesses taking hits from that move (but unable to do much more than grumble about it). In a way, it was kind of lucky for them that the virus came from China, since had it come from Japan – which they tend to like – they may have been much slower to cut the cord.

      It also means that they are forced into the international version of “independent thinking” since they get snubbed from the groupthink super spreaders, and have to go see things for themselves, evaluate the evidence independently, come to their own conclusions, and make their own decisions. They had reason to be highly skeptical of anything bodies like the WHO said, figuring (correctly, as it happens) that if the truth made China like bad, the WHO would be reluctant to publicize it.

      They sent a team of senior medical experts to Wuhan early to see for themselves what was going on, and figured out that the disease was dangerous and highly communicable a long time before the WHO admitted it.

      Note that independent thinking isn’t necessarily *good, rational* thinking, it’s just more likely to come out different from the expert consensus. In this particular case, it may have been driven early on by anti-China antagonism as much as anything else, and that prejudicial attitude just happened to pay off huge in this specific scenario.

      Perhaps they got a little lucky – or lucky in just the right way, timing when days mattered – but their performance in other ways since then indicates high levels of competence in everything else.

      If I’m right about this special factor, then that’s yet another reason to be leery of cultural and technological developments which tend to establish and spread the groupthink consensus around everywhere quickly.

      • Excellent comment.

        With heavy masks and individual protection usage, they decrease the ability of the virus to transfer from person to person enough to allow trace and isolate strategies.

  3. There is a way both the openers and the closers win the debate.

    Let’s not forget that during the shutdown we learned that 80% of COVID-19 deaths occurred in an easily identifiable 20% of the population. Forget for a moment of the horror that 25% of those we died lived in nursing homes. Now we should use that mortality information to “identify and protect” those most at risk of death or a serious health impacts from the virus. So, for the at risk group the closers like Wolfers win.

    On the other hand, we 80% of the population has little risk of succumbing to COVID-19. Having them in an open world is a huge win for bringing back the economy.

    It’s hard to imagine in our current political situation that we could have a way forward where both sides were right. But in this instance it’s clearly the case. Let’s open and close!

    • “There is a way both the openers and the closers win the debate”

      Ok, please knock it off. This is way too reasonable.

      (At this point the issue seems to be more about politics than policy, unfortunately)

    • Sounds reasonable but it won’t be acceptable to the Left until after Trump is defeated in November.

  4. Is his ridiculous claim more about the effectiveness of hospitals or that the virus does out before we all get it?

    Either way, life is too short to spend too much time listening to him

  5. I don’t think it’s about flattening the curve right now. I think it’s a belief that if everyone just does this lockdown thing, most people will be spared from getting the disease.

    People over 45 have just encountered what amounts to probably a doubling or tripling of their chance of dying. And a huge increase in their chance of having a scary, painful disease. I think many of them are experiencing something like the stages of grief as a result. The lockdown is a way of denying that our life expectancy has shortened so dramatically.

    I personally don’t see it. I don’t see how we can prevent 200 million people from getting sick in the US. It infected 20% of NYC in a couple months. It can’t be eradicated – it transmits too easily and too silently. And I just don’t see that a vaccine or cure will be available nearly soon enough.

  6. “Honestly, I think that what has happened to many people, I hope not including Wolfers, is that the lockdown has morphed into a miracle cure for the disease itself. Once that becomes your mindset, it becomes impossible for an “opener” to argue with a “closer.””

    I think it’s even worse than that.

    One of the most terrible things that can happen in our society is that some important and formerly ideologically neutral question becomes politicized, and the position one espouses about it becomes a strong signal of affiliation and loyalty to a particular team, subsequent to which all the usual mechanisms of exerting social pressure to insist upon solidarity and conformity for us, and opposition to them, go into overdrive.

    Thinking clearly about this issue was hard enough because of scientific uncertainly. It wasn’t made any easier by the emotional impact of fear, suffering, and death. But once there is a party line on the matter, thinking ends altogether.

    • “Thinking clearly about this issue was hard enough because of scientific uncertainly.”

      A quibble on an otherwise good post: there certainly is uncertainty, but not infinite uncertainty. At this point, we know quite a bit about the virus within some reasonable range.

      Read: 60+ and/or underlying health issues = significant concerns. Everyone else, not so much.

    • It’s a propaganda war in which both sides are positioning themselves for November. The cost-benefit calculus is political; it is only economic or medical insofar as those affect the political. How many independents might turn Left if Trump seems to callous in his reopening efforts? How many might turn Right if the Democrats appear too callous about destroying jobs and businesses?

      It seems clear that the nation’s hospitals could process us into herd immunity over the summer without significant under-treatment deaths. But, I would not be surprised if that resulted in close to a million deaths (which is why it won’t happen in that time frame). We’re running at the rate of about 15-20,000 deaths a week now, and that’s highly concentrated in a few big cities. Key questions are how much treatment has improved so far and how much it’s likely to improve in the next few months. What if treatment optimization is lowering mortality 10% a month? Or 20%? Improvements have clearly happened, but no one knows what the number may be. Then there’s the possibility of effective prophylaxis: a big trial in Utah is underway to show whether hydroxychloroquine works. Slowing the pandemic could save many lives beyond the problem of overwhelmed hospitals.

      • “Slowing the pandemic could save many lives beyond the problem of overwhelmed hospitals.”

        Yes, it could, but at what cost and to whom?

        (What is wholly lacking from this debate is a reasonable accounting of the costs vs. the benefits.)

        • Once you enter the business of trading money for lives, lives for money in public and at large scale–you exit the “reasonable accounting” realm. It’s a matter of subjective preferences. What further muddies any analysis is the huge amount of uncertainty inherent in any major policy decision on this issue. Maybe a giant Keynesian boost will permit a solid economic recovery after several months of torpor–and maybe not. Then there are uncertainties concerning treatments, vaccines, rate of virus transmission under different regulations, long term health effects on Covid survivors.
          In the long run, the best outcome is discrediting the corrupt, incompetent American elite and its treasonous relations with China.

      • “We’re running at the rate of about 15-20,000 deaths a week now,”

        The U.S. peaked at 2,600 deaths on April 15 and from April 16 to April 24, the average number of deaths has been 2,000 a day so 14,000 deaths the past week but this will soon decline to maybe 1,500 a week, 900 a week, 600 a week, 300 a week, etc.

        • That should be “decline to maybe 1,500 a day, 900 a day, 600 a day, 300 a day, etc.”

  7. Don’t see how lockdown helps the people in nursing homes: https://www.frontpagemag.com/fpm/2020/04/1-5-coronavirus-deaths-could-have-been-prevented-daniel-greenfield/#disqus_thread or prisons, or hospitals, which together are the bulk of infections.

    And it should be no surprise that the deadliest settings are also the most heavily regulated because by now it is no secret that our regulators are a net drain on human welfare.

    Most common sense people are going to re-enter an “opened” world defensively: masks, hand sanitizer, maintaining distance. The most vulnerable will continue to isolate. The Risky Randy population will have big parties, pretty much like they are doing anyway under lockdown. Since they are ignoring the lockdown now anyway, how much incremental difference will opening make anyway. But how many Florida spring break cases were there? 5? And did any of them turn severe?

    The real burden will be upon employers to provide low-risk working space. Employers unable to do so will remain closed until they can due to potential legal liability. Walmart and the grocery stores seem to have all developed sufficiently defensive protective regimes that other employers can emulate. Eliminating face to face meetings in the typical meeting room will probably increase productivity anyway.

    At this point the opposition to opening seems largely a matter of political optics. The problem with wild predictions, and we have 50 years of experience with this in climate change, is that nobody pays a price for prophesying doom. I wonder if Wolpers would bet money on a million dead.

    • Don’t see how lockdown helps the people in nursing homes

      I think the idea is: lots of people come from outside every day to work in nursing homes. Lots of people get the virus and are contagious but don’t show symptoms, at least for a while. Once those people get into a nursing home, they can cause a lot of sickness and death–because, to a first approximation, the virus only kills old sick people.

      The more the outside is “locked down”, the fewer people come in contact with the virus and the fewer “asymptomatic spreaders” go into the nursing homes.

  8. On the one hand, a Wuhan-style lockdown — prohibit everyone from leaving their house unless they are part of the military, have the military deliver food for about four weeks, and be very strict about procedures by those food deliverers — probably could drop the number of new cases to nearly zero. That kind of lockdown would be almost a miracle cure, with terrible side effects.

    On the other hand, no country except China will do that, and it would be a gamble on the “probably” part.

    On the gripping hand, the more we learn about treatments and acute care for the disease, the better the chances are for people whose cases are delayed by lockdowns. For example, doctors have found a number of alternatives to putting Covid patients on ventilators that seem to have better outcomes and use fewer scarce resources. This seems to me like the strongest argument in favor of current lockdowns, although it is not dispositive. As Dr. Kling points out, it assumes that most of the deaths to date are preventable by some protocol that we have not yet settled on, but that we will discover in reasonably short order. In that respect, it does assume an almost miraculous cure for Covid deaths.

  9. The core logic here is hopelessly flawed.

    It isn’t sooner rather than later. We cannot spread this out so half or so of us get it by August or September. At any rate sufficient to do that, we would be unable to control the spread. The should be obvious to anyone looking at this.

    So, the only real choice that means anything becomes “constrained” or “unconstrained”. Not sooner rather than later but “all at once”, or about what we have now. And I would argue that “all at once” will eviscerate any benefits to opening up within 30 days and will soon after be several orders of magnitude worse than what we have now. Protecting personal freedom will seem like a sick joke at that point.

    So, any “openers” need to explain how our society can survive processing a 50% infection rate over a few months. How would that work?

    • “So, any “openers” need to explain how our society can survive processing a 50% infection rate over a few months. How would that work?”

      And any “closers” need to explain how our society can afford to continue with large scale government lockdowns. How would that work?

      • The answer is obvious. All economic activity must stay within the containment bounds. Doing nothing can’t work. Engaging in economic activity that breaks containment would only be sustainable for a few weeks. We painstakingly explore what can and cannot be done within those bounds, and we move forward.

        We will find out that we can do quite a bit, but much less than we would want. Believe in markets to figure it out.

        • “The answer is obvious. All economic activity must stay within the containment bounds. Doing nothing can’t work. Engaging in economic activity that breaks containment would only be sustainable for a few weeks.”

          And, you know this how?

          • Well, it’s sort of self defining, no? Unconstrained infections means growth in the number of concurrent infections.

            Do you think it is controversial that the more infections there are, the harder it is to prevent further spread? That’s pretty safe to assume at this point, right?

            Do you think we could sustain economic activities if we had 20 times the infections we have now? Is it a reasonable guess to say no?

          • “Well, it’s sort of self defining, no?”

            Sorry, didn’t know that it was self-evident. My bad!

          • Bob

            Maybe you could clue me into what you find controversial.

            Is it your opinion that we could open up the economy broadly, while accepting a tradeoff of significant growth in the rate of infections, and still gain some net economic benefit? That we could continue to manage that infection rate? That such a strategy would be sustainable economically?

    • In a new article, Dr Geach and others do just that:

      “Based on these studies, it is clear that COVID-19 is more prevalent and less deadly than previous testing has shown. In addition, it has become clear that the disease progression is generally mild in those under 65 with limited co-morbidities and even less of a threat to children. Knowing how low the risk is for the vast majority of Americans, which is rapidly approaching the mortality rate of the flu, we can safely reopen most of the country’s economy while continuing to protect those who are most at risk from COVID-19. Once shelter-in-place orders have been lifted, those who remain fearful are free to continue to shelter in place for as long as they so choose. There is more than enough information available at this time to reopen America.”

      https://spectator.org/mapping-the-mortality-maze-how-deadly-is-covid-19/

      The evidence cited In the article seems persuasive enough.

      The testers want testing unless testing results don’t support the predetermined narrative it is supposed to.

    • Sorry I couldn’t reply to your most recent comment for some reason so I’m sticking this here.

      This is Arnold’s point isn’t it? You seem to think that states lifting lockdown orders would instantly result in massive activity that creates a 20x infection rate and a surge in deaths. But openers believe many covid customs will persist and the gradual scenario you’ve outlined will occur without the state order anyway. Difference being that the state isn’t using its infinite wisdom to choose what’s safe in a blanket manner. Openers seem to think that infections will likely increase but they won’t overwhelm most hospital systems (too late for NY) and IF infection provides a degree of immunity this will diminish future waves that lockdowns will only make worse.

  10. At the start, Henderson says that the dramatic reduction in covid-19 deaths from 2 million to the currently estimated 60,000 has been in part due to the lockdowns “but some, and possibly a large part of it, is due to our private voluntary activity.” That may be true but almost certainly the main reason we don’t have anywhere near 2 million deaths is because the quickly wriiten-up internal paper that Imperial College released was wildly off to begin,

  11. Imposing a lockdown to “buy time for a vaccine” seems insane to me. Only an academic with guaranteed income and working comfortably at home would defend this.

    Imposing a lockdown to “prevent the swamping of our hospitals” is a much tougher call.

    I do not have enough knowledge to fully answer this, but here are a few observations:

    a. The only hospitals that have been swamped so far appear to be in the poor sections of large cities. Vast numbers of hospitals in suburban and rural America look awfully stable to me, though again I am not in the industry.

    b. Caring for late stage coronavirus patients is wildly expensive. High staffing ratio and every staff member needing elaborate protections, and patients on a ventilator for weeks and needing extra respiratory therapy afterwards….

    What we need is the equivalent of the Depression-era Reconstruction Finance Corporation for health care, to allocate staff and get the protective equipment needed.
    If we had this, a lockdown would be less needed.

    c.If America had a more efficient federal system of unemployment benefits, there would be less hatred of lockdowns. An appalling number of persons are out of work and getting no money.

    If you go to war, including a war on the virus, two things to remember:

    1. Mistakes will be made

    2. You have to take care of your allies.

    Laid off Americans are allies in this war. You must take care of them even if it means vast federal spending.

  12. n a remarkable study, Maryam Farboodi (MIT), Gregor Jarosch (Princeton), and Robert Shimer (Chicago) explain that a standard epidemiological model (SIRD) ignores spontaneous, private social distancing (adaptation by individuals and firms):
    https://bfi.uchicago.edu/wp-content/uploads/BFI_WP_202047.pdf

    The authors sketch and critique a naïve model, with facts in hand:

    “The pandemic unfolds rapidly even though only one out of one million individuals is initially infected. After several weeks, a sizable share of the population is infected. The infection rate peaks after seven weeks above 31 percent. As a consequence of the height of the peak, the benefits of herd immunity do not kick in before almost everyone is sick. By 14 weeks into the infection, only 5.3 percent of the population remains susceptible and 0.19% of the population has died, more than 600,000 people in a country the size of the United States. Although the pandemic ends quickly, the cost of the disease is substantial. […] Of course, this model completely fails to capture the experience in places that did not institute any restrictions on social activity. For instance, Sweden hit 1040 total cases on March 15, 2020. By mid April, this number had risen to about 11,927 confirmed cases despite the laissez-faire approach taken by the Swedish government. In contrast, our calibrated SIRD model predicts an increase by a factor of more than 8,000-fold, during this period. Likewise, the SafeGraph micro-data document a remarkably uniform decline in individual social activity. The fact that this decline happened across the board in the US despite the large differences in policies also suggests that the basic SIRD model fails to capture a key aspect of this epidemic, namely that individual behavior responds to the risk of infection. […]” (p. 18)

    In other words, the authors estimate that the number of deaths in the US would be 600,000 if no one adapts behaviorally and if the pandemic burns through the populace in one peak. If their analysis is sound, then Justin Wolfers’ estimate of 1,000,000 deaths is hyper-naïve.

    In other words, What Arnold Kling said!

    • Their critique of the SIRD model makes sense, but I wonder what hat they pulled that mortality rate out of? NYC’s rate is already higher than that, as is Lombardy’s.

  13. I have been quite amazed how “we can never open up at all in any way” has become a party line for the Great and the Good. Especially since they held the opposite in February.

    TDS has made everyone insane.

  14. One of the reasons given for flattening the curve was saving medical resources for other causes. However, the actual practice has been to do no other medical care during the lockdown that isn’t emergency care. So, no routine examinations, no routine bloodwork done, no routine scans done for the last 7 weeks. There will be a price paid for some of this lost time in medical care, it just won’t show up for another year.

    • See Kieran F. Docherty et al., “Deaths from Covid-19: Who are the forgotten victims?”:

      https://www.medrxiv.org/content/10.1101/2020.04.21.20073114v1.full.pdf

      Summary of findings:

      “With the spreading global pandemic of coronavirus disease 2019 (Covid-19) there has been disruption to normal clinical activity in response to the increased demand on health services. There are reports of a reduction in non Covid-19 emergency presentations. Consequentially, there are concerns that deaths from non-Covid-19 causes could increase. We examined recent reported population-based mortality rates, compared with expected rates, and compared any excess in deaths with the number of deaths attributed to Covid-19. […]
      There was an increase in observed, compared with expected, mortality in Scotland (+27%), England and Wales (+35%), the Netherlands (+60%) and New York state (+26%). Of these deaths, only 43% in Scotland and England and Wales, 49% in the Netherlands and 30% in New York state were attributed to Covid-19 leaving a number of excess deaths not attributed to Covid-19. […] A substantial proportion of excess deaths observed during the current COVID-19 pandemic are not attributed to COVID-19. This may indicate an increase in non-COVID-19 deaths due to changes in routine health care delivery during this pandemic.”

  15. The idea that we could lockdown for more than a month or two was always ridiculous. The canary you should be watching carefully here are the meat packing plant closures- these will continue to spread unless we bite the bullet and start accepting that we have to reopen. To do otherwise will be self-defeating- you can’t expect the people involved in food production, preparation, processing, shipping, and retail to continue to accept the risks no one else is prepared to accept because you are undermining your own arguments.

  16. I think Arnold has stated explicitly before that his understanding of flattening the curve to merely delay, and not reduce, total infections. That’s not quite right, as I understand it. Flattening the curve (including by voluntary actions) is also suppose to reduce total infections to some extent because we would not overshoot the point of herd immunity by as much. In other words, if herd immunity sets in at, say, 70% immunity, if people tried to carry on as normal (not a realistic counterfactual), the infection would spread so fast that we may reach 90% infected. But if you flatten the curve significantly and slow the spread, then maybe you only go to 75% infected.

    Not saying that justifies lockdowns or that Woflers is right, etc, etc. Just a point about how I understand “flattening the curve” to work.

  17. Woflers did state later in the debate where he got his prediction that 1 million persons could die from the virus without intervention.

    He states that one person in every 582 residents of New York City has died so far.

    If you apply that ratio to the US population of 330 million, you do get about 1 million deaths.

    So the question would be, is New York City abnormal re: the virus?

    I would certainly think so, due to housing density. There was not time to challenge Wolters on this during the debate, and for all his virtues David Henderson probably could not have done so.

    • Right. It’s not like farmers in Idaho take the subway every day, and go back to their apartments where they share elevators with 6 neighbors at a time. There are reasons New York was hit hardest, and there is heterogeneity of towns and cities in the US.

      But the question still may be: is this thing so contagious that NYC’s attributes just accelerated the same inevitable spread that other locations will eventually have. Or maybe we eventually reach herd immunity primarily from high spread in dense cities first, and less dense areas will be mostly spared.

  18. it also seems that for many, the position taken on lockdown is a proxy for position on Trump.
    I’m not saying this is the case with Wolfers, but it is a very high correlation among my acquaintances

    • I live in SF and keep an eye on the papers, and I agree that this is very much how things seem to be shaping up—no need to think, just do the opposite of what Trump and his supporters do!

      Which of course is also just what Trump and his supporters are thinking with SF liberals as their reference point.

      We really have to try to get ideology down and pragmatism up.

Comments are closed.