Heterodoxy in an Emergency


Consider two recent examples of heterodox thought on COVID-19.

In a blog post, Robin Hanson (an economics professor at George Mason) was more worried than most about the novel coronavirus. He concluded that "we are probably already past this point of no return" and that one significant problem is "that our medical systems have limited capacities." This post was published in mid-February, well before the idea to "flatten the curve" and maximize our health care response to the SARS-CoV-2 virus itself spread virally across the Internet. Hanson considered whether "controlled exposure," that is "deliberately exposing particular people at particular times, according to a plan" might be a good way of dealing with the epidemic.

Such a plan shouldn't just expose random people early, as they'd be likely to infect others around them. Instead, groups might be taken together to isolated places to be exposed, or maybe whole city blocks could be isolated and then exposed at once. Exposed groups should be kept strongly isolated from others until they are not longer very infectious.

Those who work in critical infrastructure, especially medicine, are ideal candidates to go early. Such a plan should only expose a small fraction of each critical workforce at any one time, so that most of them remain available to keep the lights on. If critical workers could be moved around fast enough, perhaps different cities could be exposed at different times, with critical workers moving to each new city to be ready to keep services working there.

Angry reaction predictably followed, leading to the following Hanson tweet:

Hanson was undeterred and elaborated on his proposal in a series of posts, plus a spreadsheet model, with modifications discussed here. Lately, Hanson has made a distinct but related argument, for variolation, i.e. voluntary deliberate infection at a low dose, as was used for smallpox.

Meanwhile, Richard Epstein wrote an article on March 16 entitled "Coronavirus Perspective." The article in its current form offers the following conclusion:

From this available data, it seems more probable than not that the total number of cases world-wide will peak out at well under 1 million, with the total number of deaths at under 50,000 (up about eightfold). In the United States, if the total death toll increases at about the same rate, the current 67 deaths should reach about 5000 ….

In the original publication, he concluded that U.S. deaths would reach 500; he later attributed this to a math error, acknowledging that there was no reason to think that the United States would account for only 1% of the death rate. In the revision (on March 24), he acknowledged that his revised estimate "could prove somewhat optimistic."

Epstein claimed that other models "underestimate the rate of adaptive responses, which should slow down the replication rate." On one hand, Epstein's thesis is that death rates will be low because people are taking steps to minimize contact with one another. On the other hand, he calls the measures that the government has taken "draconian," arguing "that even though self-help measures like avoiding crowded spaces make abundant sense, the massive public controls do not." Thus, if Epstein's bottom-line forecast of 5,000 deaths proves roughly correct, that would not resolve the underlying question whether Epstein's public policy recommendation was correct. If deaths are relatively low, we won't know (though could estimate with various assumptions) whether this is because of government mandates or because of voluntary distancing. Some but not all of what the government mandates would occur voluntarily, and it's hard to determine how much. Epstein's forecast is falsifiable (and, many sophisticated observers believe, likely to be falsified, as "superforecasters" estimate that it is probable that there will be between 35,000 and 350,000 deaths), and if his forecast is wrong, that would cast serious doubt on his public policy recommendation.

Epstein also made an evolutionary argument, that adaptive responses such as handwashing will exert selection pressure on the virus, so that the strains of the virus that survive will tend to be less virulent strains:

Start with the simple assumption that there is some variance in the rate of seriousness of any virus, just as there is in any trait for any species. In the formative stage of any disease, people are typically unaware of the danger. Hence, they take either minimal or no precautions to protect themselves from the virus. In those settings, the virus—which in this instance travels through droplets of moisture from sneezing and bodily contact—will reach its next victim before it kills its host. Hence the powerful viruses will remain dominant only so long as the rate of propagation is rapid. But once people are aware of the disease, they will start to make powerful adaptive responses, including washing their hands and keeping their distance from people known or likely to be carrying the infection. Various institutional measures, both private and public, have also slowed down the transmission rate.

At some tipping point, the most virulent viruses will be more likely to kill their hosts before the virus can spread. In contrast, the milder versions of the virus will wreak less damage to their host and thus will survive over the longer time span needed to spread from one person to another. Hence the rate of transmission will trend downward, as will the severity of the virus. It is a form of natural selection.

One key question is how rapidly this change will take place. There are two factors to consider. One is the age of the exposed population, and the other is the rate of change in the virulence of the virus as the rate of transmission slows, which should continue apace. By way of comparison, the virulent AIDS virus that killed wantonly in the 1980s crested and declined in the 1990s when it gave way to a milder form of virus years later once the condition was recognized and the bath houses were closed down. Part of the decline was no doubt due to better medicines, but part of it was due to this standard effect for diseases. Given that the coronavirus can spread through droplets and contact, the consequences of selection should manifest themselves more quickly than they did for AIDS.

The Epstein article appears to have had greater influence on public officials than Hanson's. The Washington Post reported, "Conservatives close to Trump and numerous administration officials have been circulating an article by Richard A. Epstein of the Hoover Institution, titled 'Coronavirus Perspective,' that plays down the extent of the spread and the threat." One can speculate that the article may have affected policy, but it is hard to tell for sure.

The New Yorker today published an interview by Isaac Chotiner of Epstein, in which Epstein elaborates in particular on his evolutionary view. The interview includes parenthetical responses from infectious-disease experts and epidemiologists disagreeing with some of Epstein's claims. For example, one expert denies that there is any "evidence that there are strong and weak variations of the coronavirus circulating." And another states that there is no evolutionary tendency of the virus to weaken, explaining, "To the extent we see that evolution taking place it is usually over a much vaster timescale."

There is a great deal of criticism of Epstein's responses in the interview on Twitter, much of it by scholars I greatly admire. Some of the critique uses the episode to take a swipe at the legal academy in general. One of the most thoughtful critiques is by Rex Douglas, a data scientist. Among other points, he argues that state-of-the-art epidemiological models do take into account that R0 (the measure of infectiousness) is likely to decline over time. It's not clear to me whether that's because of an evolutionary tendency or simply because there are fewer people to infect. But I think that it's a fair point that of course epidemiological models take into account how infection rates change over time.

At least to me, the exchange has been helpful in highlighting the particular areas of disagreement: Epstein, who points out that he has read widely in evolutionary theory, would expect the virus to weaken greatly through natural selection. At least a few epidemiologists, no doubt also conversant in evolutionary theory, do not believe that this is a significant consideration, and in any event, epidemiological models already account in at least some way for decreasing infectiousness over time. My own instinct at this point is to think that Epstein's estimate of fatalities is too low and also that Epstein's confidence in his critique was too high. But I would welcome further discussion and explanation, if for no other reason that I am curious about how much if at all evolutionary forces matter within the relevant time frame.

It is often useful for an intelligent, well-read outsider to a literature to push the experts to explain their assumptions more clearly. Perhaps the vast majority of the time, when an outsider to the literature makes a critique, it will turn out that the critique is flawed, and maybe it will waste the time of insiders who feel obliged to respond and clarify. But time spent clarifying foundational assumptions and points that a thoughtful outsider might miss is not really wasted. And every once in a while, an outsider may identify a genuine problem. Intellectual history is littered with once accepted beliefs ultimately changed as a result of the insistence of contrarians. Epstein complains in his interview that Chotiner is looking to make Epstein out to be a crackpot. I don't think that Epstein is a crackpot, but even if he were, we all know that sometimes the crackpot hits the jackpot. If there were some major flaw in conventional thinking, we would like it to be exposed during an emergency as quickly as possible. Change within a literature often occurs over generations. Policymakers should provisionally accept widely-held views of experts, but outsiders can be helpful in probing those views.

Epstein's article, moreover, can be credited with highlighting the point that even if the government does not mandate social distancing and lockdowns, people will not behave as they ordinarily would. Neil Ferguson originally predicted that the U.K. could suffer 500,000 deaths, but now projects more like 20,000. This is not a change in the underlying model, but rather an updated calculation based on an exogenous change in government policy. But can we really credit policy change for the entire difference? The 500,000 figure is based on a model of "what would have happened if no interventions were implemented (and Rt = R0 i.e. the initial reproduction number estimated before interventions)." But surely, Epstein is right that even absent government interventions, voluntary actions from people worried about getting sick (and worried also about then passing along their sickness) would lower the initial reproduction number. It is very difficult to disentangle reductions in the reproduction number attributable to government action from reductions in the reproduction number attributable to voluntary action, because governments will tend to act when people start to get worried.

Epstein's work thus adds some nuance to the debate (though that nuance was not the central point and has been lost in the broader discussion). We need to consider the marginal effect of various types of government interventions, relative to the behavioral changes that people would engage in on their own. That will allow for better assessment of marginal benefits and marginal costs of particular interventions. I don't know that I would look to epidemiologists to answer that question (though they may have produced some useful work on this issue of which I am unaware). It seems like more of a law-and-economics issue, well within Epstein's expertise. Unfortunately, it's also not an issue on which there exists a methodology to produce conclusive answers, at least not yet. My own instinct is different from Epstein's here. I tend to believe that fairly draconian government interventions are justified for the time being. But that instinct is also based in part on a law-and-economics point, that there is option value to delaying the virus and waiting for better information to develop, both about governmental interventions and about medical ones.

Could Epstein have been more careful here? He did conclude his article by stating, "Perhaps my analysis is all wrong, even deeply flawed. But the stakes are too high to continue on the current course without reexamining the data and the erroneous models that are predicting doom." Maybe this should have been his first paragraph, and maybe he should have omitted the word "erroneous," as that seems to me a premature judgment. Perhaps one should be especially careful about such disclaimers in a time of emergency, especially when political leaders might be more inclined to trust the advice of someone who is generally aligned with them on policy issues.

But Epstein is, to my mind, one of the great legal thinkers of our age. And because legal scholarship aims to influence public policy, it is inherently interdisciplinary. Legal scholars, especially those who read as broadly as Epstein, should not stay in their lanes. And, to switch sports metaphors, they should swing for the fences, even if that means that they might strike out. Like other heterodox scholars, Epstein is unfraid of making controversial arguments, like his criticism of employment discrimination law. I don't agree with that argument but think that public discourse is better when someone advances deeply thoughtful cases against propositions that almost all of us hold dear.

Hanson isn't afraid to offend people either. He swings for the fences all the time and, however many strikeouts he may have accumulated, has hit at least three home runs in very different parts of the ballpark: first, with his groundbreaking work on prediction markets; second, in his discussion of the Great Filter in the context of the Fermi Paradox; and third, in his provocative dystopian analysis of a future world with many simulated brains. (I don't mean to imply that I agree with all his work in these areas, but surely I can score an academic idea as a home run even without agreeing it.) It isn't a coincidence that his greatest work is published outside of conventional academic journals.

Hanson included disclaimers too. He acknowledged that controlled exposure was a "disturbing" idea. He noted that the system might be implemented, albeit not as efficiently, through "volunteers." He merely suggested that "authorities, and the rest of us, should at least consider controlled infection as a future option." Disclaimers, it turns out, don't do much to reduce internet outrage. Is Hanson right? His model seems plausible to me, but I haven't examined it closely enough or heard enough from critics to assess. But we need creative ideas during a pandemic at least as much as during ordinary times. Time is critical in this fight, and if he and critics have given us a head start in analyzing unconventional strategies like variolation, perhaps that will help us better assess those strategies should we come to a point where people are desperate enough to consider counterintuitive approaches.

Disclosure: I know both Hanson (who was an interdepartmental colleague of mine when I was a law professor at George Mason) and Epstein (who was at University of Chicago when I visited there in 2005).

NEXT: "Wuhan COVID-19 Death Toll May Be in the Tens of Thousands"

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Let me guess the TLDR summary: It’s hard to make predictions, especially about the future.

    (Attributed to the late Yogi Berra)

    1. They swung for the fences and struck out. (More like Reggie Jackson than Yogi Berra.)

      But having said that, I think OP misses that law professors shouldn’t really be sticking their necks out and sucking up valuable oxygen from scientists who actually study this stuff. Epstein’s arrogance in that interview with Chotiner is just astounding- he really thinks because he spent maybe 2 percent of his career studying medical issues, that he’s as qualified as an epidemiologist to be opining about this stuff.

      1. Meh. Expertise is highly overrated. This Chotiner guy seems a repulsive little weasel.

        1. So is bullshit. Which is all Epstein offers.

          1. Eh, it’s just empty populist signaling with an insult so you know he’s serious.

  2. Neither of these people have any business making health predictions or recommendations.

    They lack training, experience and education in the complex medical areas one needs.

    Its reckless, their institutions probably need to discipline them.

    1. Anybody can make predictions, recommendations, etc. Whether or not they should be listened to is another story.

      1. ” Whether or not they should be listened to is another story.”

        Yes, that’s not inconsistent with what I said. Opinions are like a**holes after all.

        They are using their reputations in unrelated fields to bolster their opinions. That is what is bad.

        1. I agree 100%. In the interview Epstein comes across as a know-it-all jerk, and much of what he says sounds like word salad.

        2. Bob, you are going to get some bipartisan agreement on this one.

    2. Since when has that ever stopped anyone on the internet?

    3. Which institutions are you proposing discipline people for having opinions?

  3. Volunteers for exposure? Found on the beaches in Florida and California. Both young and healthy. Probably also not well-informed, but you can’t always get what you are looking for in volunteers.

  4. My prediction is that the heterodoxy will outlast the emergency. In particular there will be no consistency in the method of counting cases and deaths from country to country (never mind China’s invented numbers.)

    What counts as a coronavirus death will be different from place to place, and will vary according to political expedience. And there will be many learned papers on overall death rates, which will be largely ignored because the moment will have passed.

  5. “…and that one significant problem is “that our medical systems have limited capacities.”” Can you show me a country where medical systems have unlimited capacities? I’ll try to emigrate to there.

    1. Communist China comes to mind.

      However, it turns out that welding you into your apartment is considered medical treatment.

  6. Let them have a go at one of the Millenium Problems if they want a challenge.

  7. You can’t really attribute any government action in the US for any reductions yet. Long incubation period means you won’t see any results of government mandates until after Easter. Everything before that was already locked in.

    Why are no epidemiologists not talking about what the long incubation period means for the effectiveness of interventions or the reality of likely current infection numbers? It’s absolutely shameful that they’re pretending to the public that stuff the government does today can matter tomorrow or next week.

    Which is another reason outsiders should be encouraged to talk – healthcare professionals have an incentive to deceive the public to pursue their agenda, and aren’t trained to evaluate indirect health consequences (or non-health consequences) in emergency situations. (Example: Stress from lockdowns will probably kill more people than covid-19 ever would: https://consistentprinciples.wordpress.com/2020/03/23/the-powerful-reason-covid-19-lockdowns-may-cause-more-deaths-than-they-prevent/ , includes cites to relevant medical literature)

  8. So Robin Hanson wants to round people up and infect them with a potentially fatal disease, and gets his feelings hurt when he gets rude, disrespectful responses?

    Well, here’s a rude response: Fuck him.

    1. We’re such babies about risk these days. Before vaccines, people had to live with a much higher likelihood of dying or severe disfigurement from polio, just to pick one disease.

      Allowing people in low risk groups to *volunteer* for intentional exposure (possibly with compensation) is an eminently reasonable policy suggestion that has the fastest route to herd immunity. Not to mention the route with least casualties overall (because it would result in fewer vulnerable people being exposed). And if it’s volunteers, then its a risk they can choose to take or not.

      1. We’re such babies about risk these days. Before vaccines, people had to live with a much higher likelihood of dying or severe disfigurement from polio, just to pick one disease.

        Unlike you, I don’t long for a return to those glorious times.

        Allowing people in low risk groups to *volunteer* for intentional exposure

        I don’t see anything about volunteers in,

        “maybe whole city blocks could be isolated and then exposed at once.”

        Allowing people in low risk groups to *volunteer* for intentional exposure (possibly with compensation) is an eminently reasonable policy suggestion that has the fastest route to herd immunity.

        Well, it’s a policy suggestion, anyway. And you seem awfully confident about that herd immunity business. Got evidence?

        1. I don’t long for a return to the past and its problems, but just because we’ve solved some risk doesn’t mean people shouldn’t be willing to live with a reasonable low risk in times of crisis. Putting the entire economy on hold for disease risk vastly less than that present in 1950 seems ridiculous. (Indeed, risk lower than the individual risk for each of many diseases that were endemic in human populations in 1950). (Not to mention that we do live with some risks every day, some of which kill more people than covid-19 ever will. And there’s decent evidence the stress from the lockdown will kill more people than the virus – linked in a post above).

          Evidence for herd immunity? It’s a general concept where enough people are immune that it prevents most transmission of a disease from happening. The only way it doesn’t work is if a disease mutates rapidly enough that you can’t develop immunity (see, e.g., the common cold or flu).

          Basics of herd immunity: https://www.heart.org/en/news/2020/03/25/covid-19-science-understanding-the-basics-of-herd-immunity

          If covid-19 immunity isn’t possible, then there will never be an effective vaccine, and we might as well get on with our lives and accept some people are going to die. (Well, we need to accept some people are going to die anyway, but accept more than that are going to die).

          I’ve seen estimates of 60 to 70 percent of the population needing to be immune (that is, in the absence of a vaccine, have had and recovered from Covid-19) to achieve herd immunity. It depends on the R-value of the virus, which is probably somewhere between 2 and 3. (R measures how many people an infected person will infect, on average. The flu is close to 1 and Mumps is ~12).

          That particular proposal didn’t mention volunteers, but another one did (the variolation one, iirc). And how extreme an approach needs to be taken depends on how serious covid-19 actually is. I don’t feel covid-19 is so serious as to involuntarily expose people, but letting people volunteer seems like a reasonable step, especially if it lets people get back to work.

          1. I know what herd immunity is.

            What I wanted evidence for was the claim that Hanson’s strategy is the fastest way to get it.

            I’m glad you see that Hanson’s first proposal didn’t involve volunteers. What do you think is the appropriate response? How about from someone like me, who is well into the age group at high risk?

            Somehow, “Sure, Robin, come around and kill me, and a lot of my friends and relatives, whenever you’re ready, if it will help test your theory,” seems inadequate. For me, “Fuck you, Hanson,” seems more appropriate.

            1. Well, my initial thought is the vulnerable should isolate, and everyone else should work or go to school as normal.

              Intentional exposure to low risk volunteer cohorts is a viable strategy, but I think that’s more than warranted given the data we have currently.

              I think it’s terribly flawed to send 20-something college students home to their likely over-50 parents, putting possible carriers in close proximity to people at increased risk. (FWIW, I’d call 50 the break point where lethality starts increasing to risky enough that you should consider isolating).

              I also think we desperately need a serological test on a population-wide basis that can tell us *how many people have been infected*, not just those who go to a hospital with severe enough symptoms to be tested. If there’s a government response here, that’s what the focus should be.

              Relative to real infections, our data is woefully incomplete and at least a week behind actual. Death data is even farther out of date (some people now hospitalized will die). A primary question here is if any government response will do or have done *anything* to halt the spread of the disease, because we’re flying blind on actual infection rates. If the 86% undiagnosed (because extremely mild or asymptomatic) estimate from Wuhan is even close to being true, much of the population, especially in places like New York, have likely already been exposed, and quarantines are too late to do much of anything.

              Anyway, since you know what herd immunity is, then the fastest way to get it means finding the fastest way to expose a lot of people to the virus. Intentionally exposing people would be faster and more systematic then just letting the virus wander around the population, and could specifically target low risk members. Basically, speed to herd immunity varies directly with exposing people (who recover) to the virus. How do you imagine intentional exposure would be slower than some other method?

              (The other bonus of intentional exposure is you know *who* is immune after they recover, so you can use immune people for tasks like taking care of vulnerable members or medical staffing for covid-19 patients with less risk than presently. A population-focused serological test could help determine the same thing, if enough people have already been exposed).

              1. Well, my initial thought is the vulnerable should isolate, and everyone else should work or go to school as normal.

                Right. So someone with autoimmune Crohn’s disease is vulnerable, right? Answer is “Yes,” of course. Vulnerable but conscientious, say. Self-isolates perfectly. Then gets a bowel bleed, because Crohn’s. Time for the hospital? Well, no. Because the hospital is busy cleaning up the wreckage of Epstein’s experiment, which makes it way too risky to go there. Dozens of hospital staff have already been infected themselves. Or, wait a minute. Yes, go to the hospital, because that bowel bleed could kill you right now! Bowel bleeds can go bad really fast.

                You fell for it. You went to the hospital and you got the virus—and you’re 63-years old, with a cranky immune system. Triage says, no treatment for you. So long. Too bad your exposure at the hospital also gave the virus to your widow, before you became symptomatic.

                The Epstein schemes tout a degree of control which in a world of happenstance and virulent viruses simply cannot be delivered. Deliberately infecting younger people would jam the hospitals with coronavirus patients anyway, and infect hospital staff, just as a do-nothing policy would do. To maintain the kind of security an Epstein-style city-block experiment would require, you would have to be willing to kill not only anyone who developed a bad case of the virus, but also anyone who had any other need for emergency medical care. Once you let anyone out of the virus compound, for any reason, the virus goes out too. And once out—in an un-locked-down society—it spreads beyond control.

                The alternative is bad, but better. Everyone needs to be locked down, to attenuate the overall virus concentration, and preserve to the maximum extent possible the medical capacity to deal with continuing pandemic. While waiting, build capacity—hospital beds, testing capacity, contact tracing, masks and other protective gear, and ventilators—so that when economic exigencies leave no choice, as they will eventually leave no choice, we can ease the lockdown and deal better with what happens. Someday, there is a vaccine.

                As I write this, Massachusetts has been generally locked down for nearly 3 weeks. Initial compliance was spotty, but after the first week, rush hour traffic was down to about 10% of normal. During the first 10 days, the state-wide number of confirmed cases wasn’t notably larger than it would be in one of Epstein’s city-block tests, climbing slowly through the low-to-mid hundreds. As I write this, the state-wide case total is pushing 6,000, and will surely surpass that number today.

                The state locked down too late, and exponential growth is taking off. The next few generations will probably take the case total over 100,000, and guarantee the deaths of 1,500 Massachusetts victims or more. Epstein’s experiments would seed exactly that same outcome, wherever they were tried.

                The three principal hospitals in Boston—including Brigham and Womens, and Mass General, and Dana Farber—among the best hospitals in the world—report among them more than 150 staff infected with coronavirus. They probably ought to be no-go zones for anyone except corona virus patients, but of course that cannot happen.

                This is a grim situation. That must be beyond controversy. The situation in New York is already many times worse than in Massachusetts. This is no time to teach the controversy, as the OP and Epstein seem to be advocating.

              2. So do you or do you not find Hanson’s idea of involuntarily exposing city blocks at a time acceptable?

  9. Epstein is just an arrogant Dunning-Kruger case with an axe to grind. He cannot be trusted to accurately recount the history he is making arguments from; why would anyone listen to him about a specialty he suddenly found himself interested in?


    I don’t mind outsiders taking interest in my field (not epidemiology); I try to be kind to cranks and help those with sincere interest. What I hate are dilettantes, frequently looking for press, abruptly spreading misinformation. And I don’t deal with life-or-death issues.

    Epstein should be ashamed of himself. But he won’t be.

  10. Legal scholars, especially those who read as broadly as Epstein, should not stay in their lanes.


    Lawyers should stop thinking they know everything. Epstein says,”I’m trained in all of these things,” referring presumably to epidemiology, virology, statistics, etc. I see no evidence on his CV.

    “I’ve worked on evolutionary theory for forty years in its relationship to law.”

    WTF does that mean? His claims about the changes in the AIDS virus are refuted by Daniel Kuritzkes, the chief of the infectious-diseases division at Brigham and Women’s Hospital, in Boston, who actually knows something.

    1. I know nothing about this, but from this piece it looks as if some serious scientists agree with Epstein about the AIDS virus: https://www.bbc.com/news/health-30254697

  11. Epstein hilariously calls Chotiner an amateur, totally unaware tat he himself is the worst kind of amateur – one who thinks he’s a pro.

    His discussion of modeling is incoherent.

    There’s an underlying, standard model that you want to use, and the question is how you stuff it full of parameters. That is, numbers you add into it to make what’s going on.

    OK. Fair enough

    And, so, the situation that you get is you cannot use any exponential system because essentially then everybody is going to be dead, because things just keep doubling, doubling, and doubling.

    They are not using exponential systems, partly for that very reason.

    So you have to develop a model which is going to explain why there’s a fairly rapid increase at the outset, and then why the thing starts to turn flat, ultimately down, and then disappears. That’s the strategy that you have to do.

    And Epstein is the guy to do it? I thought he wanted to use the standard model. Look, what he is saying is that all the smart people – smarter than him – who have spent decades, careers, working on this – have missed a simple, blindingly obvious point that he, the Great Epstein, sees, and if they would only listen to his blather about general equilibrium and Darwinian economics they would see the light.

  12. Epstein’s article, moreover, can be credited with highlighting the point that even if the government does not mandate social distancing and lockdowns, people will not behave as they ordinarily would.

    Duh. There’s an insight. People get cautious when there’s a dangerous contagious disease running around. Only one of our leading legal scholars would have noticed that.

  13. One of the huge problems with prophylactic measures is that if the measures work, the threat disappears, and so does the evidence supporting their use. The more effective the prophylactic measures are, the more unnecessary and irrational they will appear in hindsight. This is especially true of long-term prophylactic measures, which if widely practiced and successful may appear as superstitions to later generations disconnected to the problem the measures were trying to prevent.

    Chief Justice Roberts addressed the issue in his famous “elephant whistle” comment. But the problem is much harder than he lets on. Counterfactuals are notoriously hard to prove. When the elephants don’t arrive, we don’t actually know for sure whether the whistle kept them away or not.

    In general, people have much more confidence in their state of knowledge than they have any right to be, especially when it comes to predicting the future. And the sort of people who become lawyers and judges seem to be especially overconfident, often simply assuming that since they have a higher social position, their ideological predispositions and gut instincts must be more reliable than other peoples’, if not completely infallible. It isn’t necessarily so.

    1. In Roberts’ case the elephants showed up as soon as the whistle stopped.

  14. There is too much rally around the flag spirit. Too much call to “listen to the experts” Too much rush to silence other opinions. Epstein may be wrong, but he doesn’t sound like a crackpot.

    Why do I say too much? Because I think the experts are caught in a bald faced lie about the masks. They only had enough masks for health care workers (actually not even enough for them) so they told the public that masks for everyone would have no value. Actually, credible evidence from Asia suggests that masks for everyone slows transmission of the virus by 50%. It may be one of the reasons why Asia may fare better than other continents in this virus crisis.

    I believe that if we had 100 masks for every person, that our authorities would have mandated that we wear them. To claim zero value, was a lie.

    Now that our most trusted authorities are caught giving fake news, I am much more open to listening to alternative sources of advice. Putin and Trump are hardly the only purveyors of fake news.

    1. Except I didn’t hear experts saying masks weren’t useful for people, just vastly more useful for health care workers.

      Maybe there were some experts that were buying into the ‘tell them what they need to hear not the truth’ but I don’t think it was anywhere near the consensus your comment implies.

      1. They were lying. It was undoubtedly necessary to lie but they were lying their asses off. To someone like me who uses N95 masks all the time in woodworking the lies were risible: they’ll put them on upside down, they won’t pinch the clips, they won’t be able to wear them for more than 15 minutes. Ridiculous assertions every one. When masks become widely available will they be able to convince people to wear them?

        1. My point is that the news I got said masks would be useful to everyone. I’m not hoovering up every bit of Covid news, but I never heard the lie/lies you and anorlunda describe.

          Not saying the lies didn’t happen, but you both rest your argument on a large-scale consensus position that I’m am not tracking.

      2. sarcastro : Except I didn’t hear experts saying masks weren’t useful for people, just vastly more useful for health care workers.

        Well, let’s go all the way back to……yesterday :


        Virologists are sceptical about their effectiveness against airborne viruses.
        Since the start of the coronavirus outbreak, the official advice from the World Health Organization has been clear. Only two types of people should wear masks: those who are sick and show symptoms, and those who are caring for people who are suspected to have the coronavirus.

        So, for the general public, they are not recommended because:
        • they are not reliable protection (they can get contaminated by other people’s coughs and sneezes)
        • Washing your hands often and keeping your distance from others is a more effective way to avoid coronavirus
        • there is a risk of contamination when you put masks on and take them off
        • masks might offer a false sense of security

        Homemade facemasks are not recommended because there is no guarantee that they will provide enough protection, even when used in the correct way.

        Note, this doesn’t mention anything about masks being effective, but in times of shortage best kept for the medical profession. This is solely about whether masks work for the general public.

        So let’s knock over those pins, one by one.

        1. “they are not reliable protection” – neither is an NBC suit. Nothing is reliable at a rate of 100%. The question is whether they are useful or not. Do they slow the spread of the virus ?

        2. “washing your hands and keeping you distance from others is more effective” – ditto living in a plastic bubble all on your lonesome. Again the point is not merely irrelevant but bonecrushingly stupid. The question is whether using a mask helps slow the spread of the virus, not whether washing your hands helps more.

        3. “there is a risk of contamination when you put masks on and take them off” – yes there is. Ditto your spectacles, your pants, your false teeth. And what contaminant might be on the mask, that you risk contaminating yourself with when you take it off ? Why those very same droplets that would have hit you smack in the face if you hadn’t been wearing a mask.

        4. “masks might offer a false sense of security” – altogether now “don’t wash your hands as it might offer a false sense of security.”

        It’s not that I mind them being stupid. My objection is to their absolute insistence that everyone else is feed-your-husband-fish-tank-cleaner stupid.

        1. Come on Lee. I specifically said Not saying the lies didn’t happen, and Maybe there were some experts that were buying into the ‘tell them what they need to hear not the truth’.

          You saw something incorrect on the bbc. And now you’re on a crusade against all experts for deciding to lie to you for your own good? That’s confirmation bias.

          (Note: I don’t think such dissemination is actually for your own good. I’m not one to believe in the efficacy misleading the public)

          1. 1. I am not – on this occasion 🙂 – criticising you.

            2. I am merely pointing out that the line that masks are of no use for the general public has been widely publicised.

            3. The fact that it is one story from the BBC does not mean that there has only been one story, and only from the BBC. As the story notes :

            Since the start of the coronavirus outbreak, the official advice from the World Health Organization has been clear.

            This a parroting of the official WHO line, which has been duly parroted by all the media parrots, not just the BBC parrot. And not just in the distant past, but throughout.

            4. You may have missed it. I have not. For Mrs Moore, from whom I am currently at a social distance, keeps on sending me links to ever stupider stories with appropriately waspish remarks. Along the lines of “Good grief – these people are stupider than you !”

            1. 1. OK. Well then, criticize away. I am very much against the idea of experts gaslighting the public towards whatever policy they want. Experts should stick to actual facts, and our duly elected/appointed policymakers should turn those into the appropriate policy.

              2. Noted; agreed.

              3. I do NPR, CNN, WaPo, and various political forums around the Internets. Never heard a breath of ‘don’t use masks they’re not useful to you.’
              That being said, while I’m against this overgeneral ‘experts all suck’ nonsense, you make a great case for condemnin the WHO for stepping way outside their lane and confirming every elitist stereotype.

              4. Fair enough. I don’t have such a news aggregator; my girlfriend sticks to economics which has it’s own set of ridiculous hot takes. My political forums are a sight to see though – from the ‘it’s a hoax and revolt is coming’ to ‘The real story is Biden is a rapist!’

              1. 3. I do NPR, CNN, WaPo, and various political forums around the Internets. Never heard a breath of ‘don’t use masks they’re not useful to you.’

                Then you weren’t paying attention. It was all over the news.

                I can’t post links because Reason’s software will eat the comment if I do. But here’s a particularly egregious example where they’re trying to scare people into not wearing them:

                Masks may actually increase your coronavirus risk if worn improperly, surgeon general warns

                1. And since you mentioned the outlet, here’s the Washington Post:

                  How to prepare for coronavirus in the U.S. (Spoiler: Not sick? No need to wear a mask.)

                  Although the article is datelined March 12, the URL implies it was originally posted on February 26. And it naively suggests that all people need to do is wash their hands and stay home if they’re sick, with no other changes to their lives.

                  And it quotes CDC and Surgeon General as saying not to wear masks.

        2. The deception in that statement is that it focuses entirely on protecting the person wearing the mask. The value to humanity is that wearing the mask slows the spread of the virus to other people.

          It’s just common sense. Why have surgeons worn surgical masks for centuries? Not to protect themselves but to protect the patient from drool dripping into the open cavity. (Yuck)

  15. Here is a takedown of Epstein that is long, but worth reading.

    1. Since you don’t seem to care about offense, I’ll ask. I know you’ve been here longer than me, so I’m curious; what exactly about you, if anything, is libertarian? I read this takedown (the fact you’re even phrasing it as such is troubling enough) and it is one of the most authoritarian things I’ve ever seen from a logical perspective.

      Introduction: the very premise of needing a code or procedure for “non-experts” to discuss a subject, while simultaneously claiming to be conscious of gate keeping, is laughable. Any standard devised is gate keeping. Someone has proclaimed themselves the expert and judges who shall be qualified.

      Lesson 1: this is bullshit of the highest order and I must admit I have a very poor opinion of you now having shared it voluntarily. This is literally a baseless assumption and character assassination. This is the scientific method at work? Assuming that someone who questions widely and readily accepted theories is essentially a troll and that they are “dead” to the author of this polemic? I mean holy fucking shit, how dense are you? You set yourself up to apply the scientific method and not even a sentence later throw out the baby and the bathwater. On the internet, we would consider this a highly effective form of trolling by dismissing a serious and genuine effort by someone to participate in a discussion.

      Lesson 3: the author continued from this initial point by begging the question. Here’s this awesome article that’s totally worth reading and should help non-experts understand how to approach the subject and I swear I’m not gatekeeping, but I’m going to accuse a non-scientist of gaslighting for not making corrections in a scientific manner that I approve of. Also I’m not even going to address my assertion about the logic not being addressed, even though Epstein’s footnote clearly addressed that logic. He claimed 500 US casualties based on the rate of change globally and not based on the US proportion of the world population. That’s not tweaking a zero or a comma. That’s a fundamental error in logic he made initially and corrected because people noticed it makes no sense. But hey, Epstein is the one gaslighting, not the author. Trust me, I’m the expert.

      Lesson 4, State of the Art: this part was fucking rich to read because immediately after it, the author goes on about strawmen. Great way to sabotage your own argument about strawmen by prefacing it with a strawman of your own. I’m struggling to find anywhere in Epstein’s drafts any claim that R0 is assumed to be constant. What I do see is a concern over omitted variable bias. He feels that the factors he addressed are not adequately addressed by the models. Welcome to the world of statistics, where a bunch of people argue about what belongs and why.

      Lesson 4, Basic Facts: This is gatekeeping at its finest. No, a cursory search does not suggest an incubation period of 5 days. A cursory search suggests a MEDIAN incubation period of 5 days, meaning 50% of infected exhibit symptoms before then and 50% after. Nearly all infected exhibit symptoms within a two week window, which is probably why he went with that timeframe. After all, it would be pretty fucking stupid to tell someone after 5 days they’re in the clear. Even the CDC lists the timeframe as 2-14 days.

      Lesson 4, Cherry Picking: using South Korea’s data to estimate the true CFR is not cherry picking, it is common sense. A country that fails to contain the virus properly and has deaths resulting from purely economic factors doesn’t give you any idea of how likely people are to die from the disease itself. The whole purpose of regression analysis is to hold other variables constant. If we’re trying to be as truthful as possible, it would be best to use data that can include robust testing in its model, as opposed to other data sources which will have larger stochastic error terms.

      Lesson 4, Consensus: since we’re all pretending to be alert and questioning underlying assumptions, why does the author use an equivocation fallacy here? They claimed that “Young Americans” still require medical treatment to survive. Never mind that hospitalization does not mean they won’t survive otherwise, but this “young” category requires that we consider the 45+ crowd, which comprises 80% of hospitalizations, to be “old” and 20-44 to be “young.” Sorry man, but nobody thinks that 40 is young. You’re basically a boomer at that point.

      Lesson 5: some more irony here, but the author cites NY and MA in a rather meaningless way. He cites the rate of change in deaths as being rather high, but this doesn’t mean much on its own. An increase in deaths from 1 to 2 is a 100% increase. This is a common tactic that ideologues engage in, switching back and forth between nominal values and ratios, whichever suits their argument best. Ironically, the very same cherry pick the author accused Epstein of.

      Lesson 6: it is not contradictory, nor unprincipled, to suggest that the US trends will “break” just like other nations, but also to argue that we should not copy the measures they have adopted. In case you forgot, this is part of a libertarian site and most of us aren’t willing to turn the world into a police state just so that we can virtue signal about trying everything we could to save just one life, consequences be damned. Furthermore, the “contradiction” at the end about certain factors applying and others not is hogwash. One case could very well be analogous with regards to spread, but perhaps not analogous with regards to death.

      Lesson 7: congratulations for realizing the weakness of statistical modeling and regression analysis; incomplete data. Unfortunately, that’s not an argument.

      Conclusion: just lol. Seriously, the author is a condescending asshole. Didn’t his mother ever tell him about the golden rule?

      Here’s my own conclusion: this is why everyone hates academics. I don’t know if it’s collegiate culture or what, but people like this argue about the dumbest shit. It took me a lot of time just to read through this bullshit and offer an initial reaction to something that I could immediately tell was wrong. I too am not an epidemiologist, I’m just a dumb kid starting his career in finance with a BA in Business and Economics with a focus in Econometrics.

      The author made me laugh a bit at the end complaining about how much easier it is to make an argument than it is to take one apart. Yes it is. That’s how logic works. Taking apart an argument by just assuming that someone else meant something is meaningless. Address the person directly, ask for clarification, and stop assuming things so damn often. When you said lawyers need to stop thinking they know everything, I spit my drink. Take some of your own medicine.

  16. Epstein is wrong for a number of reasons.
    — His conclusions are largely based on China’s reported death tolls. We know Chinese numbers are propaganda. They reported numbers large enough to be semi-believable, without being so large as to scare the world. A true Wuhan death toll of 40,000-100,000 wipes out his entire argument.

    — His argument about the virus evolving to be less virulent over time is also bogus. This particular virus is not a particularly virulent one, as it kills only 0.5-1.0% of the infected (about 5x-10x seasonal flu), but is unique in that most spread is by asymptomatic people and the disease has a long incubation period. From the perspective of the virus, there is no advantage to being less lethal. It already spreads just fine, and the asymptomatic infected can be contagious for weeks or longer. The relatively small number of deaths are just minor collateral damage from what the virus does best – spreading. But a 1% death rate does equal 3 million dead in the US in a short amount of time.

  17. Regarding angry reaction by idiot masses, around 1990, the UK had a TV chess match between a grandmaster and people calling in to vote on the next move. Would they play a bad game, or the game of a science fiction multi-consciousness god?

    Take a guess.

    Twitter has turned received wisdom into the burblings of the same folks, so no discussion of thoughtful positions need occur.

    1. Except that the populist side is Epstein, and the ‘idiot masses’ are endorsing the experts here.

      You can’t have your elitist cake and eat it too.

      And BTW I use elitist not in the pejorative sense, just in the opposite of populist sense.

  18. Epstein seems like a crackpot to me.

    He claims that a “the virulent AIDS virus that killed wantonly in the 1980s…gave way to a milder form of the virus.” The data he links to don’t demonstrate this. It shows that the amount of time that people live after being diagnosed with AIDS increased over time. Epstein acknowledges that, “Part of the decline was no doubt due to better medicines,” but asserts that, “part of it was due to this standard effect for diseases.” That’s a pure assertion. He offers us no reason to suppose that the *all* of the decline is not due to better medicines. I’d add that there are other possibilities besides these two. Epstein may have merely been speaking loosely when he claimed that AIDS kills, but AIDS is not fatal by itself, so anything that prevents people with AIDS from being killed by other diseases despite having a weakened immune system will extend the life of people with AIDS.

    Variants of the AIDS virus are important for developing treatments because a medication that works against one variant of a virus won’t necessarily work against another. That means there may very well research that would tell us whether Epstein’s assertion is true or false. Epstein doesn’t give any indication that he’s looked for it.

    Making confident assertions without any evidentiary support, and apparently lacking having any interest in looking at the evidence that may exist? Sounds like a crackpot to me.

    1. Right.

      Look. If Epstein wants to challenge prevailing opinion that’s fine. But what he’s doing is bullshit.

      He essentially claims that the models the experts are using are wrong, because they fail to take important factors into account. OK, Richard, show us your model. Tell us what parameter values you would use. Show us how it was more accurate than standard models in the AIDS crisis, say.

      But he didn’t do that, because he wouldn’t know where to start, and if someone told him where to start he wouldn’t know where to go from there.

      A bunch of BS about general equilibrium and “evolutionary theory … in its relationship to law” is not an argument.

      1. He essentially claims that the models the experts are using are wrong, because they fail to take important factors into account. OK, Richard, show us your model. Tell us what parameter values you would use. Show us how it was more accurate than standard models in the AIDS crisis, say.

        I’m not sure you have to have your own model to be allowed to criticise other people’s.

        Same principle as science in general. You can throw stones from anywhere.

        Indeed, even the fearless reptiles of the press are allowed to criticise without being in full possession of their own grand plan.

        1. I’m not sure you have to have your own model to be allowed to criticise other people’s.

          You do when you start making projections as Epstein does.

          If he just wants to say that he thinks the epidemiologists are overlooking an important point, then fine, but leave it at that. And if they say it’s not important, and give reasons, then explain why you think it is, with some actual facts. Don’t just say, “Gee, I’ve read a lot and I think you’re wrong.”

          Even if his description of the evolutionary process is correct, hs conclusion doesn’t automatically follow. Even if there are multiple strains, that doesn’t necessarily imply deaths will be less than projected. There’s work to be done to establish that. I doubt he’s capable of doing it.

  19. public discourse is better when someone advances deeply thoughtful cases against propositions that almost all of us hold dear.

    Indeed it is. And if Epstein had advanced such a case it would have improved public discourse. As it was, it not only didn’t improve discourse, it did damage to the country because the imbeciles running the country bought into it.

    Here’s the point. He is not being attacked for presenting an unorthodox argument. He is being attacked because his non-argument is incoherent nonsense.

  20. Still waiting for a discussion of how we will ever develop ‘herd immunity’ if we hide in our prison-homes for months at a time.
    I guess when starvation from lack of work kills off the younger and more healthy, and loneliness and starvation from lack of open stores kills off the elderly, we will find out that if the herd consists of only one animal, it really doesn’t matter.

    1. A great example of uninformed surety.

      It’s a pandemic. Even the current measures mean a whole bunch of people will get it. Virgin soil epidemics have a key word to them, and you’re only virgin once.

      Starvation is not going to be an issue; food stores are not closing.

      1. you’re only virgin once

        In the epidemic sense, yes and no.

        Being infected once does not guarantee lifelong immunity. Whether you become immune after infection and recovery, and for how long, varies from pathogen to pathogen.

        1. Not lifelong, but from what I understand, if people lose their immunity within a year or so, that’d be a pretty novel trick for a virus.

          There will always be some small amount of reinfections; immune response is not perfect. But if such an effect gets statistical, I’ll be pretty interested from both a ‘we’re all going to die’ and a scientific point of view.

          1. The common cold has achieved this trick.

            Admittedly it has a whole golf bag of different viruses to choose from, all of which keep evolving.

            If wikipedia is to be believed – we manage to get infected two or three times a year, each; or twice as often if we’re kids. So that implies a modest level of short term immunity developing as we grow up. But not much.

            1. The common cold isn’t actually a single virus so I’m not very concerned about the trick of being countless different species.

        2. Correct. Which is a rather large flaw in Hanson’s suggestions.

  21. In case someone wants an example from the left, here is a New York Times journalist accidentally (presumably) getting into it with a Yale University epidemiologist: https://twitter.com/jmartNYT/status/1244975910350192640

  22. Yesterday the Imperial College’s team of researchers issued a paper estimating the number of infections (actual infections, not reported cases of COVID) in several countries. Rather quietly (or at least I haven’t heard much reporting) the modeling used by the researchers predicted that there were “orders of magnitude fewer infections detected than true infections, most likely due to mild and asymptomatic infections as well as limited testing capacity.” The model spit out a prediction that the following countries had approximately this many infections (Low end estimate – medium – high with 95% confidence within the three intervals) as of March 28:

    Austria 97,042
    Belgium 421,800
    Denmark 61,633
    France 2,009,700
    Germany 596,088
    Italy 5,927,040
    Norway 22,008
    Spain 6,999,000
    Sweden 313,720
    Switzerland 274,240
    United Kingdom 1,793,880

    The death rate from COVID is not obvious at this point. Some countries may be undercounting, some overcounting. Different countries will have different infection fatality ratios based on behavior, demographics, medical infrastructure, etc. And the death rate will lag the infection rate, since many people who are infected but who will die from COVID, will not die for days (or even weeks). This effect should not be ignored, but I don’t know how to calculate it. However, assuming that the Imperial Academy’s infection prediction was correct, and overlaying that with the reported deaths from those countries as of March 28, the IFR for each drops to…

    1. Because this would be a crude IFR, due to the lag in deaths, I have decided to use the high end (95% confidence) deaths on March 28th from each country. This may still be seriously undercounting the infection:death ratio. But we get:

      Austria 0.134%
      Belgium 0.1%
      Denmark 0.149%
      France 0.124%
      Germany 0.069%
      Italy 0.219%
      Norway 0.150%
      Spain 0.087%
      Sweden 0.038%
      Switzerland 0.091%
      United Kingdom 0.061%

      Again, the above crude IFR is low (because it does not account for the people who were infected as of March 28th but who will die) and high (because I have intentionally used the high end estimated deaths as of March 28th). But even so, the figure should be very comforting. If actual IFRs for the disease are closer to .1% than 1%, that presages an approximate order of magnitude fewer deaths.

    2. Note that I’ve used the reported attack rate, which is somewhere between their low and high end (95% confidence interval).

  23. I think that the point of this article was missed.

    Everyone has focused on what Epstein wrote. What I haven’t seen anyone else address is the larger issue surrounding appeals to authority. It’s rather scary seeing so called libertarians here telling someone to stay in their lane and not to “abuse” their authority. Excuse me, but since when did being respected by someone forbid you from exercising free speech?

    Expertise is a label. It is not truth itself and we should never confuse the two. Just like how they are many untruthful facts, there are many incorrect experts. It is a deadly and foolish assumption to create an unwelcome and outright hostile environment for critique, especially with something as serious as disease.

Please to post comments