Richard Epstein Cops to a 'Stupid Gaffe' in Controversial Coronavirus Essay That Caught Trump Administration Attention

But he stands by his reasoning and predicts that global deaths will peak under 50,000.


Last week, I interviewed New York University law professor Richard Epstein based on a contrarian article he had written for the Hoover Institution, where he is a senior fellow (video of the interview is here and a podcast version is here). The Washington Post reports that "conservatives close to Trump and numerous administration officials have been circulating" the law prof's article, which "plays down the extent of the spread and the threat" of coronavirus.

While Epstein does not have a medical or epidemiological background, he is a fellow at the Center for Medical Ethics at the University of Chicago and had worked on research about the spread of AIDS during the 1980s and 1990s. Amid what he called a "full state of panic," his March 16 article boldly declared that the total number of deaths from COVID-19 would top out under 50,000 and that in the United States we would see "about 500 deaths at the end."

In fact, at least 537 Americans have died from COVID-19 as of today. I wrote Epstein last night to ask for a response and he emailed me to say that he had committed "a stupid gaffe the first time round" and that correct number for U.S. deaths should have been 5,000. He pointed me to his latest column for Hoover, "Coronavirus Overreaction," which was published yesterday. From that article:

In my column last week, I predicted that the world would eventually see about 50,000 deaths from the novel coronavirus, and the United States about 500. These two numbers are clearly not in sync. If the first number holds, the total US deaths should be about 4 to 5 percent of that total, or about 2,000–2,500 deaths. The current numbers are getting larger, so it is possible both figures will move up in a rough proportion from even that revised estimate. Indeed, the recent run-ups in Italy and perhaps Spain suggest that those countries have yet to turn the corner.

Yet Epstein remains steadfast in his belief that current projections of millions of deaths globally are based upon faulty extrapolations from the number of existing cases.

The implicit assumption behind…my earlier column is that the worst way to model growth in deaths from the coronavirus is through a geometric progression that runs rapidly through a large number of cycles, each of which generates more cases than the cycle before. If the exponent at each of these stages is greater than one, the model will quickly explode. Thus, if we assume that each infected person infects 2.3 other persons, a world that starts with 100 infections in the first period will have about 2,800, infections by the fifth period, and the rate would grow even more rapidly after that. If the periods are close together, it is easy to see how the fearful analyst could conclude that the world will be soon consumed….

But there is no reason to assume that the doubling is a constant, and therefore there is no reason to accept the ratio of 800 to 1 for true cases compared to deaths. The situation in South Korea alone should dispel that narrative, where pervasive COVID-19 testing has revealed 8,897 cases against only 104 deaths…. Political leaders predict rising rates of infection, running perhaps for three months, while more cautious analysts think that the cycles turn down far sooner than the doomsday models predict. All of the stay-at-home orders that we see assume that the growth of cases (and of deaths) will be exponential, which is the sole justification for imposing the draconian measures that have wrecked both the economy and upended the lives of millions of people.

Epstein believes that such straight-line extrapolations fail to account for changes in behavior and new information that inevitably reduces the case fatality rate (CFR), or the number of infected people who die. In this, Epstein is hardly alone. There are at least two reasons why the worst-case scenarios for COVID-19 seem unrealistic.

First is that when more people get tested, the CFR declines as patients with few or no symptoms are included in official statistics (due to a lack of tests, only people in high-risk categories or with visible symptoms have been tested so far). As Brett Giroir, assistant secretary for health at the U.S. Department of Health and Human Services, explained, early CFRs of 2 percent to 3 percent are almost certainly too high. "The best estimates now for the overall mortality rate for COVID-19 is somewhere between 0.1 percent and 1 percent," said Giroir, who is in agreement with a New England Journal of Medicine (NEJM) commentary co-authored by the Centers for Disease Control and Prevention (CDC)'s Anthony Fauci. Based on a study in Science that used comprehensive infection data from China, Sullum estimates that the CFR is actually closer to 0.2 percent rather than "the current crude CFR of 1.2 percent." For his part, Fauci has amended his position since the NEJM article and now believes coronavirus fatalities are likely to be about 10 times worse than seasonal flu and thus have a CFR of 1 percent.

Second, the rate of transmission responds to changes in behavior. The CDC's worst-case scenarios assume that efforts to reduce infections are largely unsuccessful and that each carrier infects another 2.3 people, leading ultimately to about two-thirds of Americans being infected. However, the Science study found that the infection rate in China was reduced from 2.4 to 0.99 in short order once lockdown measures were put in place. So there are ways to reduce the spread, although the methods used in China counter the tactics supported by Epstein.

"We need a public debate on the political response to COVID-19, and we need it now," insists Epstein in his new column. He argues that high-risk populations such as people over 70 are already removing themselves "out of harm's way" and that members of lower-risk groups should be given more latitude in their work and social movements. This sort of targeted response is appealing and fits with the proscriptions of David L. Katz, the president of True Health Initiative and the founder of the Yale-Griffin Prevention Research Center, but other epidemiologists warn that it will become impossible to sequester the old for very long under such circumstances. "Our government fiats will probably save very few, if any, lives saved over what we can obtain through more focused voluntary precautions," Epstein writes, another prediction that will be tested over the coming months.

NEXT: U.S. Needs Billions of New Masks to Combat Coronavirus. Feds Say It'll Take Up To 90 Days to Approve New Mask-Making Facilities.

Coronavirus Epidemiology Public Health Donald Trump FDA CDC

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 or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

Please to post comments

42 responses to “Richard Epstein Cops to a 'Stupid Gaffe' in Controversial Coronavirus Essay That Caught Trump Administration Attention

  1. Enough with the amateur prognostication, these newly minted experts who were just plain old journos yesterday arent going to do any better that the professionals, and the professionals have been awful.

    1. +1, I mean +10

      1. I mean, I'm not arguing with his larger point, which I agree with...
        "But there is no reason to assume that the doubling is a constant, and therefore there is no reason to accept the ratio of 800 to 1 for true cases compared to deaths. The situation in South Korea alone should dispel that narrative, where pervasive COVID-19 testing has revealed 8,897 cases against only 104 deaths"
        Kinda contradicts what he's trying to say (I think). Indeed, 104/8897 isn't a 1/800 death/cases rate... it's 1/89
        Which looks a lot worse, and I'd have to assume is another typo/gaffe

        1. I think what he was referring to is the rate of spread of infection. As more people become infected the virus has a more difficult time finding new hosts to infect, thus the rate drops.
          People with no experience of outbreaks are crudely calculating the time to infect the whole population of the US and inferring massive death rates from this.

        2. I think he is saying the prediction of 800 true cases (unknown) for every 1 death (known) looks grossly exaggerated when the ROK has found only 89 cases per death. Almost an order of magnitude fewer than predicted.

          1. Ah
            Thank you for that clarification

    2. Make 6150 bucks every month... Start doing online computer-based work through our website. I have been working from home for 4 years now and I love it. I don't have a boss standing over my shoulder and I make my own hours. The tips below are very informative and anyone currently working from home or planning to in the future could use this website...
      More Read Here

  2. The goal of the media Propagandists is to get rid of Trump and destroy the US economy no matter what the facts and realities of the Coronavirus cold due to cough actually are.

    Americans are going back to their regular lives by this weekend.

    1. I do hope that next winter when Trump is reelected and some existing or new virus comes out of Asia, Trump is able to limit sick non-Americans and Americans entering the USA until they are quarantined for recommended periods.

      We are remember when Lefties went after Trump for even raising the possibility of restriction non-American travelers for safety or disease reasons.

    2. Did the liberal media cause India and China to shut down as well? You're the propagandist whether you realise it or not. Obviously you don't realize it or you wouldn't be parroting out of date Trump cult propaganda from two weeks ago. Let me help you friend. Haven't you heard we were invaded by a Chinese virus and it is wartime but our Dear Leader is on top of everything and has also known the enemy was coming and has always prepared for this evil Chinese disease.

      1. Look you feckless cunt just because you suck democrat dick every day of the week doesn't make you an authority on disease vectors, or public crisis management. It makes you an expert on techniques for swallowing. Be the ball.

  3. "We need a public debate on the political response to COVID-19"

    Well, at least it's not a "national conversation" this time.

    Seriously, what does Epstein want? One of Reason's debates in which whoever sways more of the studio audience wins?

    1. Epstein was willing to speak up in writing for the record in real time that he thought the political response was not what’s needed. Too few people with public audiences were speaking up then and now. The public debate Epstein wants is more people with sizable public audiences speaking up in writing for the record in real time, before government actions are fait accomplis.

  4. Trump wants the country running again by Easter. He's the first of our illustrious leadership to make that suggestion, and i think Americans will beat that by 2 or 3 weeks. One bright spot is that Pelosi and Schumer have poisoned the bailout so badly that the recovery will be here before before congress votes on anything. So at least we don't have the green new deal riding on the back of the pandemic.

    1. Yeah well nobody is listening to Trump. He might as well be commanding the tide.

    2. 2 or 3 weeks? Easter is in 19 days.

  5. due to a lack of tests, only people in high-risk categories or with visible symptoms have been tested so far

    Stop with your perpetual head-up-ass stupidity Gillespie. If you can't keep up with what is happening NOW, then just stop repeating shit that may (or may not) have been true three weeks ago. Ideologues like Gillespie are as rigid re reality as conservatives are.

    TX - 11,167 tests - 410 cases - 3.7% positive rate
    CA - 26,400 tests - 1733 cases - 6.6% positive
    FL - 14,540 tests - 1412 cases - 9.7% positive (also - total of 3090 were 'risk'/traces)
    NY - hell if I can find out how many tests but with 25,000+ cases they are way beyond general testing doing anything now
    WI - 8694 tests - 457 cases - 5.2% positive
    WA - 33,933 tests - 2221 cases - 7% positive

    WTF is your intention of this testing Gillespie? To prove that we can waste money testing 100 people for every one who is infected? 1000? And then repeat again next week? And the week after? In order to ignore the problem in hopes that this seasonal flu?

    Same is true of many states. We are not universally testing at-risk groups (medical workers or all inpatients in hospitals) anywhere AFAIK. The only 'negatives' we should be vacuuming up in prep for changing their workplace post-haste. Some places tested contact tracing. But many, if not most, have clearly turned the new availability of testing into a way to profit off either hypochondriacs or self-important assholes who want to get some column-inches re their test results.

    Hint to you twits at Reason -- CDC no longer receives negative test results from non-CDC kits once the positives test out the same as CDC. So CDC numbers are crap now. Some states don't even get those results from their own counties or direct from private labs. Stop relying on shitty data. The US now has shittier data than Italy - still beating Iran though.

    1. All licensed labs have to report positive results to the public health labs which in turn pass the data to the CDC.

      1. Yes. Not NEGATIVE results. Which is - based on above states - 90%+ of the total test results and thus of the total tests given.

        1. That happens informally. There’s no legal requirement to report the negatives but the CDC still gets the data. I know, because we and and all the other labs are doing this.
          What’s your first hand experience of lab reporting?

          1. The CDC ain't putting it on their site. So it really doesn't matter one whit whether it is in some manila folder in some warehouse being researched informally by Top Men. What is transparent is what gets reported. And if no one knows how many negatives have been tested, then the working assumption by everyone outside some little clique is that those tests are NOT being done.

              1. He prefers chinese propaganda sites sense they have had a better response in his view.

              2. And?

                The number on the CDC site says 83,391 tests
                The number in the 5 states (x NY) I mentioned is 94,734 tests. And that state data is most likely up to one day old. Plus NY and 44 other states

                So please spare me that the CDC data is worth a shit. It is - at best - the number of CDC kits + the number of non-CDC kit positives. But who really cares what it actually is. It's shit. Move on.

                1. So what exactly is your original point then?
                  People who are symptomatic are the priority for testing and of those tested only a small percentage have Coronavirus. The rest have influenza-like illness.

    2. Here's the link you are looking for. It shows cases and tests for every state. NY is clearly behind the testing curve.

  6. It has been interesting to see the Covid-19 debacle become just another episode in the 'the people vs. the experts'. The truth lies on neither side of that equation, but there's something just nuts about how in the midst of a pandemic and economic collapse everyone is still heartily fighting the culture war.

  7. Clearly he screwed up with the 500 deaths comment, but he's not wrong in his overall point. That's what I don't get about all these people who are so quick to "listen to the experts!" and proclaim how scientifically literate they are because they can see that because x% of positive cases died in Lombardy, then clearly that same percentage extrapolated to the entire US population will die if we don't DO SOMETHING. There are so many other factors, and without all this stupid "social distancing," just getting people to be extra aware, stay home from work/school if you have any cold or flu symptoms even if you normally wouldn't have, and make a special effort to wash/sanitize you hands more would probably have gotten us 90% there. Plus, this thing is obviously going to have a higher fatality rate early on- it's going to burn through not just old people, but really sick old people, and they will die at a much higher rate than the average person would be expected to. But there's only so many of them, and as we've seen even among 80+ year olds that don't have major underlying conditions, 85% still live. So that early trend is not going to continue, and the fatality doubling rate obviously can't even come close to continuing for long. Otherwise, by my calculations everyone in the country will be dead by sometime in June.

    1. There are so many other factors

      Actually there really aren't that many factors. The social adjustment stuff merely slows down the spread.

      Reason roundup linked to this am re Scott Gottlieb

      The last column - headed 'case-fatality' is the range of fatalities by age assuming that the hospitals don't fill so everyone can get treated heroically. Ensuring the hospitals don't fill is the purpose of all this social adjustment and economic lockdown stuff. Personally, I have little interest in saving the 'heroic care' model - but those who do are the ones who are imposing all this stuff on everyone.

      The middle column - headed 'ICU admission' is the range of fatalities once ICU beds and/or ventilators are all full. None of those folks who need that and don't get that will survive. Period. That is Lombardy for the last week or two. Hubei for the first week or so of Feb

      The first column - headed 'hospitalization' is IMO irrelevant. I suspect right now it is way way overstated simply because hospitals aren't full so cases are being admitted to hospital not because they really need it but because doctors want to monitor/watch/learn. Once hospitals are full, then everything outside the ICU will be 'waiting line for ICU'. There will be a triage for going into ICU - and another triage outside the hospital door. The fatality rate will depend more on those triage decisions - but regardless, whoever gets ill enough to normally 'go to hospital' and doesn't get admitted will have the worst experience of their lives (unless they have lived in the Third World in an epidemic).

      Triage in both Lombardy and China was heavily age-based. So the young got treatment (third column) and, for awhile at least, the older didn't (second column). I have no idea how that triage will go in the US. But in some places in the US, we will fill the ICU's and hospitals for awhile in this first wave. And we are not remotely capable of handling the aftermath of that triage since we can't even seem to discuss it as it looms.

      1. The first column – headed ‘hospitalization’ is IMO irrelevant. I suspect right now it is way way overstated simply because hospitals aren’t full so cases are being admitted to hospital not because they really need it but because doctors want to monitor/watch/learn.
        At least a couple weeks ago, that number was for VISITS, not actually getting checked in.

      2. These scenarios would be dramatically improved by saturation off-label use of hydroxychloroquine and azithromycin. The FDA people, with their randomized-trial dictates that deny care to people who aren’t in the paperwork loop to receive medicine, and with their denials of care to half the people who do make it through the paperwork hurdles and enroll, will fall back on dark-ages pre-Bayesian statistics to do despicable things to us to guard their undeserved reputation and their individual self-esteems. Here, all doctors could provide these treatments off-label, and any who misrevere the FDA branding will contribute to greater needs for heroic measures and greater incidences of even-worse outcomes.

        These scenarios would also be somewhat prevented in the first place if our Center for Disease Control and Prevention would inform us about how to prevent ourselves from getting infected at all by boosting our immune systems, instead of solely informing us how to not infect others.

        I don’t find it necessary to imagine any malevolence on the part of any of these people. I can easily visualize the FDA people running their dark-ages tests, the doctors carefully choose to not prescribe off-label, and the CDC people thoughtfully focusing away from immune strengthening all believing they’re focusing all their attentions on the best options, all confident they know just what they’re doing. I see their individual actions as each being understandably human; and yet I fall short at being able to imagine how it is that all these errors just happen to act only in the directions that hurt people. Visualizing that these people could choose these careers and yet still have these blind spots makes it all not tragically lame but instead Kafkaesque-dark.

    2. As an anecdotal aside - a series of tweets from an ER doc right now in I assume Seattle or NYC

      Pretty sure he's not saying this isn't even the flu

      1. Yes, doctors see sick people. COVID cases aren't even 1% of total cases right now, so how can they be overwhelming the system?

  8. "This sort of targeted response is appealing and fits with the proscriptions of David L. Katz"

    I believe you mean "prescriptions" here, Nick.

    1. To someone like Nick, saying don't do something actually makes him want to do it, ergo a proscription is like a prescription.

  9. "a stupid gaffe the first time round"

    gaffe correction, or backpedaling... Let's see what happens if deaths clear 5,000. (I hope he's right, and we don't hit that number, but I don't think he is)

  10. So he changes his numbers as the facts change

    real US deaths should be 5000, uh yeah, so we screw up and manage to have 10 percent of the global deaths

    I don't know if the projections are correct, but the ones that made them are smarter than this boob

    1. My meaning being if he sticks with his 50k global, where we are at 20k right now also seems unlikely

  11. ★My last month paycheck was for 11000 dollars… All i did was simple online work from comfort at home for 3-4 hours/day that I got from this agency I discovered over the internet and they paid me for it 95 bucks every hour... .....Read More

  12. In 1918, approx 650K died here in the US, roughly 12x the number we lost in WWI. At that time, the total US population was about 1/3 what it is today. So what makes Epstein so sanguine (if projecting 50K deaths in this country is a "sanguine" conclusion)? We have no established treatment except supportive care, including respirators for the critically ill and antibiotics for opportunistic pneumonias, which while helpful are not going to make the corvid-19 a pussycat, and a vaccine and/or proven drugs are an unknown ways off.

    Does Epstein know that this corona virus is more "benign" or less virulent than the so-called Spanish Flu of 1918 or the corona virus we have experienced before (SARS and MERS). Or his own massaging of published data lead him to his confident conclusions? Does he have any confidence in the leadership and his political backers, who have been and still are so egregiously wrong about this plague upon us? (What will it be like when Easter is here? What will it be like when Passover is here and the Ten Plagues inflicted on the Egyptian oppressors?)

  13. Hey Epstein, You should just get it over with and apologize for this dumb prediction today. The longer you wait, the more credibility you lose.

    While you're at it, you should admit that humans cause climate change and that cutting taxes does not increase revenue. Then you should commit to not writing about anything other than the law ever again, because you are a hack.

Comments are closed.