Yesterday, I posted Part I of my commentary on Alex Tabarrok’s talk on Covid failures and successes.
Here’s Part II. It starts at about the 41:00 point. Alex discusses four failures: failure to do an Operation Warp Speed for masks and trials; not getting the vaccine early to nursing homes; not experimenting and studying the virus; and not doing human challenge trials.
On failure to do an OWS for masks and trials, I don’t have much to say. I don’t know how feasible that would have been in the time they had. Alex himself shows how heavy-footed and slow the government is on many of these things. We don’t know that OWS for masks and trials would have worked well.
Not getting the vaccine early to nursing homes. This is a biggie. I totally agree with him on this one. As he notes, getting it to nursing homes 5 weeks earlier might have (Alex says “would have”) saved 14,000 lives. On the issue of nursing homes, that would have been the obvious to mention another huge nursing home failure in the spring of 202o: the decision of at least 3 governors of major states, Governor Cuomo of New York, Governor Phil Murphy of New Jersey, and Governor Wolf of Pennsylvania, to insist that nursing homes take in people who tested positive for Covid. Alex says nothing about this.
Not funding experimentation and study of the virus. Alex points out that Fast Grants, run by Tyler Cowen and Patrick Collison, handed out $60 million to study the virus before the National Institutes of Health had reviewed even one Covid proposal. Alex is justifiably upset by this. He also points out that Fast Grants was the entity that funded Professor Anne Wyllie of Yale to do her Saliva Direct project. This is Yale, which, Alex points out, has a $42 billion endowment. So neither the government nor the non-profit sector shone on this, to put it mildly.
In a toting of costs and benefits, though, which is what Alex’s talk is about, I wish he had commented retrospectively on whether it was a good idea for Emergent Ventures, another fund run by Tyler Cowen, to give a grant to Neil Ferguson and his Imperial College colleagues for their model that, as noted in Part I, way overpredicted deaths for Sweden without non-pharmaceutical interventions. Does Alex think that funding Ferguson was a good move? We don’t know.
Human challenge trials of the vaccine. Watch from 50:40 to 55:30. Alex is almost pitch perfect on this. Why almost? He should have a statement in there that anyone who engages in a human challenge trial does so voluntarily. When you say it, everyone gets it, but it needs to be said.
One thing I wonder that it would have been nice for Alex to address. He points out (at about 49:20) that Moderna’s vaccine was designed on January 13, 2020. When he, Kremer, Athey and the others proposed Operation Warp Speed later in 2020, did they know that? If so, how did it affect their thinking? If not, would knowing that have changed their thinking?
READER COMMENTS
AMW
Mar 17 2022 at 6:59pm
I’ll just take this opportunity to note that the last time David brought up the Ferguson model, I pointed out that his model for the UK included a simulation which produced a multi-peaked pandemic that resulted in between 5,600 and 120,000 deaths over a two-year period. Almost exactly two years out total deaths now stand at about 163,000 for the UK. Not exactly an overestimate.
Todd K
Mar 18 2022 at 9:37am
Ferguson’s model showed 510,000 Covid deaths in the UK by October 2020 if there was no lockdown. It also showed 490,000 Covid deaths in Japan if there was no lockdown. There have been 25,000 Covid deaths there after two years and no lockdown but a national emergency kept many at home for six weeks while many still worked. However, the declared emergency didn’t have an effect on cases as those were already decelerating.
Ferguson’s model also predicted 90,000 Covid deaths in Sweden by June 2020 if there was no lockdown. Sweden didn’t lock down and there were 5,500 Covid deaths by mid-June and 18,000 as of this month.
Mark Brophy
Mar 17 2022 at 9:16pm
I believe that 6 states insisted that nursing homes take in people who tested positive for Covid, Michigan plus 5 in the northeast: MA, NY, NJ, CT, RI. Mass murder isn’t a crime but the Governor of New York lost his job for harassing women.
steve
Mar 18 2022 at 9:32pm
I practice in PA. In our network a covid pt could not go to a nursing home until they were both stable and no longer infectious. This was true for all of the facilities in our network. Also true for all of the hospitals with which we were in communication. Economists were able to figure out that sending infectious pts back would likely create more pts. Doctors were also able to figure this out.
Steve
Tyler Cowen
Mar 18 2022 at 5:44am
David, this is a simple misrepresentation. Fast Grants did not give anything to Ferguson (neither did Emergent Ventures, but that is a longer story). Get your facts straight!
J Mann
Mar 18 2022 at 9:44am
That’s a big mistake, and I hope Henderson corrects it soon.
I’d be interested in hearing the longer story of why Emergent Ventures did not give anything to Ferguson. Did he decline the grant or redirect it or something?
https://marginalrevolution.com/marginalrevolution/2020/03/emergent-ventures-prize-winners-for-coronavirus-work.html
Alan Goldhammer
Mar 18 2022 at 10:10am
The research group at Imperial College was amply funded by the UK government.
J Mann
Mar 21 2022 at 10:54am
Thanks – does that mean that Ferguson declined the grant, or otherwise didn’t receiv it?
David Henderson
Mar 18 2022 at 11:08am
Thanks, Tyler. Correction made. I think you need to correct your own comment, though. Emergent Ventures did give money to the Ferguson team, as you noted in your March 21, 2020 post that J Mann noted in his comment.
Phillip Magness
Mar 20 2022 at 5:51pm
It would be helpful if Tyler Cowen would fully disclose what, exactly, he did award to Ferguson and Imperial College for their disastrous covid-19 model.
This was widely reported on at the time of the grant, including in a university press release from GMU. https://www.gmu.edu/news/2020-03/masons-tyler-cowen-leads-efforts-incentivize-coronavirus-response
I believe that Tyler has soured on lockdowns since that time in the wake of their poor performance over the last 2 years. But I don’t think he’s ever revisited or clarified his own original error of throwing his support behind Ferguson in March 2020. That’s a fairly severe oversight, given that he’s repeatedly criticized others for erroneous early covid predictions that had far fewer harmful consequences than the Ferguson model.
Phillip Magness
Mar 20 2022 at 5:58pm
Also if I’m not mistaken, the Fast Grants program was created in the early weeks of the pandemic shortly *after* the Ferguson award was announced by GMU (3/23/20). As far as I can tell, the Fast Grants call went out on 4/7/20 – https://www.mercatus.org/features/mercatus-emergent-ventures-launches-fast-grants-fight-covid-19
The 3/23/20 GMU press release says that the Ferguson award was funded through Emergent Ventures, and the language of the press release touts it as an attempt to fulfill a nearly-identical purpose to what Fast Grants – created two weeks later – sought to achieve.
So David is not in error to suggest a kinship between the two programs. Nor is it unreasonable to ask what the deal was with Cowen’s very public support for Ferguson’s model in those early days of the pandemic and lockdowns.
J Mann
Mar 21 2022 at 12:12pm
(1) Tyler said that Emergent Ventures didn’t give anything to Ferguson, but that’s a “longer story.” I’m inferring that it’s something on the order of Ferguson declined the grant or asked that it be given to an unrelated charity or something, but since it’s a long story, it’s probably something more complicated than that. (It’s also possible that Emergent Ventures is some kind of Producers-style scheme and Tyler absconded with the money to fund his patronage of interesting restaurants, but that seems unlikely.)
(2) It’s possible that Ferguson’s model was excellent and worthy of recognition, even if it was wrong in some particulars. Has anybody, especially Ferguson, looked back on (a) where the model was likely wrong and (b) how much weight was rational to apply to it based on what we knew at the time?
Thomas Lee Hutcheson
Mar 18 2022 at 7:57am
I only wish Alex had addressed the way MSM failed to cover these failures. Even when they were being recognized as failures, reporters were not asking why the failures were occurring.
Alan Goldhammer
Mar 18 2022 at 10:07am
It is not clear to me what this means. Operation Warp Speed was a government originated program based on a public/private partnership approach. The first announcement came at the end of April 2020 and it was ramped up very quickly with a top notch vaccine expert Moncef Slaoui (who was unfairly criticized by a number of Democrats because he worked in the pharma industry).
Regarding Human Challenge Trials (HCTs), it really would not have sped development up. The real opportunity was missed when all Zika vaccine development ceased as the virus pretty much went away on its own. there were a couple of mRNA vaccine prototypes around and these could have entered early phase trials where some very good antibody titer and safety data would have been acquired. The process could have been validated during this time period which would have shaved more time of development of a Covid vaccine as the technology was understood. Most of those advocating for HCTs do not understand vaccine development. The need for safety data is critical as vaccines are administered to healthy people (and of course children, who receive the majority of vaccines at extremely young ages). Covid vaccines are no different in this regard.
While you can get reliable antibody titer data with smaller numbers of patients who might be in a HCT, there is no way to get reliable safety data with this cohort. Another pitfall is that all of the proposals for HCTs involve health ‘young’ people. This might lead to the development of a dose that is not effective for older people. It’s important to note that those over 60 years of age receive a high dose flu vaccine to provoke a better immune response. this approach came from several years of work. FDA’s approach of permitting the vaccines to be administered prior to full licensure was correct as the manufacturers continued to complete their clinical trials (in most cases that I am aware of the trials continued for several months in order to full acquire all the safety data on the vaccine. IIRC, they were required to follow those in the trial for six months regarding adverse effects from the vaccine).
David Henderson
Mar 18 2022 at 2:32pm
You asked what my questions mean. They mean what they say.
Charles Hooper
Mar 21 2022 at 5:36pm
Even if, say, six months of safety data is required for approval, human challenge trials can speed up the testing of efficacy to a point where everything else is sped up: manufacturing, distribution, FDA approval contingent on six months of safety data, etc. Then, once the safety data is complete, everything is already teed up and ready to go.
But there’s another way that HCT’s could really speed up the process and that’s if people were free to take the vaccine with less than six months of safety data. Six months is no magic length of time. One person might prefer 12 months while another would be fine with three.
A number of problems arise by telling people what they should accept instead of actually asking them what they would accept.
Comments are closed.